Article(id=1319955578244420547, tenantId=1146029695717560320, journalId=1278651732997652489, issueId=1319955375944745964, articleNumber=null, orderNo=null, doi=10.12086/oee.2026.250208, pmid=null, cstr=32245.14.oee.2026.250208, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1752508800000, receivedDateStr=2025-07-15, revisedDate=1758643200000, revisedDateStr=2025-09-24, acceptedDate=1759075200000, acceptedDateStr=2025-09-29, onlineDate=1792631533442, onlineDateStr=2026-10-22, pubDate=1771948800000, pubDateStr=2026-02-25, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1792631533442, onlineIssueDateStr=2026-10-22, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1792631533442, creator=13701087609, updateTime=1792631533442, updator=13701087609, issue=Issue{id=1319955375944745964, tenantId=1146029695717560320, journalId=1278651732997652489, year='2026', volume='53', issue='2', pageStart='250117', pageEnd='1', issueExtLink='null', onlineDate='null', pubDate='1771948800000', pubDateStr='2026-02-25', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1792631485210, creator='13701087609', updateTime=1792631485210, updator='13701087609', preIssue=null, nextIssue=null, articleTotal=null, ext=null, issueFiles=null, downloadFileDto=null}, startPage=250208, endPage=1, ext={EN=ArticleExt(id=1319955578500273095, articleId=1319955578244420547, tenantId=1146029695717560320, journalId=1278651732997652489, language=EN, title=A small target detection algorithm for UAV aerial images based on frequency domain perception feature fusion, columnId=1279511634116841602, journalTitle=Opto-Electronic Engineering, columnName=Article, runingTitle=null, highlight=null, articleAbstract=
Objective

Small object detection remains a significant and persistent challenge within the computer vision domain. This challenge is particularly acute in the context of drone aerial imagery, where objects captured from high altitudes appear extremely small, densely packed, and often exhibit blurred features due to low resolution and environmental factors. Traditional object detection algorithms frequently suffer from performance degradation in these scenarios, manifesting as high rates of missed detections and false positives. To address these critical limitations, this paper proposes a novel object detection algorithm named FFC-YOLO (FASFF-FreqFusion-CAA YOLO), which is built upon a frequency-aware feature fusion framework.

Methods

The core of the FFC-YOLO algorithm lies in three systematic and synergistic innovations designed to enhance feature representation, fusion, and contextual understanding specifically for small and dense objects. First, to bolster multi-scale feature extraction, we redesign the detection head by incorporating a Four Adaptively Spatial Feature Fusion (Detect-FASFF) structure. This is augmented with an additional P2 detection layer dedicated to capturing finer-grained spatial information crucial for identifying minuscule targets. The FASFF mechanism leverages four adaptive branches to dynamically aggregate features across different scales, thereby significantly improving the model's sensitivity and discriminative power for small objects across varying sizes.Second, we comprehensively optimize the feature fusion pathway to counteract the information loss common in conventional sampling methods (e.g., up-sampling and down-sampling). We introduce a Frequency-aware Feature Fusion (FreqFusion) module. This module operates by decomposing features into different frequency components within the frequency domain, allowing for more deliberate and effective integration of high-frequency details (essential for edge and texture information of small objects) and low-frequency semantics. This FreqFusion module is then elegantly coupled with a Bidirectional Feature Pyramid Network (BiFPN) structure. The resulting fusion architecture enables efficient, multi-level bidirectional cross-scale information flow, effectively mitigating the problems of feature obliteration, category misclassification, and bounding box localization drift in scenes crowded with small objects.

Third, to enhance the model's ability to leverage surrounding semantic context—a vital clue when object appearance is ambiguous—we integrate a Context Anchor Attention (CAA) mechanism into the C2PSA module. This attention mechanism functions by establishing dynamic associations between anchor points and their contextual surroundings, enabling the feature extraction process to selectively focus on and amplify informative regions around potential targets. This contextual awareness strengthens feature representation and aids in distinguishing small objects from complex backgrounds.

Results and Discussions

The efficacy of the proposed FFC-YOLO algorithm is rigorously validated through extensive experiments on two benchmark datasets. On the public VisDrone2019 dataset, a standard benchmark for drone vision, FFC-YOLO achieves a mean Average Precision (mAP@0.5) of 40.0%. This represents a substantial performance gain of 8.0%, 7.6%, and 7.8% over the widely-recognized YOLOv8n, YOLOv10n, and YOLOv11n baselines, respectively. To further verify its generalization capability and prowess in detecting extremely small targets, we conducted additional experiments on a custom-built, challenging dataset named "tiny-data", which contains three categories of progressively smaller persons ("sperson", "lperson", "wperson"). On this dataset, FFC-YOLO demonstrates superior results compared to the strong YOLOv11n baseline, with improvements of 9.2% in mAP@0.5, 8.7% in Precision (P), and 5.9% in Recall (R). These consistent gains across both public and private datasets robustly confirm that the FFC-YOLO algorithm possesses strong generalization ability and excels in the demanding task of drone-based aerial image analysis, particularly for long-range and dense small object detection.

Conclusions

In conclusion, this paper presents a comprehensive and effective solution to the small object detection problem in aerial imagery. By innovating at the levels of feature extraction (FASFF head), feature fusion (FreqFusion-BiFPN), and contextual modeling (CAA attention), the FFC-YOLO framework successfully addresses key shortcomings of prior methods. The experimental evidence underscores its potential for practical application in UAV-based surveillance, traffic monitoring, and search-and-rescue operations. For future work, we plan to explore model compression techniques such as pruning and knowledge distillation to streamline the model for efficient deployment on resource-constrained edge devices, thereby bridging the gap between high accuracy and real-time performance.

, authors=Xiaojun Jia1, 2, Biao Tang1, 2, Shuyun Luo2, Nuolei Liang1, 2, Chunhua Zhang1, authorsList=Xiaojun Jia, Biao Tang, Shuyun Luo, Nuolei Liang, Chunhua Zhang, authorCompany=null, correspAuthors=Xiaojun Jia, authorNote=null, correspAuthorsNote=
, copyrightStatement=Copyright © 2026 Opto-Electronic Engineering. All rights reserved., copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=null, magXml=null, pdfUrl=null, pdf=null, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=null, mapNumber=null, fund=null), CN=ArticleExt(id=1319955588847620138, articleId=1319955578244420547, tenantId=1146029695717560320, journalId=1278651732997652489, language=CN, title=基于频域感知特征融合的无人机航拍图像小目标检测算法, columnId=1319955120360636489, journalTitle=光电工程, columnName=研究论文, runingTitle=null, highlight=null, articleAbstract=

小目标检测一直是计算机视觉处理中一项极具挑战性的任务,针对传统无人机航拍图像小目标多且密集,而传统检测算法存在漏检、误检的问题,提出一种基于频域感知特征融合的FFC-YOLO (FASFF-FreqFusion-CAA YOLO)目标检测算法。首先,通过增加P2检测层,重新设计了针对小目标检测的Detect-FASFF (four adaptively spatial feature fusion)检测头结构。然后,将传统的特征融合采样方法替换为频率感知特征融合FreqFusion (frequency-aware feature fusion)方法并结合BiFPN (bidirectional feature pyramid network)结构提升了算法处理密集小目标图像的能力。最后,在C2PSA模块中加入CAA (context anchor attention)注意力机制,增强了目标上下文特征信息的关联。在VisDrone2019数据集中,基于FFC-YOLO目标检测算法的mAP@0.5为40%,较传统算法如Fast-RCNN、RetinaNet分别提升了28.4%、18.6%,较YOLO算法如YOLO v8n、v10n、v11n分别提升了8.0%、7.6%、7.8%。在自建小目标数据集tiny-data中,基于FFC-YOLO的3类目标sperson、lperson和wperson的检测结果较YOLOv11n,mAP@0.5、P、R分别提升了9.2%、8.7%、5.9%。实验结果表明,FFC-YOLO小目标检测算法能够更好地应用于无人机航拍图像检测领域。

, authors=贾小军1, 2, *, *, 唐彪1, 2, 骆淑云2, 梁诺雷1, 2, 张春花1, authorsList=贾小军, 唐彪, 骆淑云, 梁诺雷, 张春花, authorCompany=null, correspAuthors=贾小军, authorNote=

作者贡献

唐彪、梁诺雷:研究资料收集、分析、论文写作;贾小军、洛淑云、张春花:指导论文选题、论文修改、研究资料收集。

贾小军(1974-),四川邻水人,嘉兴大学教授,研究方向为计算机视觉与模式识别,图像处理。E-mail:

, correspAuthorsNote=
贾小军,
, copyrightStatement=版权所有©《光电工程》编辑部 2026, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=c9A+GBnznnkCbrixLukc3g==, magXml=hnC7XYFIgmisdFpIz4Y9TA==, pdfUrl=null, pdf=ni+e1EexkNVBLZ3cisZa8A==, pdfFileSize=8806113, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=PJKCqEwQzKi/W7RiaII1Xw==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=2FIbY9PE+PjScl7UyUIpRw==, mapNumber=null, fund=null)}, authors=[Author(id=1319955589258661938, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=xjjiad@sina.com, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1319955589346742325, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, authorId=1319955589258661938, language=EN, stringName=Xiaojun Jia, firstName=Xiaojun, middleName=null, lastName=Jia, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1College of Artificial Intelligence, Jiaxing University, Jiaxing, Zhejiang 314001, China
2School of Information Science and Engineering, Zhejiang Sci-Tech University, Hangzhou, Zhejiang 310018, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1319955589426434102, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, authorId=1319955589258661938, language=CN, stringName=贾小军, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, *, *, address=1嘉兴大学人工智能学院,浙江 嘉兴 314001
2浙江理工大学信息科学与工程学院,浙江 杭州 310018, bio={"img":"nnI7zyTp5E5cjxsBk2UQIQ==","content":"

贾小军(1974-),四川邻水人,嘉兴大学教授,研究方向为计算机视觉与模式识别,图像处理。E-mail:

"}, bioImg=nnI7zyTp5E5cjxsBk2UQIQ==, bioContent=

贾小军(1974-),四川邻水人,嘉兴大学教授,研究方向为计算机视觉与模式识别,图像处理。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1319955589078306859, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, xref=1, ext=[AuthorCompanyExt(id=1319955589082501164, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589078306859, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Artificial Intelligence, Jiaxing University, Jiaxing, Zhejiang 314001, China), AuthorCompanyExt(id=1319955589090889773, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589078306859, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1嘉兴大学人工智能学院,浙江 嘉兴 314001)]), AuthorCompany(id=1319955589162192942, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, xref=2, ext=[AuthorCompanyExt(id=1319955589170581551, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589162192942, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2School of Information Science and Engineering, Zhejiang Sci-Tech University, Hangzhou, Zhejiang 310018, China), AuthorCompanyExt(id=1319955589178970160, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589162192942, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2浙江理工大学信息科学与工程学院,浙江 杭州 310018)])]), Author(id=1319955589518708792, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1319955589644537915, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, authorId=1319955589518708792, language=EN, stringName=Biao Tang, firstName=Biao, middleName=null, lastName=Tang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1College of Artificial Intelligence, Jiaxing University, Jiaxing, Zhejiang 314001, China
2School of Information Science and Engineering, Zhejiang Sci-Tech University, Hangzhou, Zhejiang 310018, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1319955589761978428, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, authorId=1319955589518708792, language=CN, stringName=唐彪, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1嘉兴大学人工智能学院,浙江 嘉兴 314001
2浙江理工大学信息科学与工程学院,浙江 杭州 310018, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1319955589078306859, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, xref=1, ext=[AuthorCompanyExt(id=1319955589082501164, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589078306859, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Artificial Intelligence, Jiaxing University, Jiaxing, Zhejiang 314001, China), AuthorCompanyExt(id=1319955589090889773, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589078306859, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1嘉兴大学人工智能学院,浙江 嘉兴 314001)]), AuthorCompany(id=1319955589162192942, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, xref=2, ext=[AuthorCompanyExt(id=1319955589170581551, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589162192942, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2School of Information Science and Engineering, Zhejiang Sci-Tech University, Hangzhou, Zhejiang 310018, China), AuthorCompanyExt(id=1319955589178970160, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589162192942, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2浙江理工大学信息科学与工程学院,浙江 杭州 310018)])]), Author(id=1319955591448088638, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1319955591586500672, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, authorId=1319955591448088638, language=EN, stringName=Shuyun Luo, firstName=Shuyun, middleName=null, lastName=Luo, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2School of Information Science and Engineering, Zhejiang Sci-Tech University, Hangzhou, Zhejiang 310018, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1319955591695552577, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, authorId=1319955591448088638, language=CN, stringName=骆淑云, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2浙江理工大学信息科学与工程学院,浙江 杭州 310018, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1319955589162192942, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, xref=2, ext=[AuthorCompanyExt(id=1319955589170581551, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589162192942, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2School of Information Science and Engineering, Zhejiang Sci-Tech University, Hangzhou, Zhejiang 310018, China), AuthorCompanyExt(id=1319955589178970160, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589162192942, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2浙江理工大学信息科学与工程学院,浙江 杭州 310018)])]), Author(id=1319955591775244355, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1319955591871713350, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, authorId=1319955591775244355, language=EN, stringName=Nuolei Liang, firstName=Nuolei, middleName=null, lastName=Liang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1College of Artificial Intelligence, Jiaxing University, Jiaxing, Zhejiang 314001, China
2School of Information Science and Engineering, Zhejiang Sci-Tech University, Hangzhou, Zhejiang 310018, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1319955591972376647, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, authorId=1319955591775244355, language=CN, stringName=梁诺雷, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1嘉兴大学人工智能学院,浙江 嘉兴 314001
2浙江理工大学信息科学与工程学院,浙江 杭州 310018, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1319955589078306859, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, xref=1, ext=[AuthorCompanyExt(id=1319955589082501164, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589078306859, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Artificial Intelligence, Jiaxing University, Jiaxing, Zhejiang 314001, China), AuthorCompanyExt(id=1319955589090889773, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589078306859, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1嘉兴大学人工智能学院,浙江 嘉兴 314001)]), AuthorCompany(id=1319955589162192942, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, xref=2, ext=[AuthorCompanyExt(id=1319955589170581551, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589162192942, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2School of Information Science and Engineering, Zhejiang Sci-Tech University, Hangzhou, Zhejiang 310018, China), AuthorCompanyExt(id=1319955589178970160, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589162192942, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2浙江理工大学信息科学与工程学院,浙江 杭州 310018)])]), Author(id=1319955592060457033, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, orderNo=4, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1319955592152731723, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, authorId=1319955592060457033, language=EN, stringName=Chunhua Zhang, firstName=Chunhua, middleName=null, lastName=Zhang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1College of Artificial Intelligence, Jiaxing University, Jiaxing, Zhejiang 314001, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1319955592219840588, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, authorId=1319955592060457033, language=CN, stringName=张春花, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1嘉兴大学人工智能学院,浙江 嘉兴 314001, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1319955589078306859, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, xref=1, ext=[AuthorCompanyExt(id=1319955589082501164, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589078306859, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Artificial Intelligence, Jiaxing University, Jiaxing, Zhejiang 314001, China), AuthorCompanyExt(id=1319955589090889773, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589078306859, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1嘉兴大学人工智能学院,浙江 嘉兴 314001)])])], keywords=[Keyword(id=1319955592341475405, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=EN, orderNo=1, keyword=FFC-YOLO), Keyword(id=1319955592412778574, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=EN, orderNo=2, keyword=small target detection), Keyword(id=1319955592505053263, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=EN, orderNo=3, keyword=UAV aerial image), Keyword(id=1319955592597327952, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=EN, orderNo=4, keyword=FreqFusion), Keyword(id=1319955592697991249, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=EN, orderNo=5, keyword=YOLOv11n), Keyword(id=1319955593008369746, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=CN, orderNo=1, keyword=FFC-YOLO), Keyword(id=1319955593364885588, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=CN, orderNo=2, keyword=小目标检测), Keyword(id=1319955593461354581, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=CN, orderNo=3, keyword=无人机航拍图像), Keyword(id=1319955593801093206, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=CN, orderNo=4, keyword=频域感知特征融合), Keyword(id=1319955594249883735, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=CN, orderNo=5, keyword=YOLOv11n)], refs=[Reference(id=1319955598846840949, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=1, rfOrder=0, authorNames=null, journalName=null, refType=null, unstructuredReference=李心仪, 王振飞, 吴晗. 面向仿真场景的无人机目标跟踪算法研究[J]. 系统仿真学报, 2025., articleTitle=null, refAbstract=null), Reference(id=1319955600537145462, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=1, rfOrder=1, authorNames=null, journalName=null, refType=null, unstructuredReference=Li X Y, Wang Z F, Wu H. Research on UAV target tracking algorithm for simulation scenarios[J]. J Syst Simul, 2025., articleTitle=null, refAbstract=null), Reference(id=1319955600692334711, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=2, rfOrder=2, authorNames=null, journalName=null, refType=null, unstructuredReference=曹凯, 李宁, 王立志, 等. 基于单眼视觉的巡检无人机避障策略研究[J]. 电子器件, 2025, 48(1): 86−90., articleTitle=null, refAbstract=null), Reference(id=1319955600851718264, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=2, rfOrder=3, authorNames=null, journalName=null, refType=null, unstructuredReference=Cao K, Li N, Wang L Z, et al. Research on obstacle avoidance strategy of patrol UAV based on monocular vision[J]. Chin J Electron Science Devices, 2025, 48(1): 86−90., articleTitle=null, refAbstract=null), Reference(id=1319955600923021433, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=3, rfOrder=4, authorNames=null, journalName=null, refType=null, unstructuredReference=庄瑜, 傅晓锦, 李莎, 等. 基于XMB-YOLOv5s的无人机小目标检测[J]. 计算机与现代化, 2025, (4): 29−35,41., articleTitle=null, refAbstract=null), Reference(id=1319955601002713210, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=3, rfOrder=5, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhuang Y, Fu X J, Li S, et al. UAV small target detection based on XMB-YOLOv5s[J]. Comput Mod, 2025, (4): 29−35,41., articleTitle=null, refAbstract=null), Reference(id=1319955601069822075, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=4, rfOrder=6, authorNames=null, journalName=null, refType=null, unstructuredReference=Ghen M Q, Yu L J, Zhi C, et al. Improved faster R-CNN for fabric defect detection based on Gabor filter with genetic algorithm optimization[J]. Comput Ind, 2022, 134: 103551., articleTitle=null, refAbstract=null), Reference(id=1319955601162096764, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=5, rfOrder=7, authorNames=null, journalName=null, refType=null, unstructuredReference=Terven J, Córdova-Esparza D M, Romero-González J A. A comprehensive review of YOLO architectures incomputer vision: from YOLOv1 to YOLOv8 and YOLO-NAS[J]. Mach Learn Knowl Extr, 2023, 5(4): 1680−1716., articleTitle=null, refAbstract=null), Reference(id=1319955601241788541, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=6, rfOrder=8, authorNames=null, journalName=null, refType=null, unstructuredReference=Wang A, Chen H, Liu L H, et al. YOLOv10: real-time end-to-end object detection[C]//Proceedings of the 38th International Conference on Neural Information Processing Systems, 2025: 3429., articleTitle=null, refAbstract=null), Reference(id=1319955601313091710, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=7, rfOrder=9, authorNames=null, journalName=null, refType=null, unstructuredReference=Khanam R, Hussain M. YOLOv11: an overview of the key architectural enhancements[Z]. arXiv: 2410.17725, 2024. https://arxiv.org/abs/2410.17725., articleTitle=null, refAbstract=null), Reference(id=1319955601384394879, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=8, rfOrder=10, authorNames=null, journalName=null, refType=null, unstructuredReference=张杨, 程正东, 朱斌. 基于改进Faster R-CNN的无人机小目标检测算法[J]. 量子电子学报, 2022, 39(3): 354−363., articleTitle=null, refAbstract=null), Reference(id=1319955601447309440, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=8, rfOrder=11, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhang Y, Cheng Z D, Zhu B. Small target detection algorithm of drones based on improved faster R-CNN[J]. Chin J Quantum Electron, 2022, 39(3): 354−363., articleTitle=null, refAbstract=null), Reference(id=1319955601589915777, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=9, rfOrder=12, authorNames=null, journalName=null, refType=null, unstructuredReference=李钟华, 林初俊, 朱恒亮, 等. 基于结构感知和全局上下文信息的小目标检测[J]. 计算机工程与应用, 2024, 60(9): 292−298., articleTitle=null, refAbstract=null), Reference(id=1319955601673801858, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=9, rfOrder=13, authorNames=null, journalName=null, refType=null, unstructuredReference=Li Z H, Lin C J, Zhu H L, et al. Small object detection based on structure perception and global context information[J]. Comput Eng Appl, 2024, 60(9): 292−298., articleTitle=null, refAbstract=null), Reference(id=1319955601745105027, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=10, rfOrder=14, authorNames=null, journalName=null, refType=null, unstructuredReference=李利霞, 王鑫, 王军, 等. 基于特征融合与注意力机制的无人机图像小目标检测算法[J]. 图学学报, 2023, 44(4): 658−666., articleTitle=null, refAbstract=null), Reference(id=1319955601828991108, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=10, rfOrder=15, authorNames=null, journalName=null, refType=null, unstructuredReference=Li L X, Wang X, Wang J, et al. Small object detection algorithm in UAV image based on feature fusion and attention mechanism[J]. J Graphics, 2023, 44(4): 658−666., articleTitle=null, refAbstract=null), Reference(id=1319955601954820229, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=11, rfOrder=16, authorNames=null, journalName=null, refType=null, unstructuredReference=韩俊, 袁小平, 王准, 等. 基于YOLOv5s的无人机密集小目标检测算法[J]. 浙江大学学报(工学版), 2023, 57(6): 1224−1233., articleTitle=null, refAbstract=null), Reference(id=1319955602076455046, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=11, rfOrder=17, authorNames=null, journalName=null, refType=null, unstructuredReference=Han J, Yuan X P, Wang Z, et al. UAV dense small target detection algorithm based on YOLOv5s[J]. J Zhejiang Univ (Eng Sci), 2023, 57(6): 1224−1233., articleTitle=null, refAbstract=null), Reference(id=1319955602160341127, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=12, rfOrder=18, authorNames=null, journalName=null, refType=null, unstructuredReference=牛为华, 魏雅丽. 基于改进YOLOv7的航拍小目标检测算法[J]. 电光与控制, 2024, 31(1): 117−122., articleTitle=null, refAbstract=null), Reference(id=1319955602235838600, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=12, rfOrder=19, authorNames=null, journalName=null, refType=null, unstructuredReference=Niu W H, Wei Y L. Small target detection in aerial photography images based on improved YOLOv7 algorithm[J]. Electron Opt Control, 2024, 31(1): 117−122., articleTitle=null, refAbstract=null), Reference(id=1319955602344890505, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=13, rfOrder=20, authorNames=null, journalName=null, refType=null, unstructuredReference=常凯旋, 黄建华, 孙希延, 等. 基于双模态图像融合的无人机光学小目标检测算法[J]. 激光与光电子学进展, 2025, 62(4): 0428001., articleTitle=null, refAbstract=null), Reference(id=1319955602441359498, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=13, rfOrder=21, authorNames=null, journalName=null, refType=null, unstructuredReference=Chang K X, Huang J H, Sun X Y, et al. Optical small target detection method by drone based on dual-modal image fusion[J]. Laser Optoelectron Prog, 2025, 62(4): 0428001., articleTitle=null, refAbstract=null), Reference(id=1319955602562994315, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=14, rfOrder=22, authorNames=null, journalName=null, refType=null, unstructuredReference=郑华伟, 王飞, 高建邦. DES-YOLO: 一种更精确的目标检测方法[J]. 光电工程, 2024, 51(11): 240212., articleTitle=null, refAbstract=null), Reference(id=1319955602672046220, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=14, rfOrder=23, authorNames=null, journalName=null, refType=null, unstructuredReference=Zheng H W, Wang F, Gao J B. DES-YOLO: a more accurate object detection method[J]. Opto-Electron Eng, 2024, 51(11): 240212., articleTitle=null, refAbstract=null), Reference(id=1319955602751737997, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=15, rfOrder=24, authorNames=null, journalName=null, refType=null, unstructuredReference=张润梅, 肖钰霏, 贾振楠, 等. 改进YOLOv7的无人机视角下复杂环境目标检测算法[J]. 光电工程, 2024, 51(5): 240051., articleTitle=null, refAbstract=null), Reference(id=1319955602831429774, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=15, rfOrder=25, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhang R M, Xiao Y F, Jia Z N, et al. Improved YOLOv7 algorithm for target detection in complex environments from UAV perspective[J]. Opto-Electron Eng, 2024, 51(5): 240051., articleTitle=null, refAbstract=null), Reference(id=1319955602902732943, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=16, rfOrder=26, authorNames=null, journalName=null, refType=null, unstructuredReference=唐彪, 贾小军, 骆淑云, 等. BL-YOLO: 无人机航拍图像目标检测算法[J]. 光电子·激光, 2025. http://kns.cnki.net/kcms/detail/12.1182.o4.20250423.1727.002.html., articleTitle=null, refAbstract=null), Reference(id=1319955602999201936, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=16, rfOrder=27, authorNames=null, journalName=null, refType=null, unstructuredReference=Tang B, Jia X J, Luo S Y, et al. BL-YOLO: a target detection algorithm for aerial images captured by drones[J]. J Optoelectron·Laser, 2025. http://kns.cnki.net/kcms/detail/12.1182.o4.20250423.1727.002.html., articleTitle=null, refAbstract=null), Reference(id=1319955603108253841, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=17, rfOrder=28, authorNames=null, journalName=null, refType=null, unstructuredReference=马滔, 熊英灼, 潘庆娜, 等. 基于VSA-YOLO 11n的无人机红外小目标检测[J]. 红外技术, 2025. http://kns.cnki.net/kcms/detail/53.1053.tn.20250509.1602.002.html., articleTitle=null, refAbstract=null), Reference(id=1319955603229888658, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=17, rfOrder=29, authorNames=null, journalName=null, refType=null, unstructuredReference=Ma T, Xiong Y Z, Pan Q N, et al. Infrared small object detection for UAV imagery based on VSA-YOLO11n[J]. Infrared Technol, 2025. http://kns.cnki.net/kcms/detail/53.1053.tn.20250509.1602.002.html., articleTitle=null, refAbstract=null), Reference(id=1319955603322163347, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=18, rfOrder=30, authorNames=null, journalName=null, refType=null, unstructuredReference=叶荣, 马自飞, 高泉, 等. 基于改进YOLOv5s-ECA-ASFF算法的茶叶病害目标检测[J]. 中国农机化学报, 2024, 45(1): 244−251., articleTitle=null, refAbstract=null), Reference(id=1319955605033439380, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=18, rfOrder=31, authorNames=null, journalName=null, refType=null, unstructuredReference=Ye R, Ma Z F, Gao Q, et al. Target detection of tea disease based on improved YOLOv5s-ECA-ASFF algorithm[J]. J Chin Agric Mech, 2024, 45(1): 244−251., articleTitle=null, refAbstract=null), Reference(id=1319955605176045717, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=19, rfOrder=32, authorNames=null, journalName=null, refType=null, unstructuredReference=Chen L W, Fu Y, Gu L, Yan C, et al. Frequency-aware feature fusion for dense image prediction[J]. IEEE Trans Pattern Anal Mach Intell, 2024, 46(12): 10763−10780., articleTitle=null, refAbstract=null), Reference(id=1319955605343817878, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=20, rfOrder=33, authorNames=null, journalName=null, refType=null, unstructuredReference=Cai X H, Lai Q X, Wang Y W, et al. Poly kernel inception network for remote sensing detection[C]//Proceedings of the 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024: 27706–27716. https://doi.org/10.1109/CVPR52733.2024.02617., articleTitle=null, refAbstract=null), Reference(id=1319955605406732439, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=21, rfOrder=34, authorNames=null, journalName=null, refType=null, unstructuredReference=Tan M X, Pang R M, Le Q V. EfficientDet: scalable and efficient object detection[C]//Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020: 10778–10787. https://doi.org/10.1109/CVPR42600.2020.01079., articleTitle=null, refAbstract=null)], funds=null, companyList=[AuthorCompany(id=1319955589078306859, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, xref=1, ext=[AuthorCompanyExt(id=1319955589082501164, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589078306859, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Artificial Intelligence, Jiaxing University, Jiaxing, Zhejiang 314001, China), AuthorCompanyExt(id=1319955589090889773, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589078306859, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1嘉兴大学人工智能学院,浙江 嘉兴 314001)]), AuthorCompany(id=1319955589162192942, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, xref=2, ext=[AuthorCompanyExt(id=1319955589170581551, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589162192942, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2School of Information Science and Engineering, Zhejiang Sci-Tech University, Hangzhou, Zhejiang 310018, China), AuthorCompanyExt(id=1319955589178970160, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589162192942, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2浙江理工大学信息科学与工程学院,浙江 杭州 310018)])], figs=[ArticleFig(id=1319955594686091353, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=EN, label=Fig.1, caption=FFC-YOLO network structure, figureFileSmall=kKwin+AV8MqBAjl/882D2Q==, figureFileBig=PJKCqEwQzKi/W7RiaII1Xw==, tableContent=null), ArticleFig(id=1319955596061823066, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=CN, label=图1, caption=FFC-YOLO网络结构, figureFileSmall=kKwin+AV8MqBAjl/882D2Q==, figureFileBig=PJKCqEwQzKi/W7RiaII1Xw==, tableContent=null), ArticleFig(id=1319955596405755995, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=EN, label=Fig.2, caption=FASFF structure, figureFileSmall=4jSFIY97ujpGnB3r2FFAEA==, figureFileBig=BMXZOI7OBKSJgFNlc0ZbLg==, tableContent=null), ArticleFig(id=1319955596489642076, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=CN, label=图2, caption=FASFF结构, figureFileSmall=4jSFIY97ujpGnB3r2FFAEA==, figureFileBig=BMXZOI7OBKSJgFNlc0ZbLg==, tableContent=null), ArticleFig(id=1319955596573528157, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=EN, label=Fig.3, caption=The structural composition of the key components of FreqFusion;(a) ALPF generator; (b) Offset generator; (c) AHPF generator, figureFileSmall=N6IYSGVUiyBaIG10ZKUPzQ==, figureFileBig=lAh0+gDLNFK7gDDc4302Tg==, tableContent=null), ArticleFig(id=1319955596657414238, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=CN, label=图3, caption=FreqFusion主要组件结构;(a) ALPF生成器;(b) Offset生成器;(c) AHPF生成器, figureFileSmall=N6IYSGVUiyBaIG10ZKUPzQ==, figureFileBig=lAh0+gDLNFK7gDDc4302Tg==, tableContent=null), ArticleFig(id=1319955596774854751, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=EN, label=Fig.4, caption=FreqFusion structure, figureFileSmall=sGDZJ79OEriES5wG52ga5g==, figureFileBig=8mk6o4lIZ2EbL3poV+x6Mw==, tableContent=null), ArticleFig(id=1319955596862935136, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=CN, label=图4, caption=FreqFusion结构, figureFileSmall=sGDZJ79OEriES5wG52ga5g==, figureFileBig=8mk6o4lIZ2EbL3poV+x6Mw==, tableContent=null), ArticleFig(id=1319955596988764257, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=EN, label=Fig.5, caption=CAA structure, figureFileSmall=qCDURvneAzNjkOQpvWDgMA==, figureFileBig=1QpfPxkVdsGdVm88t8LSwA==, tableContent=null), ArticleFig(id=1319955597227839586, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=CN, label=图5, caption=CAA结构, figureFileSmall=qCDURvneAzNjkOQpvWDgMA==, figureFileBig=1QpfPxkVdsGdVm88t8LSwA==, tableContent=null), ArticleFig(id=1319955597303337059, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=EN, label=Fig.6, caption=Comparison of mAP@0.5 values for ten recognition targets; (a) Detection results of YOLO v11n; (b) Detection results of FFC-YOLO, figureFileSmall=eV3X0RGZB7KS4henJ0CLHA==, figureFileBig=SiLxmXBUtKjxW+sc4I/uSw==, tableContent=null), ArticleFig(id=1319955597370445924, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=CN, label=图6, caption=十种识别目标mAP@0.5值对比。(a) YOLOv11n检测结果;(b) FFC-YOLO检测结果, figureFileSmall=eV3X0RGZB7KS4henJ0CLHA==, figureFileBig=SiLxmXBUtKjxW+sc4I/uSw==, tableContent=null), ArticleFig(id=1319955597462720613, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=EN, label=Fig.7, caption=Detection results between YOLOv11n and FFC-YOLO. (a-c) Detection results of the YOLOv10n; (d-f) Detection results of the FFC-YOLO, figureFileSmall=5lavJR78RKTgG/MkkSnSEQ==, figureFileBig=NgSGX6DNh6Gv7LQvlj4D4g==, tableContent=null), ArticleFig(id=1319955597521440870, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=CN, label=图7, caption=YOLOv11n与FFC-YOLO检测结果。(a-c) YOLOv11n检测结果;(d-f) FFC-YOLO检测结果, figureFileSmall=5lavJR78RKTgG/MkkSnSEQ==, figureFileBig=NgSGX6DNh6Gv7LQvlj4D4g==, tableContent=null), ArticleFig(id=1319955597588549735, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=EN, label=Fig.8, caption=Detection results of YOLOv11n and FFC-YOLO on tiny-data. (a, b) The detection results of YOLOv11n; (c, d) The detection results of FFC-YOLO, figureFileSmall=6VSQCBsM7X7/cVEZNqGf7Q==, figureFileBig=LbxZosBt6z29gt44cheLKQ==, tableContent=null), ArticleFig(id=1319955597743738984, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=CN, label=图8, caption=YOLOv11n与FFC-YOLO对于tiny-data的检测结果。(a, b) YOLOv11n检测结果;(c, d) FFC-YOLO检测结果, figureFileSmall=6VSQCBsM7X7/cVEZNqGf7Q==, figureFileBig=LbxZosBt6z29gt44cheLKQ==, tableContent=null), ArticleFig(id=1319955597806653545, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=EN, label=Tab.1, caption=

VisDrone2019 dataset

, figureFileSmall=null, figureFileBig=null, tableContent=
ClassNumber
Pedestrian79337
People27059
Bicycle10477
Car144865
Van24950
Truck12871
Tricycle4803
Awning-tricycle3243
Bus5926
Motor29642
), ArticleFig(id=1319955597877956714, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=CN, label=表1, caption=

VisDrone2019数据集

, figureFileSmall=null, figureFileBig=null, tableContent=
ClassNumber
Pedestrian79337
People27059
Bicycle10477
Car144865
Van24950
Truck12871
Tricycle4803
Awning-tricycle3243
Bus5926
Motor29642
), ArticleFig(id=1319955597949259883, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=EN, label=Tab.2, caption=

Detection results of different networks on VisDrone2019 dataset

, figureFileSmall=null, figureFileBig=null, tableContent=
NetworksmAP@0.5/%Params/MP/%R/%FPS
Fast-RCNN11.628.412.210.36
RetinaNet21.419.826.934.222
YOLOv3-tiny23.19.5239.529.4124
YOLOv5n31.22.1842.131.3138
YOLOv8n32.02.6942.332.6147
YOLOv10n32.42.6942.232.9136
YOLOv11n32.22.5841.532.9126
YOLOv11s39.319.849.337.9132
FFC-YOLO40.04.0249.038.8106
), ArticleFig(id=1319955598007980140, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=CN, label=表2, caption=

不同网络在VisDrone2019数据集上的检测结果

, figureFileSmall=null, figureFileBig=null, tableContent=
NetworksmAP@0.5/%Params/MP/%R/%FPS
Fast-RCNN11.628.412.210.36
RetinaNet21.419.826.934.222
YOLOv3-tiny23.19.5239.529.4124
YOLOv5n31.22.1842.131.3138
YOLOv8n32.02.6942.332.6147
YOLOv10n32.42.6942.232.9136
YOLOv11n32.22.5841.532.9126
YOLOv11s39.319.849.337.9132
FFC-YOLO40.04.0249.038.8106
), ArticleFig(id=1319955598112837741, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=EN, label=Tab.3, caption=

Dense small object detection mAP@0.5 results comparison

, figureFileSmall=null, figureFileBig=null, tableContent=
ClassmAP@0.5/%Improved value
YOLOv11nFFC-YOLO
Pedestrian34.146.412.3
People26.537.511.0
Bicycle7.313.46.1
Car75.382.67.3
Van38.345.37.0
Truck27.832.64.8
Tricycle20.225.04.8
Awning-tricycle11.514.32.8
Bus45.854.78.9
Motor35.148.113.0
), ArticleFig(id=1319955598226083950, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=CN, label=表3, caption=

密集小目标检测mAP@0.5值结果对比

, figureFileSmall=null, figureFileBig=null, tableContent=
ClassmAP@0.5/%Improved value
YOLOv11nFFC-YOLO
Pedestrian34.146.412.3
People26.537.511.0
Bicycle7.313.46.1
Car75.382.67.3
Van38.345.37.0
Truck27.832.64.8
Tricycle20.225.04.8
Awning-tricycle11.514.32.8
Bus45.854.78.9
Motor35.148.113.0
), ArticleFig(id=1319955598347718767, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=EN, label=Tab.4, caption=

Comparative experiment results with adding different attention mechanisms or model optimization modules

, figureFileSmall=null, figureFileBig=null, tableContent=
TypemAP@0.5/%Params/MP/%R/%
Base39.44.0448.839.0
Base+caa39.74.2243.132.9
Base+C2PSA-CAA40.04.0249.038.8
Base+C2PSA-CBAM39.64.0448.237.8
Base+C2PSA-ECA39.84.0249.138.2
Base+C2PSA-iRMB39.84.0650.338.2
Base+C2PSA-EMA39.74.0048.838.8
), ArticleFig(id=1319955598439993456, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=CN, label=表4, caption=

添加不同注意力机制及模型优化模块的对比实验结果

, figureFileSmall=null, figureFileBig=null, tableContent=
TypemAP@0.5/%Params/MP/%R/%
Base39.44.0448.839.0
Base+caa39.74.2243.132.9
Base+C2PSA-CAA40.04.0249.038.8
Base+C2PSA-CBAM39.64.0448.237.8
Base+C2PSA-ECA39.84.0249.138.2
Base+C2PSA-iRMB39.84.0650.338.2
Base+C2PSA-EMA39.74.0048.838.8
), ArticleFig(id=1319955598540656753, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=EN, label=Tab.5, caption=

Detection results of FFC-YOLO on tiny-data dataset

, figureFileSmall=null, figureFileBig=null, tableContent=
NetworksClassmAP@0.5/%P/%R/%
YOLO v11nSperson67.564.465.8
Lperson62.061.160.8
Wperson49.157.645.6
All59.561.057.4
FFC-YOLOSperson75.473.369.9
Lperson73.067.373.2
Wperson57.868.446.9
All68.769.763.3
), ArticleFig(id=1319955598616154226, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=CN, label=表5, caption=

FFC-YOLO在tiny-data数据集上的检测结果

, figureFileSmall=null, figureFileBig=null, tableContent=
NetworksClassmAP@0.5/%P/%R/%
YOLO v11nSperson67.564.465.8
Lperson62.061.160.8
Wperson49.157.645.6
All59.561.057.4
FFC-YOLOSperson75.473.369.9
Lperson73.067.373.2
Wperson57.868.446.9
All68.769.763.3
), ArticleFig(id=1319955598687457395, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=EN, label=Tab.6, caption=

Ablation experiment results

, figureFileSmall=null, figureFileBig=null, tableContent=
NetworkFASFFFreqFusionC2PSA-
CAA
mAP@0.5
/%
Params/
M
P/%R/%
Baseline32.22.5841.532.9
A√37.24.0547.336.5
B√32.72.4443.132.9
C√√39.44.0448.839.0
D√√32.62.4243.732.7
E√√37.04.0447.536.2
F√√√40.04.0249.038.8
), ArticleFig(id=1319955598758760564, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, language=CN, label=表6, caption=

消融实验结果

, figureFileSmall=null, figureFileBig=null, tableContent=
NetworkFASFFFreqFusionC2PSA-
CAA
mAP@0.5
/%
Params/
M
P/%R/%
Baseline32.22.5841.532.9
A√37.24.0547.336.5
B√32.72.4443.132.9
C√√39.44.0448.839.0
D√√32.62.4243.732.7
E√√37.04.0447.536.2
F√√√40.04.0249.038.8
)], attaches=null, journal=Journal(id=1278641367198941188, delFlag=0, nameCn=光电工程, nameEn=Opto-Electronic Engineering, nameHistory1=null, nameHistory2=null, issn=1003-501X, eissn=2097-4019, cn=51-1346/O4, coden=null, periodic=0, language=CN, oaType=null, ccby=null, superviseOffice=null, ownerOffice=null, pubOffice=null, editorOffice=null, officeType=null, aims=null, clcCode=null, officeProv=null, officeCity=null, officeAddr=null, officeZip=null, officeEmail=null, officePhone=null, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=4Vimkd+qXLWNxtdpr9mFNw==, journalPrice=null, startedYear=null, abbrevIsoEn=Opto-Electronic Engineering, journalRemark=null, publicationField=null, createdTime=1782781457950, updatedTime=1784021784694, createdBy=18614031015, updatedBy=13041195026, firstLetterCn=G, firstLetterEn=G, subjectCode=Engineering, subjectName=null, subjectCodeEn=Engineering, subjectNameEn=null, picCn=4Vimkd+qXLWNxtdpr9mFNw==, picEn=vAq9s20WLs1ODDfbWq+Gjg==, jcr=null, cjcr=null, exts=[JournalExt(id=1283843675721536154, language=CN, name=光电工程, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1784021784954, updatedTime=1784021784954, createdBy=13041195026, updatedBy=13041195026, submissionGuidelinesUrl=, submissionAuthorUrl=http://www.manuscripts.com.cn/gdgc, submissionEditorUrl=http://www.manuscripts.com.cn/gdgc, submissionReviewUrl=http://www.manuscripts.com.cn/gdgc, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1283843675771867803, language=EN, name=Opto-Electronic Engineering, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1784021784966, updatedTime=1784021784966, createdBy=13041195026, updatedBy=13041195026, submissionGuidelinesUrl=, submissionAuthorUrl=http://www.manuscripts.com.cn/gdgc, submissionEditorUrl=http://www.manuscripts.com.cn/gdgc, submissionReviewUrl=http://www.manuscripts.com.cn/gdgc, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1278651732997652489, websiteList=[Website(id=1278723867418018151, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1278651732997652489, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/oee/CN, language=CN, createTime=1782801127533, createBy=18614031015, updateTime=1782804494442, updateBy=18614031015, name=光电工程-中文, tplId=1146099689490845704, title=光电工程, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1278738091150128034, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1278723867418018151, code=articleTextType, value=kx, createTime=1782804518735, updateTime=1782804518735, creator=18614031015, updator=18614031015), WebsiteProps(id=1278738091120767903, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1278723867418018151, code=banner, value=null, createTime=1782804518728, updateTime=1782804518728, creator=18614031015, updator=18614031015), WebsiteProps(id=1278738091171099557, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1278723867418018151, code=grayFlag, value=0, createTime=1782804518740, updateTime=1782804518740, creator=18614031015, updator=18614031015), WebsiteProps(id=1278738091108184990, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1278723867418018151, code=logo, value=https://castjournals.cast.org.cn/joweb/oee/CN/file/pic?fileId=A1C6uwqtMazluiWkEpR0Mg==, createTime=1782804518725, updateTime=1782804518725, creator=18614031015, updator=18614031015), WebsiteProps(id=1278738091179488167, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1278723867418018151, code=minRunFlag, value=0, createTime=1782804518742, updateTime=1782804518742, creator=18614031015, updator=18614031015), WebsiteProps(id=1278738091141739425, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1278723867418018151, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/oee/CN/file/pic, createTime=1782804518733, updateTime=1782804518733, creator=18614031015, updator=18614031015), WebsiteProps(id=1278738091175293862, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1278723867418018151, code=silenceFlag, value=0, createTime=1782804518741, updateTime=1782804518741, creator=18614031015, updator=18614031015), WebsiteProps(id=1278738091129156512, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1278723867418018151, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1782804518730, updateTime=1782804518730, creator=18614031015, updator=18614031015), WebsiteProps(id=1278738091154322339, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1278723867418018151, code=themeColor, value=null, createTime=1782804518736, updateTime=1782804518736, creator=18614031015, updator=18614031015), WebsiteProps(id=1278738091162710948, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1278723867418018151, code=themeStyle, value=null, createTime=1782804518738, updateTime=1782804518738, creator=18614031015, updator=18614031015)]), Website(id=1278723867522875769, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1278651732997652489, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/oee/EN, language=EN, createTime=1782801127558, createBy=18614031015, updateTime=1782804490442, updateBy=18614031015, name=光电工程-英文, tplId=1146101810881728533, title=Opto-Electronic Engineering, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1278738063660659607, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1278723867522875769, code=articleTextType, value=kx, createTime=1782804512181, updateTime=1782804512181, creator=18614031015, updator=18614031015), WebsiteProps(id=1278738063635493780, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1278723867522875769, code=banner, value=null, createTime=1782804512175, updateTime=1782804512175, creator=18614031015, updator=18614031015), WebsiteProps(id=1278738063924900762, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1278723867522875769, code=grayFlag, value=0, createTime=1782804512244, updateTime=1782804512244, creator=18614031015, updator=18614031015), WebsiteProps(id=1278738063606133651, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1278723867522875769, code=logo, value=https://castjournals.cast.org.cn/joweb/oee/EN/file/pic?fileId=A1C6uwqtMazluiWkEpR0Mg==, createTime=1782804512168, updateTime=1782804512168, creator=18614031015, updator=18614031015), WebsiteProps(id=1278738063937483676, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1278723867522875769, code=minRunFlag, value=0, createTime=1782804512247, updateTime=1782804512247, creator=18614031015, updator=18614031015), WebsiteProps(id=1278738063652270998, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1278723867522875769, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/oee/EN/file/pic, createTime=1782804512179, updateTime=1782804512179, creator=18614031015, updator=18614031015), WebsiteProps(id=1278738063933289371, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1278723867522875769, code=silenceFlag, value=0, createTime=1782804512246, updateTime=1782804512246, creator=18614031015, updator=18614031015), WebsiteProps(id=1278738063639688085, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1278723867522875769, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1782804512176, updateTime=1782804512176, creator=18614031015, updator=18614031015), WebsiteProps(id=1278738063664853912, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1278723867522875769, code=themeColor, value=null, createTime=1782804512182, updateTime=1782804512182, creator=18614031015, updator=18614031015), WebsiteProps(id=1278738063916512153, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1278723867522875769, code=themeStyle, value=null, createTime=1782804512242, updateTime=1782804512242, creator=18614031015, updator=18614031015)])], journalTitle=光电工程, weixinUrl=null, journalUrl=https://www.oejournal.org/oee, iacademicId=null, status=1, seqNo=null, journalTitleEn=Opto-Electronic Engineering, journalPhotoCn=4Vimkd+qXLWNxtdpr9mFNw==, journalPhotoEn=vAq9s20WLs1ODDfbWq+Gjg==, journalFirstLetter=G, journalRecommend=null, journalNew=null, journalCollection=null, jcrJf=null, cjcrJf=null, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=, provinceCode=null, provinceName=null, collectFlag=false, interPubPlatform=, interPubPlatformUrl=null), detailUrlCn=https://castjournals.cast.org.cn/joweb/oee/CN/10.12086/oee.2026.250208, detailUrlEn=https://castjournals.cast.org.cn/joweb/oee/EN/10.12086/oee.2026.250208, pdfUrlCn=https://castjournals.cast.org.cn/joweb/oee/CN/PDF/10.12086/oee.2026.250208, pdfUrlEn=https://castjournals.cast.org.cn/joweb/oee/EN/PDF/10.12086/oee.2026.250208, aliStartDate=0, aliEndDate=0, collectionFlag=false, citedCount=null, citedUrl=null, previewStatus=0, delFlag=0, hasFullText=1, orderTime=1771948800000, fullTextJson=null, articleText=null, reference=null)
收藏切换
基于频域感知特征融合的无人机航拍图像小目标检测算法
收藏切换
PDF下载
贾小军 1, 2, *, * , 唐彪 1, 2 , 骆淑云 2 , 梁诺雷 1, 2 , 张春花 1
光电工程 | 研究论文 2026,53(2): 250208-1
收起
收藏切换
光电工程 |研究论文 2026 , 53 (2) : 250208 -1
基于频域感知特征融合的无人机航拍图像小目标检测算法
全屏
[Author(id=1319955589258661938, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=xjjiad@sina.com, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1319955589346742325, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, authorId=1319955589258661938, language=EN, stringName=Xiaojun Jia, firstName=Xiaojun, middleName=null, lastName=Jia, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1College of Artificial Intelligence, Jiaxing University, Jiaxing, Zhejiang 314001, China
2School of Information Science and Engineering, Zhejiang Sci-Tech University, Hangzhou, Zhejiang 310018, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1319955589426434102, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, authorId=1319955589258661938, language=CN, stringName=贾小军, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, *, *, address=1嘉兴大学人工智能学院,浙江 嘉兴 314001
2浙江理工大学信息科学与工程学院,浙江 杭州 310018, bio={"img":"nnI7zyTp5E5cjxsBk2UQIQ==","content":"

贾小军(1974-),四川邻水人,嘉兴大学教授,研究方向为计算机视觉与模式识别,图像处理。E-mail:

"}, bioImg=nnI7zyTp5E5cjxsBk2UQIQ==, bioContent=

贾小军(1974-),四川邻水人,嘉兴大学教授,研究方向为计算机视觉与模式识别,图像处理。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1319955589078306859, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, xref=1, ext=[AuthorCompanyExt(id=1319955589082501164, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589078306859, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Artificial Intelligence, Jiaxing University, Jiaxing, Zhejiang 314001, China), AuthorCompanyExt(id=1319955589090889773, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589078306859, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1嘉兴大学人工智能学院,浙江 嘉兴 314001)]), AuthorCompany(id=1319955589162192942, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, xref=2, ext=[AuthorCompanyExt(id=1319955589170581551, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589162192942, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2School of Information Science and Engineering, Zhejiang Sci-Tech University, Hangzhou, Zhejiang 310018, China), AuthorCompanyExt(id=1319955589178970160, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589162192942, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2浙江理工大学信息科学与工程学院,浙江 杭州 310018)])]), Author(id=1319955589518708792, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1319955589644537915, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, authorId=1319955589518708792, language=EN, stringName=Biao Tang, firstName=Biao, middleName=null, lastName=Tang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1College of Artificial Intelligence, Jiaxing University, Jiaxing, Zhejiang 314001, China
2School of Information Science and Engineering, Zhejiang Sci-Tech University, Hangzhou, Zhejiang 310018, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1319955589761978428, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, authorId=1319955589518708792, language=CN, stringName=唐彪, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1嘉兴大学人工智能学院,浙江 嘉兴 314001
2浙江理工大学信息科学与工程学院,浙江 杭州 310018, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1319955589078306859, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, xref=1, ext=[AuthorCompanyExt(id=1319955589082501164, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589078306859, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Artificial Intelligence, Jiaxing University, Jiaxing, Zhejiang 314001, China), AuthorCompanyExt(id=1319955589090889773, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589078306859, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1嘉兴大学人工智能学院,浙江 嘉兴 314001)]), AuthorCompany(id=1319955589162192942, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, xref=2, ext=[AuthorCompanyExt(id=1319955589170581551, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589162192942, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2School of Information Science and Engineering, Zhejiang Sci-Tech University, Hangzhou, Zhejiang 310018, China), AuthorCompanyExt(id=1319955589178970160, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589162192942, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2浙江理工大学信息科学与工程学院,浙江 杭州 310018)])]), Author(id=1319955591448088638, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1319955591586500672, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, authorId=1319955591448088638, language=EN, stringName=Shuyun Luo, firstName=Shuyun, middleName=null, lastName=Luo, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2School of Information Science and Engineering, Zhejiang Sci-Tech University, Hangzhou, Zhejiang 310018, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1319955591695552577, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, authorId=1319955591448088638, language=CN, stringName=骆淑云, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=2, address=2浙江理工大学信息科学与工程学院,浙江 杭州 310018, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1319955589162192942, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, xref=2, ext=[AuthorCompanyExt(id=1319955589170581551, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589162192942, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2School of Information Science and Engineering, Zhejiang Sci-Tech University, Hangzhou, Zhejiang 310018, China), AuthorCompanyExt(id=1319955589178970160, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589162192942, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2浙江理工大学信息科学与工程学院,浙江 杭州 310018)])]), Author(id=1319955591775244355, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1319955591871713350, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, authorId=1319955591775244355, language=EN, stringName=Nuolei Liang, firstName=Nuolei, middleName=null, lastName=Liang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1College of Artificial Intelligence, Jiaxing University, Jiaxing, Zhejiang 314001, China
2School of Information Science and Engineering, Zhejiang Sci-Tech University, Hangzhou, Zhejiang 310018, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1319955591972376647, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, authorId=1319955591775244355, language=CN, stringName=梁诺雷, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1嘉兴大学人工智能学院,浙江 嘉兴 314001
2浙江理工大学信息科学与工程学院,浙江 杭州 310018, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1319955589078306859, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, xref=1, ext=[AuthorCompanyExt(id=1319955589082501164, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589078306859, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Artificial Intelligence, Jiaxing University, Jiaxing, Zhejiang 314001, China), AuthorCompanyExt(id=1319955589090889773, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589078306859, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1嘉兴大学人工智能学院,浙江 嘉兴 314001)]), AuthorCompany(id=1319955589162192942, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, xref=2, ext=[AuthorCompanyExt(id=1319955589170581551, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589162192942, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2School of Information Science and Engineering, Zhejiang Sci-Tech University, Hangzhou, Zhejiang 310018, China), AuthorCompanyExt(id=1319955589178970160, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589162192942, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2浙江理工大学信息科学与工程学院,浙江 杭州 310018)])]), Author(id=1319955592060457033, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, orderNo=4, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1319955592152731723, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, authorId=1319955592060457033, language=EN, stringName=Chunhua Zhang, firstName=Chunhua, middleName=null, lastName=Zhang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1College of Artificial Intelligence, Jiaxing University, Jiaxing, Zhejiang 314001, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1319955592219840588, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, authorId=1319955592060457033, language=CN, stringName=张春花, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1嘉兴大学人工智能学院,浙江 嘉兴 314001, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1319955589078306859, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, xref=1, ext=[AuthorCompanyExt(id=1319955589082501164, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589078306859, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Artificial Intelligence, Jiaxing University, Jiaxing, Zhejiang 314001, China), AuthorCompanyExt(id=1319955589090889773, tenantId=1146029695717560320, journalId=1278651732997652489, articleId=1319955578244420547, companyId=1319955589078306859, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1嘉兴大学人工智能学院,浙江 嘉兴 314001)])])]
贾小军1, 2, *, * , 唐彪1, 2, 骆淑云2, 梁诺雷1, 2, 张春花1
作者信息
  • 1嘉兴大学人工智能学院,浙江 嘉兴 314001
  • 2浙江理工大学信息科学与工程学院,浙江 杭州 310018
通讯作者:
贾小军,
作者简介:

作者贡献

唐彪、梁诺雷:研究资料收集、分析、论文写作;贾小军、洛淑云、张春花:指导论文选题、论文修改、研究资料收集。

贾小军(1974-),四川邻水人,嘉兴大学教授,研究方向为计算机视觉与模式识别,图像处理。E-mail:

A small target detection algorithm for UAV aerial images based on frequency domain perception feature fusion
Xiaojun Jia1, 2 , Biao Tang1, 2, Shuyun Luo2, Nuolei Liang1, 2, Chunhua Zhang1
Affiliations
  • 1College of Artificial Intelligence, Jiaxing University, Jiaxing, Zhejiang 314001, China
  • 2School of Information Science and Engineering, Zhejiang Sci-Tech University, Hangzhou, Zhejiang 310018, China
  • Corresponding Author:
出版时间: 2026-02-25 doi: 10.12086/oee.2026.250208
文章导航
收藏切换

小目标检测一直是计算机视觉处理中一项极具挑战性的任务,针对传统无人机航拍图像小目标多且密集,而传统检测算法存在漏检、误检的问题,提出一种基于频域感知特征融合的FFC-YOLO (FASFF-FreqFusion-CAA YOLO)目标检测算法。首先,通过增加P2检测层,重新设计了针对小目标检测的Detect-FASFF (four adaptively spatial feature fusion)检测头结构。然后,将传统的特征融合采样方法替换为频率感知特征融合FreqFusion (frequency-aware feature fusion)方法并结合BiFPN (bidirectional feature pyramid network)结构提升了算法处理密集小目标图像的能力。最后,在C2PSA模块中加入CAA (context anchor attention)注意力机制,增强了目标上下文特征信息的关联。在VisDrone2019数据集中,基于FFC-YOLO目标检测算法的mAP@0.5为40%,较传统算法如Fast-RCNN、RetinaNet分别提升了28.4%、18.6%,较YOLO算法如YOLO v8n、v10n、v11n分别提升了8.0%、7.6%、7.8%。在自建小目标数据集tiny-data中,基于FFC-YOLO的3类目标sperson、lperson和wperson的检测结果较YOLOv11n,mAP@0.5、P、R分别提升了9.2%、8.7%、5.9%。实验结果表明,FFC-YOLO小目标检测算法能够更好地应用于无人机航拍图像检测领域。

FFC-YOLO  /  小目标检测  /  无人机航拍图像  /  频域感知特征融合  /  YOLOv11n
Objective

Small object detection remains a significant and persistent challenge within the computer vision domain. This challenge is particularly acute in the context of drone aerial imagery, where objects captured from high altitudes appear extremely small, densely packed, and often exhibit blurred features due to low resolution and environmental factors. Traditional object detection algorithms frequently suffer from performance degradation in these scenarios, manifesting as high rates of missed detections and false positives. To address these critical limitations, this paper proposes a novel object detection algorithm named FFC-YOLO (FASFF-FreqFusion-CAA YOLO), which is built upon a frequency-aware feature fusion framework.

Methods

The core of the FFC-YOLO algorithm lies in three systematic and synergistic innovations designed to enhance feature representation, fusion, and contextual understanding specifically for small and dense objects. First, to bolster multi-scale feature extraction, we redesign the detection head by incorporating a Four Adaptively Spatial Feature Fusion (Detect-FASFF) structure. This is augmented with an additional P2 detection layer dedicated to capturing finer-grained spatial information crucial for identifying minuscule targets. The FASFF mechanism leverages four adaptive branches to dynamically aggregate features across different scales, thereby significantly improving the model's sensitivity and discriminative power for small objects across varying sizes.Second, we comprehensively optimize the feature fusion pathway to counteract the information loss common in conventional sampling methods (e.g., up-sampling and down-sampling). We introduce a Frequency-aware Feature Fusion (FreqFusion) module. This module operates by decomposing features into different frequency components within the frequency domain, allowing for more deliberate and effective integration of high-frequency details (essential for edge and texture information of small objects) and low-frequency semantics. This FreqFusion module is then elegantly coupled with a Bidirectional Feature Pyramid Network (BiFPN) structure. The resulting fusion architecture enables efficient, multi-level bidirectional cross-scale information flow, effectively mitigating the problems of feature obliteration, category misclassification, and bounding box localization drift in scenes crowded with small objects.

Third, to enhance the model's ability to leverage surrounding semantic context—a vital clue when object appearance is ambiguous—we integrate a Context Anchor Attention (CAA) mechanism into the C2PSA module. This attention mechanism functions by establishing dynamic associations between anchor points and their contextual surroundings, enabling the feature extraction process to selectively focus on and amplify informative regions around potential targets. This contextual awareness strengthens feature representation and aids in distinguishing small objects from complex backgrounds.

Results and Discussions

The efficacy of the proposed FFC-YOLO algorithm is rigorously validated through extensive experiments on two benchmark datasets. On the public VisDrone2019 dataset, a standard benchmark for drone vision, FFC-YOLO achieves a mean Average Precision (mAP@0.5) of 40.0%. This represents a substantial performance gain of 8.0%, 7.6%, and 7.8% over the widely-recognized YOLOv8n, YOLOv10n, and YOLOv11n baselines, respectively. To further verify its generalization capability and prowess in detecting extremely small targets, we conducted additional experiments on a custom-built, challenging dataset named "tiny-data", which contains three categories of progressively smaller persons ("sperson", "lperson", "wperson"). On this dataset, FFC-YOLO demonstrates superior results compared to the strong YOLOv11n baseline, with improvements of 9.2% in mAP@0.5, 8.7% in Precision (P), and 5.9% in Recall (R). These consistent gains across both public and private datasets robustly confirm that the FFC-YOLO algorithm possesses strong generalization ability and excels in the demanding task of drone-based aerial image analysis, particularly for long-range and dense small object detection.

Conclusions

In conclusion, this paper presents a comprehensive and effective solution to the small object detection problem in aerial imagery. By innovating at the levels of feature extraction (FASFF head), feature fusion (FreqFusion-BiFPN), and contextual modeling (CAA attention), the FFC-YOLO framework successfully addresses key shortcomings of prior methods. The experimental evidence underscores its potential for practical application in UAV-based surveillance, traffic monitoring, and search-and-rescue operations. For future work, we plan to explore model compression techniques such as pruning and knowledge distillation to streamline the model for efficient deployment on resource-constrained edge devices, thereby bridging the gap between high accuracy and real-time performance.

FFC-YOLO  /  small target detection  /  UAV aerial image  /  FreqFusion  /  YOLOv11n
贾小军, 唐彪, 骆淑云, 梁诺雷, 张春花. 基于频域感知特征融合的无人机航拍图像小目标检测算法. 光电工程, 2026 , 53 (2) : 250208 -1 . DOI: 10.12086/oee.2026.250208
Xiaojun Jia, Biao Tang, Shuyun Luo, Nuolei Liang, Chunhua Zhang. A small target detection algorithm for UAV aerial images based on frequency domain perception feature fusion[J]. Opto-Electronic Engineering, 2026 , 53 (2) : 250208 -1 . DOI: 10.12086/oee.2026.250208
近年来,随着无人机 (unmanned aerial vehicle, UAV)技术的快速发展,航拍图像在灾害救援、农业监测、交通管理、军事侦察等领域得到了广泛应用。然而,航拍图像中的小目标检测(如车辆、行人、小型设施等)仍面临严峻挑战。这类目标通常仅占图像中极少的像素(如小于32 pixel×32 pixel),且受限于无人机飞行高度、复杂背景干扰、多尺度变化及光照条件等因素,导致目标特征模糊、信噪比低,传统检测方法难以实现精准识别。
传统基于手工特征的检测算法如HOG[1]、SIFT[2]等,因依赖人工设计特征,泛化能力不足,难以应对小目标的多样性与复杂性。随着深度学习技术的兴起,基于卷积神经网络CNN的方法显著提升了小目标检测性能。研究者通过多尺度特征融合FPN、注意力机制CBAM[3]、超分辨率重建等技术增强小目标特征表达,并优化Faster R-CNN[4]、YOLO系列[5-7]网络的结构设计和训练策略以提升检测精度。然而,如何用无人机有限的算力平衡检测精度与实时性,仍是实际部署的核心难题。
优化提高无人机航拍图像的小目标检测精度已成为计算机视觉的主流研究方向,张杨等[8]提出使用卷积层更少的ResNet-18作为骨干网络改进Faster R-CNN,优化特征融合方法增强了目标特征与背景特征的对比度。李钟华等[9]在YOLOv5框架的基础上设计了多尺度结构感知模块,并借助Transformer设计了全局上下文信息模块,提高了平均精度,但针对小目标检测的改进不足。李利霞等[10]为了增加小目标特征信息,提出浅层特征增强模块来学习底层特征信息,在VisDrone2019数据集上提升了3.5%的精度,但并未对模型做轻量化处理,精度还有提升空间。韩俊等[11]以YOLOv5为基础,构造了多尺度特征提取模块LM-fem增强特征提取能力,设计了新的混合域注意力模块S-ECA和自适应权重动态融合结构AFF,在VisDrone2021数据集上提升了6.1%的检测精度,但因未对网络模型轻量化处理,所需算力过大。牛为华等[12]针对小目标多且小目标易被噪声淹没等问题,基于YOLOv7,提出增加一个检测小目标的头部,改进特征融合方式来优化特征图的表达能力,在VisDrone2019数据集上精度提升了2.8%。常凯旋等[13]提出了一种基于双模态图像融合的算法,通过双路特征融合模块融合红外和可见光的特征和在颈部网络层引入BRA (bi-level routing attention)模块,增强了小目标的检测能力,在DroneVehicle数据集上平均精度均值提升了14.78%。郑华伟等[14]针对图像检测精度低的问题,以YOLOv5为基准网络,通过在骨干网络部分引入可变形注意力模块,网络头部加入浅层特征图等方式,在NWPU VHR-10数据集上精度达到94.3%。张润梅等[15]针对无人机航拍过程受到恶劣环境影响等问题,提出一种改进算法SSG-YOLOv7。通过引入SimAM注意力机制和GhostConv卷积模块,然后使用Soft NMS优化锚框置信度,在VisDrone2019数据集上较原算法精度提升了10.45%。唐彪等[16]以YOLOv10n为基准网络,加入一个160 pixel×160 pixel尺寸的小目标检测头并在每个检测头前加入大尺寸可分离注意力机制LSKA (large separable kernel attention),最后使用BiFPN融合更多层级特征并优化网络结构,在VisDrone2019数据集上较基准网络精度提升了4.6%。马滔等[17]将YOLOv11的骨干网络替换为VanillaNet,加入了自适应空间融合检测头,在HIT-UAV红外小目标数据集上精度达到了81.3%,但并未验证模型泛化能力,并且参数量还是较大。
综上所述,现有的无人机航拍图像小目标检测算法虽然都提高了检测精度,但都存在一定的问题。小目标的特征信息很难提取且易被背景遮挡,因此在网络中多次使用采样方法易造成特征信息丢失过多,这种情况在低分辨率图像和小目标密集图像中更为严重。为了解决以上的检测问题,本文提出一种基于频域感知特征融合的检测算法 FFC-YOLO (FASFF-FreqFusion-CAA-YOLO),该算法能够有效地提高小目标检测的精度,在自建小目标数据集上的泛化实验也验证了模型的泛化性,能够较好地适用于无人机航拍图像的小目标检测。
YOLOv11是Ultralytics团队于2024年9月发布的YOLO系列,包括YOLOv11n、s、m、l、x等。YOLOv11通过模块架构的迭代(如C3k2、C2PSA、SPFF模块)与任务导向优化,在目标检测领域实现了精度、效率与轻量化的三位一体突破。本文选用精度和体积更合适的YOLOv11n网络作为基准,优化网络结构来提高模型对于小目标的检测能力,提出一种优化算法:FFC-YOLO,旨在提升小目标的检测能力。本文的改进包括:
1) 设计自适应空间特征融合的四检测头结构FASFF (four adaptively spatial feature fusion),替换掉原YOLOv11的三检测头结构,使得模型检测小目标的能力更加优秀。
2) 用频域感知特征融合(frequency-aware feature fusion, FreqFusion)方法替换掉采样方法并与BiFPN结构结合,使模型可以实现多层级特征融合的同时还可以解决密集图像预测任务中的类别不一致和边界位移问题。
3) 在C2PSA中采用上下文锚点注意力CAA (context anchor attention),构建成C2PSA-CAA模块,以并行方式捕捉远距离的上下文信息。改进后的网络模型结构如图1所示。
在无人机航拍图像目标检测任务中,小目标具有特征不明显、尺度小等特点。YOLOv11模型采用的三检测头(1/8、1/16、1/32)虽能满足大部分目标的尺度检测,但在处理像素点小于32 pixel×32 pixel的目标时检测精度较低,因此如何提高小目标的检测精度是检测模型的改进重点。小目标检测精度低的原因在于以下两点:
1)没有针对超小目标的特征分辨率。1/8尺度特征图的分辨率为80 pixel×80 pixel的网格,难以精确地捕捉到超小目标的形态特征。
2)浅层的(1/8尺度)高分辨率特征无法有效地传递信息至检测头,且各尺度的特征信息交互很有限。原始YOLOv11的FPN-PAN结构无法充分交互浅层与深层之间的特征信息。
为此,将YOLOv11模型的Head部分检测头替换为自适应空间特征融合检测头ASFF[18]并新增了一个1/4尺度的检测头,该检测头生成的160 pixel×160 pixel的高分辨率特征图能够有效的提升模型对超小目标的检测能力,形成新的检测头称为FASFF检测头结构,如图2所示。
FASFF结构首先从骨干网络中提取底层特征C2 (高分辨率,小目标细节丰富)和高层特征C3、C4、C5 (大感受野,语义信息强)。然后通过加权双向金字塔特征网络结构与频域感知特征融合方法提高特征融合能力。通过不同层级的特征融合生成P2、P3、P4、P5,各层计算如下:
$ \left\{\begin{array}{l}P_5^{\text {out }}=\operatorname{Conv}\left(P_5^{\text {in }}\right) \\P_4^{\text {out }}=\operatorname{Conv}\left(P_4^{\text {in }}+FreqFusion\left(P_4^{\text {out }}\right)\right) \\P_3^{\text {out }}=\operatorname{Conv}\left(P_3^{\text {in }}+FreqFusion\left(P_3^{\text {out }}\right)\right) \\P_2^{\text {out }}=\operatorname{Conv}\left(P_2^{\text {in }}+FreqFusion\left(P_2^{\text {out }}\right)\right)\end{array}\right. \;.$
然后,FASFF结构为每个层级P2-P5分配空间自适应权重,再将各层特征与其对应的空间权重逐元素相乘后求和,生成融合特征。具体计算如下:
$ {\boldsymbol{y}}_{ij}^l = \alpha _{ij}^l \cdot {\boldsymbol{x}}_{ij}^{1 \to l} + \beta _{ij}^l \cdot {\boldsymbol{x}}_{ij}^{2 \to l} + \gamma _{ij}^l \cdot {\boldsymbol{x}}_{ij}^{3 \to l} + \delta _{ij}^l \cdot {\boldsymbol{x}}_{ij}^{4 \to l} \;,$
式中:$ \boldsymbol{y}_{ij}^l $表示生成的融合特征;$ \boldsymbol{x}_{ij}^{n\to l} $表示从n层逐层映射到l层$ (i,j) $位置上的特征向量;$\alpha _{ij}^l$、$\beta _{ij}^l$、$\gamma _{ij}^l$、$\delta _{ij}^l$为各层级对应的特征参数分别映射到l层的特征权重,并且满足以下关系:
$ \alpha _{ij}^l = \dfrac{{{{\mathrm{e}}^{\lambda _{{\alpha _{ij}}}^l}}}}{{{{\mathrm{e}}^{\lambda _{{\alpha _{ij}}}^l}} + {{\mathrm{e}}^{\lambda _{{\beta _{ij}}}^l}} + {{\mathrm{e}}^{\lambda _{{\gamma _{ij}}}^l}}}} \;,$
$ \alpha _{ij}^l + \beta _{ij}^l + \gamma _{ij}^l + \delta _{ij}^l = 1\;. $
标准特征融合方法是当前计算机视觉领域应用最广泛的方式,其将低级的高分辨率特征和高级的粗略特征融合,以获得细节和语义信息。标准的特征融合方法可以表示为
$ {{Y} ^l} = {\mathcal{F}^{{\text{UP}}}}({{Y} ^{l + 1}}) + {{X} ^l} \;,$
式中:${{X} ^l} \in {\mathbb{R}^{^{C \times 2H \times 2W}}}$;$ {{Y} ^{l + 1}} \in {\mathbb{R}^{C \times H \times W}} $分别代表骨架生成的第l层特征和上采样后的第l+1层特征;${\mathcal{F}^{{\text{UP}}}}$表示上采样。但是这种方式存在两个显著影响密集预测的缺陷:类别内不一致性和边界位移。因此,通过引入频域感知特征融合FreqFusion[19]方法来解决传统特征融合方法存在的问题。主要组件结构如图3所示,包括:
1)自适应低通滤波器生成器(ALPF),如图3(a)所示。预测空间变化的低通滤波器,在上采样过程中减少对象内部的高频成分,降低类别内不一致性。
2)偏移生成器(Offset),如图3(b)所示。通过重采样,用更一致的特征替换大的不一致特征,使得同一类目标特征更稳定一致。
3)自适应高通滤波器生成器(AHPF),如图3(c)所示,增强在下采样过程中丢失的高频细节边界信息。
这些组件共同组成了FreqFusion结构,如图4所示。其中,偏移量生成器$ Z^l $和$ S^l $作为输入并预测偏移量。它们由两个3×3卷积层组成,用于预测偏移量方向和偏移量尺度,$ D^l $表示偏移的方向,$ A^l $用于控制偏移的大小,$ O^l $是每个高级特征像素的最终预测偏移量,G表示偏移组的数量。r表示用于减少三个生成器后续计算成本的通道缩减率,$ \overline K $和$ \widehat K $分别表示低通滤波器和高通滤波器,Resamp代表重采样过程,$ \oplus $和$ \circledast $分别表示像素级相加和卷积操作。FreqFusion通过两个主要阶段运行,即初始融合和最终融合。初始融合首先通过像素反洗牌操作将特征的空间维度缩小一半,并利用ALPF生成器将压缩的低级特征作为输入,并生成一个初始的低通滤波器来对压缩的高级特征进行上采样,使得低电平中存在的高分辨率结构可以有利于上采样粗略的高电平特征。然后通过与像素反洗牌操作相反的像素洗牌操作后与低级特征融合,一部分输入ALPF生成器,另一部分输入AHPF生成器,AHPF生成器生成的精细调谐的高频细节丰富了特征表示,促进了更有效的下游处理。增强后的初始融合使后续生成器能够更好地适应特征内容,从而产生更精细的最终融合结果。
C2PSA (Cross stage partial with pyramid squeeze attention)是YOLOv11新增的核心模块,是CSP结构与PSA注意力机制的结合体。通过CSP分段处理把特征图分为两部分,一部分直接传递,另一部分通过PSA注意力模块融合,最终拼接融合形成C2PSA模块。PSA注意力机制通过动态调整特征来调整不同位置的关注度,增强模型对目标细节的感知,提升复杂场景下的检测精度。但是上下文特征的信息和关键区域的特征融合不足。为了充分把握检测过程中不同尺度目标的上下文相互依赖性,提高检测精度。本文引入了上下文锚点注意力CAA[20]模块,结合C2PSA模块形成全新的C2PSA-CAA模块。CAA注意力模块的核心在于充分组合和利用其独特的全局平均池化和一维带状卷积,这种策略能够使网络通过精准捕捉图像中近距离和远距离像素的潜在关系,增强中心区域特征学习能力,其结构如图5所示。首先对输入特征进行全局平均池化,提取全局上下文信息。然后通过水平和垂直方向的1D卷积,近似模拟大卷积核的效果,捕获长距离的像素关系。最后通过Sigmoid函数生成注意力权重,用于增强模型输出特征,从而提升模型对长距离上下文信息的感知能力。
实验使用VisDrone2019数据集[21],该数据集是由天津大学制作的开源无人机目标检测数据集,取自中国14个不同城市。该数据集具有环境、光照和天气复杂且多样的特点。待检测的稀疏和密集目标样本数足够多且小目标占比大。默认的训练集、验证集、测试集分别为6471张图片、548张图片和1610张图片。数据集共包含10类检测对象,分别为行人、人、自行车、汽车、面包车、卡车、三轮车、遮阳篷三轮车、公交车和摩托车,每个类别都有若干个样本,各类别的目标检测数量总和如表1所示。
实验平台的操作系统为Windows 10专业版,Intel (R) Core (TM) i9-10900K CPU@3.70 GHz,64 G,GeForce RTX 3090 24 G,编码平台为PyCharm,Python 3.9,深度学习的主要框架是Pytorch,CUDA 12.0。
网络训练共进行200轮(epoch),初始学习率0.01,目标学习率0.1。训练策略上,前3个epoch采用Warmup预热,后续则应用余弦退火算法以实现学习率的平稳衰减。
为建立标准化性能基准并实现跨模型可比性研究,本研究采用多维量化指标体系,包含以下核心评价要素:准确率(precision, P)、召回率(recall, R)、平均精度mAP@0.5、帧率(FPS)、参数量(Params)。其中P定义模型预测结果的可信度;R表征模型覆盖真实目标的能力;mAP@0.5是基于交并比阈值(IoU=0.5)的综合评估,全面反映模型定位与分类性能的协同表现;FPS是单位时间内可处理的图像帧数;Params是模型可训练参数总量,决定边缘部署可行性。计算公式如下:
$ {P = \dfrac{{TP}}{{TP + FP}}} \;, $
$ {R = \dfrac{{TP}}{{TP + FN}}} \;, $
$ A P=\displaystyle\int_0^1 P (R) {\mathrm{d}} R\;, $
$ m A P=\dfrac{1}{N} \displaystyle\sum_{i=0}^n A P_i \;,$
式中:TP (true positive)真阳性表示模型正确地检测出了真实目标;FP (false positive)假阳性表示模型错误地将背景或非目标区域识别为目标;FN (false negative)假阴性表示真实存在的目标没有被模型检测出来。
为了从数据上评估模型的性能,选取了多种主流的模型进行比较。表2展示了不同算法在相同的实验条件下在VisDrone2019数据集上的对比实验数据。由实验数据可知,FFC-YOLO相比YOLOv11n基准模型mAP@0.5提升了7.8%。相比于二阶段算法Fast-RCNN,mAP@0.5提升了28.4%,同时模型的参数量明显减少。与其他一阶段算法如RetinaNet相比,精度更优越。与YOLO算法,如YOLOv3-tiny、YOLOv5n、YOLOv8n、YOLOv10n、YOLOv11n、YOLOv11s相比,mAP@0.5分别提升了16.9%、8.8%、8.0%、7.6%、7.8%、0.7%。FFC-YOLO在参数量低于YOLOv11s模型15.78 M的情况下,精度却提升了0.7%。为了更好地比较改进模型和原始模型对每种类别的识别能力,统计mAP@0.5值,可视化结果如图6所示。图6(a)表示原始模型YOLOV11n的实验结果,对于遮挡严重的密集小目标,由于其尺寸小,容易被环境和其它目标遮挡、重叠,所以检测效果并不理想。由图6(b)可知,FFC-YOLO在所有类别的检测精度都有显著提升。更直观的展示效果如表3所示。从表中可以看出,FFC-YOLO相较于YOLOv11n,对于行人、人、自行车、汽车、面包车、卡车、三轮车、遮阳篷三轮车、公交车和摩托车这10类检测目标,mAP@0.5值分别提升了12.3%、11.0%、6.1%、7.3%、7.0%、4.8%、4.8%、2.8%、8.9%和13.0%。这充分表明,在小目标类别检测任务中,FFC-YOLO具有显著的优势。
为了验证在C2PSA模块中加入CAA注意力机制的必要性以及优越性,选取了一些主流注意力机制如CBAM和ECA,以及模型优化模块如iRMB和EMA,进行对比实验,实验结果如表4所示。其中Base为YOLOv11n中加入FASFF检测头和FreqFusion机制后的基线网络,在基线网络中每个检测头前加入并行且不融入C2PSA模块的CAA注意力机制与融入C2PSA模块后的C2PSA-CAA注意力机制作对比。可以看到,CAA注意力机制与C2PSA融合后的C2PSA-CAA模块检测精度达到了最高(40.0%),比单独加入CAA注意力机制提升了0.3%,且参数量较单独加入CAA注意力机制下降了0.2 M。在与主流注意力机制的对比中,C2PSA-CAA检测精度比CBAM和ECA分别提升了0.4%和0.2%,参数量和C2PSA-ECA相等,比C2PSA-CBAM减少了0.02 M,且R值达到了最高(38.8%)。而且,在与模型优化模块iRMB和EMA的对比中,C2PSA-CAA检测精度比iRMB和EMA分别提升了0.2%和0.3%,虽然参数量比C2PSA-EMA增加了0.02 M,但P值提高了0.2%。综上可知,C2PSA-CAA模块更加适用于航拍图像小目标检测任务。
为了验证FFC-YOLO网络模型对像素在32 pixel×32 pixel及以下小目标的检测能力,另外选择了数据集tiny-data进行模型性能测试。tiny-data数据集为无人机航拍图像下的自建自标注小目标图像数据集,数据集内标注的目标像素尺寸均小于32 pixel×32 pixel。数据集内图片通过翻转、缩放、色域调整等方式进行数据增强后,按照8:1:1分为训练集、测试集和验证集,包括训练集图片1152张,测试集图片145张,验证集图片145张。目标类别为sperson、lperson、wperson,分别代表站立状态的人,坐卧状态的人和水中状态的人;目标数量分别为35240个、26930个、18110个。YOLO v11n和FFC-YOLO在此数据集上的检测结果如表5所示。结果表明,FFC-YOLO在识别所有类别的目标都比YOLOv11n出色。FFC-YOLO在识别sperson类时表现出色,精度较高,mAP@0.5和Precision达到了最高的75.4%、73.3%。FFC-YOLO在识别wperson类精度较低,但mAP@0.5值也达到了57.8%。从实验数据来看,在面对小目标数据集tiny-data,FFC-YOLO的小目标检测效果令人满意。
为了验证FFC-YOLO网络模型中各改进模块的实际效果,对FFC-YOLO网络模型进行消融实验。实验结果如表6所示,其中:Baseline为未作任何修改的YOLOv11n的数据;模型A为替换FASFF检测头之后的数据,加入FASFF检测头后的mAP@0.5提升5.0%,P值也提升了5.8%;模型B为单独加入频域感知特征融合机制FreqFusion,可以看出加入FreqFusion机制后,参数量不仅下降了0.14 M,而且检测精度提升了0.5%,P值达到43.1%;模型C为在A的基础上加入频域感知特征融合机制FreqFusion,可以看出加入FreqFusion机制后,参数量不仅较A下降了0.01 M,而且检测精度提升了2.2%,R值达到39%;模型D为在B的基础上加入C2PSA-CAA模块,可以看出参数量较B下降了0.16 M,但检测精度下降了0.1%;模型E为在A的基础上加入CAA注意力机制,可以看出参数量基本不变,但检测精度有0.2%的下降,P值提升了0.2%,R值下降了0.3%;模型F为在C的基础上加入了C2PSA-CAA模块,降低了0.02 M参数量的同时检测精度提升了0.6%。从实验结果来看,FFC-YOLO满足了检测精度、准确率和召回率较高的特点,可以更加有效的适用于无人机航拍图像小目标检测领域。
在数据集VisDrone2019中,FFC-YOLO的小目标检测能力在图7所示的实验结果中得到了凸显。相比YOLOv11n (如图7(a-c)所示),FFC-YOLO (如图7(d-f)所示)显著提高了对远处小目标 (如行人、自行车、摩托车)的检测准确率。YOLOv11n在图7(a, b)中均存在较多小尺度车辆和行人的漏检问题,行人、摩托车和远处汽车漏检数多达80个,整体对小尺度目标敏感度不足。然而,FFC-YOLO不仅有效减少了各类尺度目标的漏检,而且检测置信度更高。综上,FFC-YOLO在处理像素小于32×32的小目标上比YOLO v11n更精准,更适用于无人机航拍图像的小目标检测。
为了验证FFC-YOLO检测小目标的优越性,将YOLOv11n和改进后的模型用于检测自建数据集tiny-data。YOLOv11n检测结果如图8(a, b)所示,FFC-YOLO检测结果如图8(c, d)所示。对于3类小目标sperson、lperson和wperson,FFC-YOLO检测出的目标数量均高于YOLOv11n。对比图8(a)和图8(c),FFC-YOLO检测出3类目标的个数比YOLOv11n分别高出了25个,1个和7个。对比图8(b)和图8(d),FFC-YOLO检测出3类目标的个数比YOLOv11n分别多出了10个,3个和14个。结果表明,对于远处或密集条件下的小目标,FFC-YOLO的检测能力更加出色。
为应对无人机航拍图像小目标检测中特征模糊、识别难度高的挑战,本文提出了新的小目标检测算法FFC-YOLO。该算法从三个层面实现了创新:1) 增强特征提取:将头部检测头替换为FASFF结构,利用其四个分支强化对小目标的关注及多尺度检测能力;2) 优化特征融合:设计频域感知特征融合策略改进传统采样过程,并将此融合模块与BiFPN结构耦合,有效提升了融合效率,解决了密集小目标场景下的特征丢失、类别误判和边界偏移问题;3) 引入上下文感知:在C2PSA模块中加入上下文锚点注意力机制,使特征提取过程能更好地利用目标周围的语义信息。在VisDrone2019数据集中,基于FFC-YOLO较传统算法如Fast-RCNN、RetinaNet分别提升28.4%、18.6%,较YOLO算法如YOLOv8n、YOLOv10n、YOLOv11n分别提升8.0%、7.6%、7.8%。对于行人、人、自行车、汽车、面包车、卡车、三轮车、遮阳篷三轮车、公交车和摩托车这10类检测目标,FFC-YOLO相较于YOLOv11n,mAP@0.5值分别提升了12.3%、11.0%、6.1%、7.3%、7.0%、4.8%、4.8%、2.8%、8.9%和13.0%。在自建小目标数据集tiny-data中,基于FFC-YOLO的3类目标sperson、lperson和wperson的检测结果较YOLOv11n,mAP@0.5、P、R分别提升了9.2%、8.7%、5.9%。在小目标类别检测任务中,FFC-YOLO检测优势较为突出。后续工作旨在利用模型剪枝和蒸馏技术精简模型,实现高效检测算法的实际部署。

参考文献 引证文献
排序方式:
1
李心仪, 王振飞, 吴晗. 面向仿真场景的无人机目标跟踪算法研究[J]. 系统仿真学报, 2025.
Li X Y, Wang Z F, Wu H. Research on UAV target tracking algorithm for simulation scenarios[J]. J Syst Simul, 2025.
2
曹凯, 李宁, 王立志, 等. 基于单眼视觉的巡检无人机避障策略研究[J]. 电子器件, 2025, 48(1): 86−90.
Cao K, Li N, Wang L Z, et al. Research on obstacle avoidance strategy of patrol UAV based on monocular vision[J]. Chin J Electron Science Devices, 2025, 48(1): 86−90.
3
庄瑜, 傅晓锦, 李莎, 等. 基于XMB-YOLOv5s的无人机小目标检测[J]. 计算机与现代化, 2025, (4): 29−35,41.
Zhuang Y, Fu X J, Li S, et al. UAV small target detection based on XMB-YOLOv5s[J]. Comput Mod, 2025, (4): 29−35,41.
4
Ghen M Q, Yu L J, Zhi C, et al. Improved faster R-CNN for fabric defect detection based on Gabor filter with genetic algorithm optimization[J]. Comput Ind, 2022, 134: 103551.
5
Terven J, Córdova-Esparza D M, Romero-González J A. A comprehensive review of YOLO architectures incomputer vision: from YOLOv1 to YOLOv8 and YOLO-NAS[J]. Mach Learn Knowl Extr, 2023, 5(4): 1680−1716.
6
Wang A, Chen H, Liu L H, et al. YOLOv10: real-time end-to-end object detection[C]//Proceedings of the 38th International Conference on Neural Information Processing Systems, 2025: 3429.
7
Khanam R, Hussain M. YOLOv11: an overview of the key architectural enhancements[Z]. arXiv: 2410.17725, 2024. https://arxiv.org/abs/2410.17725.
8
张杨, 程正东, 朱斌. 基于改进Faster R-CNN的无人机小目标检测算法[J]. 量子电子学报, 2022, 39(3): 354−363.
Zhang Y, Cheng Z D, Zhu B. Small target detection algorithm of drones based on improved faster R-CNN[J]. Chin J Quantum Electron, 2022, 39(3): 354−363.
9
李钟华, 林初俊, 朱恒亮, 等. 基于结构感知和全局上下文信息的小目标检测[J]. 计算机工程与应用, 2024, 60(9): 292−298.
Li Z H, Lin C J, Zhu H L, et al. Small object detection based on structure perception and global context information[J]. Comput Eng Appl, 2024, 60(9): 292−298.
10
李利霞, 王鑫, 王军, 等. 基于特征融合与注意力机制的无人机图像小目标检测算法[J]. 图学学报, 2023, 44(4): 658−666.
Li L X, Wang X, Wang J, et al. Small object detection algorithm in UAV image based on feature fusion and attention mechanism[J]. J Graphics, 2023, 44(4): 658−666.
11
韩俊, 袁小平, 王准, 等. 基于YOLOv5s的无人机密集小目标检测算法[J]. 浙江大学学报(工学版), 2023, 57(6): 1224−1233.
Han J, Yuan X P, Wang Z, et al. UAV dense small target detection algorithm based on YOLOv5s[J]. J Zhejiang Univ (Eng Sci), 2023, 57(6): 1224−1233.
12
牛为华, 魏雅丽. 基于改进YOLOv7的航拍小目标检测算法[J]. 电光与控制, 2024, 31(1): 117−122.
Niu W H, Wei Y L. Small target detection in aerial photography images based on improved YOLOv7 algorithm[J]. Electron Opt Control, 2024, 31(1): 117−122.
13
常凯旋, 黄建华, 孙希延, 等. 基于双模态图像融合的无人机光学小目标检测算法[J]. 激光与光电子学进展, 2025, 62(4): 0428001.
Chang K X, Huang J H, Sun X Y, et al. Optical small target detection method by drone based on dual-modal image fusion[J]. Laser Optoelectron Prog, 2025, 62(4): 0428001.
14
郑华伟, 王飞, 高建邦. DES-YOLO: 一种更精确的目标检测方法[J]. 光电工程, 2024, 51(11): 240212.
Zheng H W, Wang F, Gao J B. DES-YOLO: a more accurate object detection method[J]. Opto-Electron Eng, 2024, 51(11): 240212.
15
张润梅, 肖钰霏, 贾振楠, 等. 改进YOLOv7的无人机视角下复杂环境目标检测算法[J]. 光电工程, 2024, 51(5): 240051.
Zhang R M, Xiao Y F, Jia Z N, et al. Improved YOLOv7 algorithm for target detection in complex environments from UAV perspective[J]. Opto-Electron Eng, 2024, 51(5): 240051.
16
唐彪, 贾小军, 骆淑云, 等. BL-YOLO: 无人机航拍图像目标检测算法[J]. 光电子·激光, 2025. http://kns.cnki.net/kcms/detail/12.1182.o4.20250423.1727.002.html.
Tang B, Jia X J, Luo S Y, et al. BL-YOLO: a target detection algorithm for aerial images captured by drones[J]. J Optoelectron·Laser, 2025. http://kns.cnki.net/kcms/detail/12.1182.o4.20250423.1727.002.html.
17
马滔, 熊英灼, 潘庆娜, 等. 基于VSA-YOLO 11n的无人机红外小目标检测[J]. 红外技术, 2025. http://kns.cnki.net/kcms/detail/53.1053.tn.20250509.1602.002.html.
Ma T, Xiong Y Z, Pan Q N, et al. Infrared small object detection for UAV imagery based on VSA-YOLO11n[J]. Infrared Technol, 2025. http://kns.cnki.net/kcms/detail/53.1053.tn.20250509.1602.002.html.
18
叶荣, 马自飞, 高泉, 等. 基于改进YOLOv5s-ECA-ASFF算法的茶叶病害目标检测[J]. 中国农机化学报, 2024, 45(1): 244−251.
Ye R, Ma Z F, Gao Q, et al. Target detection of tea disease based on improved YOLOv5s-ECA-ASFF algorithm[J]. J Chin Agric Mech, 2024, 45(1): 244−251.
19
Chen L W, Fu Y, Gu L, Yan C, et al. Frequency-aware feature fusion for dense image prediction[J]. IEEE Trans Pattern Anal Mach Intell, 2024, 46(12): 10763−10780.
20
Cai X H, Lai Q X, Wang Y W, et al. Poly kernel inception network for remote sensing detection[C]//Proceedings of the 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024: 27706–27716. https://doi.org/10.1109/CVPR52733.2024.02617.
21
Tan M X, Pang R M, Le Q V. EfficientDet: scalable and efficient object detection[C]//Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020: 10778–10787. https://doi.org/10.1109/CVPR42600.2020.01079.
2026年第53卷第2期
PDF下载
134
55
引用本文
BibTeX
文章信息
doi: 10.12086/oee.2026.250208
  • 接收时间:2025-07-15
  • 首发时间:2026-10-22
  • 出版时间:2026-02-25
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2025-07-15
  • 修回日期:2025-09-24
  • 录用日期:2025-09-29
基金
作者信息
    1嘉兴大学人工智能学院,浙江 嘉兴 314001
    2浙江理工大学信息科学与工程学院,浙江 杭州 310018

通讯作者:

贾小军,
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/oee/CN/10.12086/oee.2026.250208
分享至
全文二维码

扫描看全文

引用本文
BibTeX
本文的引用情况
2种不同金属材料的力学参数
科
Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)
属
Genus
种数
Number of
species
占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏