Article(id=1149781954582770308, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1149781952959574654, articleNumber=null, orderNo=null, doi=10.12404/j.issn.1671-1815.2403972, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=1716825600000, receivedDateStr=2024-05-28, revisedDate=1735142400000, revisedDateStr=2024-12-26, acceptedDate=null, acceptedDateStr=null, onlineDate=1752058979887, onlineDateStr=2025-07-09, pubDate=1743091200000, pubDateStr=2025-03-28, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1752058979887, onlineIssueDateStr=2025-07-09, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1752058979887, creator=13701087609, updateTime=1752058979887, updator=13701087609, issue=Issue{id=1149781952959574654, tenantId=1146029695717560320, journalId=1146123166801305609, year='2025', volume='25', issue='9', pageStart='3529', pageEnd='3967', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=0, createTime=1752058979501, creator=13701087609, updateTime=1776333392421, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1251596220226027613, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1149781952959574654, language=EN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1251596220226027614, tenantId=1146029695717560320, journalId=1146123166801305609, issueId=1149781952959574654, language=CN, specialIssueTitle=, coverIllustrator=, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=3806, endPage=3812, ext={EN=ArticleExt(id=1149781954926703237, articleId=1149781954582770308, tenantId=1146029695717560320, journalId=1146123166801305609, language=EN, title=Low-power Embedded YOLOv8-NANO Intelligent Waste Rapid Sorting System, columnId=1156262729162810294, journalTitle=Science Technology and Engineering, columnName=Papers·Automation and Computational Technology, runingTitle=null, highlight=null, articleAbstract=

In order to solve the problem of automatic sorting of garbage, an intelligent sorting method for household garbage was designed based on the artificial intelligence computing platform Jetson NANO and YOLOv8 algorithm. Firstly, the lightweight YOLOv8 multi-target detection algorithm was used to classify garbage into four categories and achieve accurate recognition in the presence of occlusion. Secondly, the CoreXY coordinate transformation algorithm was used to determine the optimal angle for mechanical claw grasping and achieve precise garbage grasping. Finally, based on TensorRT optimization technology, the lightweight YOLOv8 multi-objective detection algorithm was deployed on the NANO hardware platform, completing the overall construction of the four major categories of garbage intelligent sorting system based on YOLOv8-NANO. The results show that in the self-made garbage test consisting of four major categories and a total of 13 sub categories, the sorting system can effectively achieve detection of single and multiple targets. The detection mAP0.5:0.95 value is 97.3%, and the target detection speed reaches 30.6 frames/s. The fastest garbage sorting speed could reach 6 pieces/min.

, correspAuthors=Bin-rui WANG, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, 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, authorCompany=null, fund=null, authors=null, authorsList=Chun-xiang ZHU, Hong SHU, Bo YANG, Bin-rui WANG), CN=ArticleExt(id=1149781983829651459, articleId=1149781954582770308, tenantId=1146029695717560320, journalId=1146123166801305609, language=CN, title=一种低功耗嵌入式的YOLOv8-NANO智能垃圾快速分拣系统, columnId=1156262729783567290, journalTitle=科学技术与工程, columnName=论文·自动化技术、计算机技术, runingTitle=null, highlight=null, articleAbstract=

为解决垃圾自动分拣难题,设计了以人工智能计算平台Jetson NANO和YOLOv8算法为基础的生活垃圾智能分拣系统。首先,采用轻量化YOLOv8多目标检测算法将垃圾分为四大类并实现在有遮挡情况下的准确识别;其次,基于CoreXY坐标变换算法实现机械爪抓取的最佳角度判断与垃圾精准抓取;最后,基于TensorRT优化技术实现轻量化YOLOv8多目标检测算法在NANO硬件平台的部署,完成基于YOLOv8-NANO的四大类垃圾智能分拣系统整体构建。实验验证结果表明:在自制的包含四大类共13小类垃圾测试中,分拣系统可以有效实现单目标与多目标的检测,检测mAP0.5:0.95(当IoU阈值在0.5~0.95逐步增加时的平均精度)为97.3%,目标检测速度达到30.6 帧/s,垃圾分拣速度最快可达6件/min。

, correspAuthors=王斌锐, authorNote=null, correspAuthorsNote=
* 王斌锐(1978—),男,汉族,江西人,博士,教授。研究方向:检测技术与工业机器人应用。E-mail:
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=NRoqTsqpdPGpC2GEKXdXAA==, magXml=wd3BQKHOM/UHamjc6vmiGw==, pdfUrl=null, pdf=NKTG5wCrO/q8Za2n1kJqqA==, pdfFileSize=7627688, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=G+Tyj2k3hh83sJT5xWU7EA==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=J21MRUr+8vUVpnTi0IBgCg==, mapNumber=null, authorCompany=null, fund=null, authors=

竺春祥(1988—),男,汉族,浙江绍兴人,博士,实验师。研究方向:视觉检测与嵌入式系统。E-mail:

, authorsList=竺春祥, 舒红, 杨波, 王斌锐)}, authors=[Author(id=1251249353504014654, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=16a2900069@cjlu.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1251249353629843782, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, authorId=1251249353504014654, language=EN, stringName=Chun-xiang ZHU, firstName=Chun-xiang, middleName=null, lastName=ZHU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 College of Engineering Training Center, China Jiliang University, Hangzhou 310018, China
2 Zhejiang Province Key Laboratory of On-line Testing Equipment Calibration Technology Research, Hangzhou 310018, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251249353764061515, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, authorId=1251249353504014654, language=CN, stringName=竺春祥, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 中国计量大学工程训练中心、创新学院, 杭州 310018
2 浙江省在线检测装备校准技术研究重点实验室, 杭州 310018, bio={"content":"

竺春祥(1988—),男,汉族,浙江绍兴人,博士,实验师。研究方向:视觉检测与嵌入式系统。E-mail:

"}, bioImg=null, bioContent=

竺春祥(1988—),男,汉族,浙江绍兴人,博士,实验师。研究方向:视觉检测与嵌入式系统。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251249353202024739, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, xref=1, ext=[AuthorCompanyExt(id=1251249353210413348, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, companyId=1251249353202024739, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 College of Engineering Training Center, China Jiliang University, Hangzhou 310018, China), AuthorCompanyExt(id=1251249353214607653, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, companyId=1251249353202024739, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 中国计量大学工程训练中心、创新学院, 杭州 310018)]), AuthorCompany(id=1251249353285910826, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, xref=2, ext=[AuthorCompanyExt(id=1251249353298493739, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, companyId=1251249353285910826, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Zhejiang Province Key Laboratory of On-line Testing Equipment Calibration Technology Research, Hangzhou 310018, China), AuthorCompanyExt(id=1251249353306882349, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, companyId=1251249353285910826, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 浙江省在线检测装备校准技术研究重点实验室, 杭州 310018)])]), Author(id=1251249353894084944, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, 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=1251249354120577369, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, authorId=1251249353894084944, language=EN, stringName=Hong SHU, firstName=Hong, middleName=null, lastName=SHU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 College of Engineering Training Center, China Jiliang University, Hangzhou 310018, China
2 Zhejiang Province Key Laboratory of On-line Testing Equipment Calibration Technology Research, Hangzhou 310018, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251249355701829992, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, authorId=1251249353894084944, language=CN, stringName=舒红, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, address=1 中国计量大学工程训练中心、创新学院, 杭州 310018
2 浙江省在线检测装备校准技术研究重点实验室, 杭州 310018, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251249353202024739, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, xref=1, ext=[AuthorCompanyExt(id=1251249353210413348, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, companyId=1251249353202024739, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 College of Engineering Training Center, China Jiliang University, Hangzhou 310018, China), AuthorCompanyExt(id=1251249353214607653, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, companyId=1251249353202024739, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 中国计量大学工程训练中心、创新学院, 杭州 310018)]), AuthorCompany(id=1251249353285910826, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, xref=2, ext=[AuthorCompanyExt(id=1251249353298493739, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, companyId=1251249353285910826, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Zhejiang Province Key Laboratory of On-line Testing Equipment Calibration Technology Research, Hangzhou 310018, China), AuthorCompanyExt(id=1251249353306882349, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, companyId=1251249353285910826, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 浙江省在线检测装备校准技术研究重点实验室, 杭州 310018)])]), Author(id=1251249355865407856, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, 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=1251249355957682555, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, authorId=1251249355865407856, language=EN, stringName=Bo YANG, firstName=Bo, middleName=null, lastName=YANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, address=3 Yunda Energy Technology Group Co., Ltd., Hangzhou 310012, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251249356045762945, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, authorId=1251249355865407856, language=CN, stringName=杨波, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=3, address=3 运达能源科技集团股份有限公司, 杭州 310012, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251249353399157044, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, xref=3, ext=[AuthorCompanyExt(id=1251249353407545653, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, companyId=1251249353399157044, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 Yunda Energy Technology Group Co., Ltd., Hangzhou 310012, China), AuthorCompanyExt(id=1251249353415934262, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, companyId=1251249353399157044, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 运达能源科技集团股份有限公司, 杭州 310012)])]), Author(id=1251249356154814859, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=wangbrpaper@163.com, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1251249356297421210, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, authorId=1251249356154814859, language=EN, stringName=Bin-rui WANG, firstName=Bin-rui, middleName=null, lastName=WANG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, *, address=1 College of Engineering Training Center, China Jiliang University, Hangzhou 310018, China
2 Zhejiang Province Key Laboratory of On-line Testing Equipment Calibration Technology Research, Hangzhou 310018, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1251249356427444645, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, authorId=1251249356154814859, language=CN, stringName=王斌锐, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, *, address=1 中国计量大学工程训练中心、创新学院, 杭州 310018
2 浙江省在线检测装备校准技术研究重点实验室, 杭州 310018, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1251249353202024739, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, xref=1, ext=[AuthorCompanyExt(id=1251249353210413348, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, companyId=1251249353202024739, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 College of Engineering Training Center, China Jiliang University, Hangzhou 310018, China), AuthorCompanyExt(id=1251249353214607653, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, companyId=1251249353202024739, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 中国计量大学工程训练中心、创新学院, 杭州 310018)]), AuthorCompany(id=1251249353285910826, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, xref=2, ext=[AuthorCompanyExt(id=1251249353298493739, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, companyId=1251249353285910826, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Zhejiang Province Key Laboratory of On-line Testing Equipment Calibration Technology Research, Hangzhou 310018, China), AuthorCompanyExt(id=1251249353306882349, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, companyId=1251249353285910826, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 浙江省在线检测装备校准技术研究重点实验室, 杭州 310018)])])], keywords=[Keyword(id=1251249356695880124, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, language=EN, orderNo=1, keyword=YOLOv8), Keyword(id=1251249356901401043, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, language=EN, orderNo=2, keyword=Jetson NANO), Keyword(id=1251249357115310555, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, language=EN, orderNo=3, keyword=TensorRT deployment), Keyword(id=1251249357278888420, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, language=EN, orderNo=4, keyword=multi-target detection), Keyword(id=1251249357463437814, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, language=EN, orderNo=5, keyword=recognition and capture), Keyword(id=1251249357597655555, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, language=CN, orderNo=1, keyword=YOLOv8), Keyword(id=1251249357689930254, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, language=CN, orderNo=2, keyword=Jetson NANO), Keyword(id=1251249357912228388, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, language=CN, orderNo=3, keyword=TensorRT部署), Keyword(id=1251249358033863218, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, language=CN, orderNo=4, keyword=多目标检测), Keyword(id=1251249358138720831, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, language=CN, orderNo=5, keyword=识别抓取)], refs=[Reference(id=1251249362744067013, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, doi=null, pmid=null, pmcid=null, year=2024, volume=450, issue=null, pageStart=141762, pageEnd=null, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=Dokl M, Van Fan Y, Vujanović A, journalName=Journal of Cleaner Production, refType=null, unstructuredReference=Dokl M, Van Fan Y, Vujanović A, et al. A waste separation system based on sensor technology and deep learning: a simple approach applied to a case study of plastic packaging waste[J]. Journal of Cleaner Production, 2024, 450: 141762., articleTitle=A waste separation system based on sensor technology and deep learning: a simple approach applied to a case study of plastic packaging waste, refAbstract=null), Reference(id=1251249362869896151, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, doi=null, pmid=null, pmcid=null, year=2023, volume=17, issue=6, pageStart=2789, pageEnd=2806, url=null, language=null, rfNumber=[2], rfOrder=1, authorNames=Satav A G, Kubade S, Amrutkar C, journalName=International Journal on Interactive Design and Manufacturing, refType=null, unstructuredReference=Satav A G, Kubade S, Amrutkar C, et al. A state-of-the-art review on robotics in waste sorting: scope and challenges[J]. International Journal on Interactive Design and Manufacturing, 2023, 17(6): 2789-2806., articleTitle=A state-of-the-art review on robotics in waste sorting: scope and challenges, refAbstract=null), Reference(id=1251249362978948069, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, doi=null, pmid=null, pmcid=null, year=2023, volume=21, issue=4, pageStart=1959, pageEnd=1989, url=null, language=null, rfNumber=[3], rfOrder=2, authorNames=Fang B, Yu J, Chen Z, journalName=Environmental Chemistry Letters, refType=null, unstructuredReference=Fang B, Yu J, Chen Z, et al. Artificial intelligence for waste management in smart cities: a review[J]. Environmental Chemistry Letters, 2023, 21(4): 1959-1989., articleTitle=Artificial intelligence for waste management in smart cities: a review, refAbstract=null), Reference(id=1251249363129943029, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, doi=null, pmid=null, pmcid=null, year=2024, volume=5, issue=2, pageStart=507, pageEnd=516, url=null, language=null, rfNumber=[4], rfOrder=3, authorNames=Nwokediegwu Z Q S, Ugwuanyi E D, Dada M A, journalName=Engineering Science & Technology Journal, refType=null, unstructuredReference=Nwokediegwu Z Q S, Ugwuanyi E D, Dada M A, et al. AI-driven waste management systems: a comparative review of innovations in the USA and Africa[J]. Engineering Science & Technology Journal, 2024, 5(2): 507-516., articleTitle=AI-driven waste management systems: a comparative review of innovations in the USA and Africa, refAbstract=null), Reference(id=1251249364673446911, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=8, pageStart=3298, pageEnd=3307, url=null, language=null, rfNumber=[5], rfOrder=4, authorNames=谢康康, 朱文忠, 肖顺兴, journalName=科学技术与工程, refType=null, unstructuredReference=谢康康, 朱文忠, 肖顺兴, . 一种改进YOLOX_S的火焰烟雾检测算法[J]. 科学技术与工程, 2024, 24(8): 3298-3307., articleTitle=一种改进YOLOX_S的火焰烟雾检测算法, refAbstract=null), Reference(id=1251249364845412365, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, doi=null, pmid=null, pmcid=null, year=2024, volume=24, issue=8, pageStart=3298, pageEnd=3307, url=null, language=null, rfNumber=[5], rfOrder=5, authorNames=Xie Kangkang, Zhu Wenzhong, Xiao Shunxing, journalName=Science Technology and Engineering, refType=null, unstructuredReference=Xie Kangkang, Zhu Wenzhong, Xiao Shunxing, et al. Flame smoke detection algorithm on improved YOLOX_S[J]. Science Technology and Engineering, 2024, 24(8): 3298-3307., articleTitle=Flame smoke detection algorithm on improved YOLOX_S, refAbstract=null), Reference(id=1251249365021573143, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, doi=null, pmid=null, pmcid=null, year=2023, volume=41, issue=S2, pageStart=1030, pageEnd=1033, url=null, language=null, rfNumber=[6], rfOrder=6, authorNames=柏润泚, 王凯, 俞范文, journalName=环境工程, refType=null, unstructuredReference=柏润泚, 王凯, 俞范文. 基于ResNet_Vd模型的生活垃圾智能分拣系统[J]. 环境工程, 2023, 41(S2): 1030-1033., articleTitle=基于ResNet_Vd模型的生活垃圾智能分拣系统, refAbstract=null), Reference(id=1251249365164179492, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, doi=null, pmid=null, pmcid=null, year=2023, volume=41, issue=S2, pageStart=1030, pageEnd=1033, url=null, language=null, rfNumber=[6], rfOrder=7, authorNames=Bai Runci, Wang Kai, Yu Fanwen, journalName=Environmental Engineering, refType=null, unstructuredReference=Bai Runci, Wang Kai, Yu Fanwen. ResNet_Vd based intelligent garbage sorting system[J]. Environmental Engineering, 2023, 41(S2): 1030-1033., articleTitle=ResNet_Vd based intelligent garbage sorting system, refAbstract=null), Reference(id=1251249365340340279, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=null, pageStart=1, pageEnd=5, url=null, language=null, rfNumber=[7], rfOrder=8, authorNames=Varghese M P, Anooja V S, Akhila R, journalName=null, refType=null, unstructuredReference=Varghese M P, Anooja V S, Akhila R, et al. IoT-based smart waste management system with level indicators for effective garbage waste segregation[C]//2024 Third International Conference on Distributed Computing and Electrical Circuits and Electronics (ICDCECE). Ballari: IEEE, 2024: 1-5., articleTitle=IoT-based smart waste management system with level indicators for effective garbage waste segregation, refAbstract=null), Reference(id=1251249365474558015, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=null, pageStart=1, pageEnd=7, url=null, language=null, rfNumber=[8], rfOrder=9, authorNames=Mangayarkarasi T, Sathyaraj C, Pradyumna G V, journalName=null, refType=null, unstructuredReference=Mangayarkarasi T, Sathyaraj C, Pradyumna G V, et al. Harnessing IOT for automated sorting system[C]// 2024 International Conference on Communication, Computing and Internet of Things (IC3IoT). Chennai: IEEE, 2024: 1-7., articleTitle=Harnessing IOT for automated sorting system, refAbstract=null), Reference(id=1251249365612970063, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, doi=null, pmid=null, pmcid=null, year=2021, volume=null, issue=null, pageStart=65, pageEnd=77, url=null, language=null, rfNumber=[9], rfOrder=10, authorNames=Kahra M, Sridhar V, Breuß M, journalName=International Conference on Scale Space and Variational Methods in Computer Vision, refType=null, unstructuredReference=Kahra M, Sridhar V, Breuß M. Fast morphological dilation and erosion for grey scale images using the Fourier transform[C]// International Conference on Scale Space and Variational Methods in Computer Vision. Cham: Springer International Publishing, 2021: 65-77., articleTitle=Fast morphological dilation and erosion for grey scale images using the Fourier transform, refAbstract=null), Reference(id=1251249365810102364, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, doi=null, pmid=null, pmcid=null, year=2024, volume=488, issue=null, pageStart=03025, pageEnd=null, url=null, language=null, rfNumber=[10], rfOrder=11, authorNames=Mortos Y C Y, Malbog M A, Nipas M, journalName=E3S Web of Conferences, refType=null, unstructuredReference=Mortos Y C Y, Malbog M A, Nipas M, et al. SMARTSORT: YOLOv4-driven smart sorting for household waste management[C]// E3S Web of Conferences. Paris: EDP Sciences, 2024, 488: 03025., articleTitle=SMARTSORT: YOLOv4-driven smart sorting for household waste management, refAbstract=null), Reference(id=1251249365952708710, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, doi=null, pmid=null, pmcid=null, year=2023, volume=40, issue=9, pageStart=74, pageEnd=80, url=null, language=null, rfNumber=[11], rfOrder=12, authorNames=朱辰, 张西晨, 李森杰, journalName=实验技术与管理, refType=null, unstructuredReference=朱辰, 张西晨, 李森杰, . 基于云边协同的智能垃圾分类系统设计[J]. 实验技术与管理, 2023, 40(9): 74-80, 186., articleTitle=基于云边协同的智能垃圾分类系统设计, refAbstract=null), Reference(id=1251249366078537841, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, doi=null, pmid=null, pmcid=null, year=2023, volume=40, issue=9, pageStart=74, pageEnd=80, url=null, language=null, rfNumber=[11], rfOrder=13, authorNames=Zhu Chen, Zhang Xichen, Li Shenjie, journalName=Experimental Technology and Management, refType=null, unstructuredReference=Zhu Chen, Zhang Xichen, Li Shenjie, et al. Design of intelligent waste sorting system based on cloud edge collaboration[J]. Experimental Technology and Management, 2023, 40(9): 74-80., articleTitle=Design of intelligent waste sorting system based on cloud edge collaboration, refAbstract=null), Reference(id=1251249366179201148, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, doi=null, pmid=null, pmcid=null, year=2024, volume=41, issue=S1, pageStart=43, pageEnd=50, url=null, language=null, rfNumber=[12], rfOrder=14, authorNames=郑宜健, 孟小源, 申文元, journalName=机械设计, refType=null, unstructuredReference=郑宜健, 孟小源, 申文元. 基于机器视觉的景区智能垃圾桶设计[J]. 机械设计, 2024, 41(S1): 43-50., articleTitle=基于机器视觉的景区智能垃圾桶设计, refAbstract=null), Reference(id=1251249366346973318, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, doi=null, pmid=null, pmcid=null, year=2024, volume=41, issue=S1, pageStart=43, pageEnd=50, url=null, language=null, rfNumber=[12], rfOrder=15, authorNames=Zheng Yijian, Meng Xiaoyuan, Shen Wenyuan, journalName=Journal of Machine Design, refType=null, unstructuredReference=Zheng Yijian, Meng Xiaoyuan, Shen Wenyuan. Design of intelligent trash bin in scenic spot based on machine vision[J]. Journal of Machine Design, 2024, 41(S1): 43-50., articleTitle=Design of intelligent trash bin in scenic spot based on machine vision, refAbstract=null), Reference(id=1251249366460219536, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, doi=null, pmid=null, pmcid=null, year=2022, volume=41, issue=7, pageStart=98, pageEnd=103, url=null, language=null, rfNumber=[13], rfOrder=16, authorNames=张月文, 李松恒, 张炜, journalName=实验室研究与探索, refType=null, unstructuredReference=张月文, 李松恒, 张炜, . 基于机器视觉的可回收垃圾智能分拣系统设计[J]. 实验室研究与探索, 2022, 41(7): 98-103, 107., articleTitle=基于机器视觉的可回收垃圾智能分拣系统设计, refAbstract=null), Reference(id=1251249366581854364, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, doi=null, pmid=null, pmcid=null, year=2022, volume=41, issue=7, pageStart=98, pageEnd=103, url=null, language=null, rfNumber=[13], rfOrder=17, authorNames=Zhang Yuewen, Li Songheng, Zhang Wei, journalName=Reserch and Exploration in Laboratory, refType=null, unstructuredReference=Zhang Yuewen, Li Songheng, Zhang Wei, et al. Design of intelligent sorting system for recyclable waste based on machine vision[J]. Reserch and Exploration in Laboratory, 2022, 41(7): 98-103, 107., articleTitle=Design of intelligent sorting system for recyclable waste based on machine vision, refAbstract=null), Reference(id=1251249366682517678, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=8, pageStart=68, pageEnd=73, url=null, language=null, rfNumber=[14], rfOrder=18, authorNames=欧阳飞, 吴旭, 向东升, journalName=计算机与现代化, refType=null, unstructuredReference=欧阳飞, 吴旭, 向东升. 基于改进YOLOv8的垃圾分类检测方法[J]. 计算机与现代化, 2023(8): 68-73., articleTitle=基于改进YOLOv8的垃圾分类检测方法, refAbstract=null), Reference(id=1251249366791569589, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=8, pageStart=68, pageEnd=73, url=null, language=null, rfNumber=[14], rfOrder=19, authorNames=Ouyang Fei, Wu xu, Xiang Dongsheng, journalName=Computer and Modernization, refType=null, unstructuredReference=Ouyang Fei, Wu xu, Xiang Dongsheng. Garbage classification and detection method based on improved YOLOv8[J]. Computer and Modernization, 2023(8): 68-73., articleTitle=Garbage classification and detection method based on improved YOLOv8, refAbstract=null), Reference(id=1251249366959341762, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, doi=null, pmid=null, pmcid=null, year=2021, volume=164, issue=null, pageStart=105132, pageEnd=null, url=null, language=null, rfNumber=[15], rfOrder=20, authorNames=Mao W L, Chen W C, Wang C T, journalName=Resources, Conservation and Recycling, refType=null, unstructuredReference=Mao W L, Chen W C, Wang C T, et al. Recycling waste classification using optimized convolutional neural network[J]. Resources, Conservation and Recycling, 2021, 164: 105132., articleTitle=Recycling waste classification using optimized convolutional neural network, refAbstract=null), Reference(id=1251249367089365198, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=null, pageStart=1, pageEnd=7, url=null, language=null, rfNumber=[16], rfOrder=21, authorNames=Ande A, Sundar G N, Thomas R, journalName=2024 Second International Conference on Emerging Trends in Information Technology and Engineering, refType=null, unstructuredReference=Ande A, Sundar G N, Thomas R, et al. Smart waste bins using DCNN and Internet of Things[C]// 2024 Second International Conference on Emerging Trends in Information Technology and Engineering. Vellore: IEEE, 2024: 1-7., articleTitle=Smart waste bins using DCNN and Internet of Things, refAbstract=null), Reference(id=1251249367231971549, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, doi=null, pmid=null, pmcid=null, year=2024, volume=83, issue=10, pageStart=29505, pageEnd=29529, url=null, language=null, rfNumber=[17], rfOrder=22, authorNames=Lilhore U K, Simaiya S, Dalal S, journalName=Multimedia Tools and Applications, refType=null, unstructuredReference=Lilhore U K, Simaiya S, Dalal S, et al. A smart waste classification model using hybrid CNN-LSTM with transfer learning for sustainable environment[J]. Multimedia Tools and Applications, 2024, 83(10): 29505-29529., articleTitle=A smart waste classification model using hybrid CNN-LSTM with transfer learning for sustainable environment, refAbstract=null), Reference(id=1251249367366189283, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, doi=null, pmid=null, pmcid=null, year=2024, volume=196, issue=3, pageStart=279, pageEnd=null, url=null, language=null, rfNumber=[18], rfOrder=23, authorNames=Mishra S, Yaduvanshi R, Rajpoot P, journalName=Environmental Monitoring and Assessment, refType=null, unstructuredReference=Mishra S, Yaduvanshi R, Rajpoot P, et al. An integrated deep-learning model for smart waste classification[J]. Environmental Monitoring and Assessment, 2024, 196(3): 279., articleTitle=An integrated deep-learning model for smart waste classification, refAbstract=null), Reference(id=1251249367517184242, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=227, pageEnd=232, url=null, language=null, rfNumber=[19], rfOrder=24, authorNames=Raman R, Rawandale C J, Meenakshi R, journalName=2023 Second International Conference on Smart Technologies for Smart Nation, refType=null, unstructuredReference=Raman R, Rawandale C J, Meenakshi R, et al. Real-time video management system for robotic waste sorting and recycling using IoT and machine learning[C]// 2023 Second International Conference on Smart Technologies for Smart Nation. Singapore: IEEE, 2023: 227-232., articleTitle=Real-time video management system for robotic waste sorting and recycling using IoT and machine learning, refAbstract=null), Reference(id=1251249369127796989, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, doi=null, pmid=null, pmcid=null, year=2024, volume=8, issue=2, pageStart=182, pageEnd=192, url=null, language=null, rfNumber=[20], rfOrder=25, authorNames=Mohankumar A, Gowtham R, Gokul B, journalName=Asian Journal of Applied Science and Technology, refType=null, unstructuredReference=Mohankumar A, Gowtham R, Gokul B, et al. Optimizing urban sustainability: a smart waste management system with Arduino technology[J]. Asian Journal of Applied Science and Technology, 2024, 8(2): 182-192., articleTitle=Optimizing urban sustainability: a smart waste management system with Arduino technology, refAbstract=null)], funds=[Fund(id=1251249362307859359, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, awardId=2023C03186, language=CN, fundingSource=浙江省“尖兵”“领雁”研发攻关计划(2023C03186), fundOrder=null, country=null), Fund(id=1251249362500797360, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, awardId=jg20220284, language=CN, fundingSource=浙江省高等教育“十四五”教学改革项目(jg20220284), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1251249353202024739, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, xref=1, ext=[AuthorCompanyExt(id=1251249353210413348, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, companyId=1251249353202024739, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 College of Engineering Training Center, China Jiliang University, Hangzhou 310018, China), AuthorCompanyExt(id=1251249353214607653, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, companyId=1251249353202024739, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1 中国计量大学工程训练中心、创新学院, 杭州 310018)]), AuthorCompany(id=1251249353285910826, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, xref=2, ext=[AuthorCompanyExt(id=1251249353298493739, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, companyId=1251249353285910826, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 Zhejiang Province Key Laboratory of On-line Testing Equipment Calibration Technology Research, Hangzhou 310018, China), AuthorCompanyExt(id=1251249353306882349, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, companyId=1251249353285910826, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2 浙江省在线检测装备校准技术研究重点实验室, 杭州 310018)]), AuthorCompany(id=1251249353399157044, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, xref=3, ext=[AuthorCompanyExt(id=1251249353407545653, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, companyId=1251249353399157044, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 Yunda Energy Technology Group Co., Ltd., Hangzhou 310012, China), AuthorCompanyExt(id=1251249353415934262, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, companyId=1251249353399157044, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3 运达能源科技集团股份有限公司, 杭州 310012)])], figs=[ArticleFig(id=1251249358486848107, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, language=EN, label=Fig.1, caption=Structure diagram of intelligent garbage rapid sorting system, figureFileSmall=SVMJXI2XPez4EbjOjZ7pAw==, figureFileBig=zmrSy4fmxF78SXpDpxVJ6w==, tableContent=null), ArticleFig(id=1251249358595900027, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, language=CN, label=图1, caption=智能垃圾快速分拣系统结构框图, figureFileSmall=SVMJXI2XPez4EbjOjZ7pAw==, figureFileBig=zmrSy4fmxF78SXpDpxVJ6w==, tableContent=null), ArticleFig(id=1251249360240067216, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, language=EN, label=Fig.2, caption=Overall algorithm block diagram of YOLOv8 and NANO, figureFileSmall=Wq/1CHYSkx0EqO+omaf0YQ==, figureFileBig=6ZDtG+4VtRKpKrBD/Na4XA==, tableContent=null), ArticleFig(id=1251249360323953303, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, language=CN, label=图2, caption=YOLOv8与NANO的总体算法框图, figureFileSmall=Wq/1CHYSkx0EqO+omaf0YQ==, figureFileBig=6ZDtG+4VtRKpKrBD/Na4XA==, tableContent=null), ArticleFig(id=1251249360449782436, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, language=EN, label=Fig.3, caption=Partial garbage dataset, figureFileSmall=e+Ib9xznY1P98G3i5XkmXw==, figureFileBig=YiuMmxE+zSg0TOjoRwAPQA==, tableContent=null), ArticleFig(id=1251249360625943227, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, language=CN, label=图3, caption=部分垃圾数据集, figureFileSmall=e+Ib9xznY1P98G3i5XkmXw==, figureFileBig=YiuMmxE+zSg0TOjoRwAPQA==, tableContent=null), ArticleFig(id=1251249360772743885, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, language=EN, label=Fig.4, caption=Signal filtering, figureFileSmall=ZZACTIUclEaQfofj5OjPGA==, figureFileBig=FLleJ3ru5YX00PSMp1lOEA==, tableContent=null), ArticleFig(id=1251249360890184413, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, language=CN, label=图4, caption=信号滤波效果, figureFileSmall=ZZACTIUclEaQfofj5OjPGA==, figureFileBig=FLleJ3ru5YX00PSMp1lOEA==, tableContent=null), ArticleFig(id=1251249360995042029, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, language=EN, label=Fig.5, caption=Calculation process of optimal clamping angle, figureFileSmall=mGbk7iIfeRm95Yjme6mfdg==, figureFileBig=H+yXeeTolt6AoCt6McAXhw==, tableContent=null), ArticleFig(id=1251249361146036998, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, language=CN, label=图5, caption=最佳夹持角度计算过程, figureFileSmall=mGbk7iIfeRm95Yjme6mfdg==, figureFileBig=H+yXeeTolt6AoCt6McAXhw==, tableContent=null), ArticleFig(id=1251249361250894618, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, language=EN, label=Fig.6, caption=Deployment framework for classification models based on TensorRT, figureFileSmall=UVvCO9q69ddGmfZQCtPNPA==, figureFileBig=0wW/0ewDDLgBaOaZJ8n7RA==, tableContent=null), ArticleFig(id=1251249361355752228, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, language=CN, label=图6, caption=基于TensorRT的分类模型部署框架, figureFileSmall=UVvCO9q69ddGmfZQCtPNPA==, figureFileBig=0wW/0ewDDLgBaOaZJ8n7RA==, tableContent=null), ArticleFig(id=1251249361468998452, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, language=EN, label=Fig.7, caption=Complete modeling diagram of garbage classification device, figureFileSmall=ZXV/rV0oXTqd0VJd2zyb1g==, figureFileBig=l5cPu0MRpaD/8Mvj1NP74g==, tableContent=null), ArticleFig(id=1251249361586438980, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, language=CN, label=图7, caption=垃圾分类装置完整建模图, figureFileSmall=ZXV/rV0oXTqd0VJd2zyb1g==, figureFileBig=l5cPu0MRpaD/8Mvj1NP74g==, tableContent=null), ArticleFig(id=1251249361695490896, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, language=EN, label=Fig.8, caption=Target detection results, figureFileSmall=iR/9gYo8sc7Hkl2nR1bQcA==, figureFileBig=v6r6pzZ+7diMnjfM1C9L6w==, tableContent=null), ArticleFig(id=1251249361842291555, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, language=CN, label=图8, caption=目标检测效果, figureFileSmall=iR/9gYo8sc7Hkl2nR1bQcA==, figureFileBig=v6r6pzZ+7diMnjfM1C9L6w==, tableContent=null), ArticleFig(id=1251249361930371951, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, language=EN, label=Table 1, caption=

Comparison results of various models

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 mAP0.5/% mAP0.5:
0.95/%
TensorRT
推理时间/ms
YOLOv8-NANO 100 97.37 30.85
YOLOV5-NANO 99.49 94.77 40.65
YOLOV3-NANO 99.56 94.28 64.23
), ArticleFig(id=1251249362026840957, tenantId=1146029695717560320, journalId=1146123166801305609, articleId=1149781954582770308, language=CN, label=表1, caption=

各类模型对比结果

, figureFileSmall=null, figureFileBig=null, tableContent=
模型 mAP0.5/% mAP0.5:
0.95/%
TensorRT
推理时间/ms
YOLOv8-NANO 100 97.37 30.85
YOLOV5-NANO 99.49 94.77 40.65
YOLOV3-NANO 99.56 94.28 64.23
)], attaches=null, journal=Journal(id=1146119176004939786, delFlag=0, nameCn=科学技术与工程, nameEn=Science Technology and Engineering, nameHistory1=null, nameHistory2=null, issn=1671-1815, eissn=, cn=11-4688/T, coden=null, periodic=4, language=CN, oaType=是, 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=UKU/O7GSka5polgCTkbIIw==, journalPrice=null, startedYear=null, abbrevIsoEn=Sci Technol Eng, journalRemark=null, publicationField=null, createdTime=null, updatedTime=1754445529766, createdBy=null, updatedBy=13701087609, firstLetterCn=S, firstLetterEn=S, subjectCode=Natural Sciences, subjectName=自然科学, subjectCodeEn=Natural Sciences, subjectNameEn=null, picCn=UKU/O7GSka5polgCTkbIIw==, picEn=5hwlULoNwcbj3xUmVi9MAQ==, jcr=null, cjcr=null, exts=[JournalExt(id=1159791870395564357, language=CN, name=科学技术与工程, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.stae.com.cn/jsygc/home, createdTime=1754445529793, updatedTime=1754445529793, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=http://www.stae.com.cn/jsygc/site/menus/20090429150146001, submissionAuthorUrl=http://www.stae.com.cn/jsygc/author/login, submissionEditorUrl=http://www.stae.com.cn/jsygc/editor/login, submissionReviewUrl=http://www.stae.com.cn/jsygc/reviewer/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1159791870441701702, language=EN, name=Science Technology and Engineering, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=http://www.stae.com.cn/jsygc/home, createdTime=1754445529804, updatedTime=1754445529804, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=http://www.stae.com.cn/jsygc/author/login, submissionEditorUrl=http://www.stae.com.cn/jsygc/editor/login, submissionReviewUrl=http://www.stae.com.cn/jsygc/reviewer/login, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1146123166801305609, websiteList=[Website(id=1148243202391400884, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146123166801305609, 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/kxjsygc/CN, language=CN, createTime=1751692112777, createBy=18614031015, updateTime=1753520965431, updateBy=18614031015, name=科学技术与工程-中文站点, tplId=1146099689490845704, title=科学技术与工程, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1148622798802673703, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=articleTextType, value=kx, createTime=1751782615614, updateTime=1751782615614, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798781702180, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=banner, value=null, createTime=1751782615609, updateTime=1751782615609, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798769119267, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=j86gbwi+p0Idkyl5SzIlmQ==, createTime=1751782615606, updateTime=1751782615606, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798794285094, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1751782615612, updateTime=1751782615612, creator=18614031015, updator=18614031015), WebsiteProps(id=1148622798790090789, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1148243202391400884, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1751782615611, updateTime=1751782615611, creator=18614031015, updator=18614031015)]), Website(id=1155914124811976731, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1146123166801305609, 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/kxjsygc/EN, language=EN, createTime=1753521003206, createBy=18614031015, updateTime=1753521003206, updateBy=18614031015, name=科学技术与工程-英文站点, tplId=1146101810881728533, title=Science Technology and Engineering, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1155914371227308235, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=articleTextType, value=kx, createTime=1753521061952, updateTime=1753521061952, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371210531016, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=banner, value=null, createTime=1753521061947, updateTime=1753521061947, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371202142407, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=j86gbwi+p0Idkyl5SzIlmQ==, createTime=1753521061945, updateTime=1753521061945, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371223113930, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1753521061950, updateTime=1753521061950, creator=18614031015, updator=18614031015), WebsiteProps(id=1155914371218919625, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1155914124811976731, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1753521061949, updateTime=1753521061949, creator=18614031015, updator=18614031015)])], journalTitle=科学技术与工程, weixinUrl=null, journalUrl=null, iacademicId=null, status=0, seqNo=null, journalTitleEn=Science Technology and Engineering, journalPhotoCn=UKU/O7GSka5polgCTkbIIw==, journalPhotoEn=5hwlULoNwcbj3xUmVi9MAQ==, journalFirstLetter=S, 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=null, provinceCode=null, provinceName=null, collectFlag=false), detailUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2403972, detailUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/10.12404/j.issn.1671-1815.2403972, pdfUrlCn=https://castjournals.cast.org.cn/joweb/kxjsygc/CN/PDF/10.12404/j.issn.1671-1815.2403972, pdfUrlEn=https://castjournals.cast.org.cn/joweb/kxjsygc/EN/PDF/10.12404/j.issn.1671-1815.2403972, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
一种低功耗嵌入式的YOLOv8-NANO智能垃圾快速分拣系统
收藏切换
PDF下载
竺春祥 1, 2 , 舒红 1, 2 , 杨波 3 , 王斌锐 1, 2, *
科学技术与工程 | 论文·自动化技术、计算机技术 2025,25(9): 3806-3812
收起
收藏切换
科学技术与工程 | 论文·自动化技术、计算机技术 2025, 25(9): 3806-3812
一种低功耗嵌入式的YOLOv8-NANO智能垃圾快速分拣系统
全屏
竺春祥1, 2 , 舒红1, 2, 杨波3, 王斌锐1, 2, *
作者信息
  • 1 中国计量大学工程训练中心、创新学院, 杭州 310018
  • 2 浙江省在线检测装备校准技术研究重点实验室, 杭州 310018
  • 3 运达能源科技集团股份有限公司, 杭州 310012
  • 竺春祥(1988—),男,汉族,浙江绍兴人,博士,实验师。研究方向:视觉检测与嵌入式系统。E-mail:

通讯作者:

* 王斌锐(1978—),男,汉族,江西人,博士,教授。研究方向:检测技术与工业机器人应用。E-mail:
Low-power Embedded YOLOv8-NANO Intelligent Waste Rapid Sorting System
Chun-xiang ZHU1, 2 , Hong SHU1, 2, Bo YANG3, Bin-rui WANG1, 2, *
Affiliations
  • 1 College of Engineering Training Center, China Jiliang University, Hangzhou 310018, China
  • 2 Zhejiang Province Key Laboratory of On-line Testing Equipment Calibration Technology Research, Hangzhou 310018, China
  • 3 Yunda Energy Technology Group Co., Ltd., Hangzhou 310012, China
出版时间: 2025-03-28 doi: 10.12404/j.issn.1671-1815.2403972
文章导航
收藏切换

为解决垃圾自动分拣难题,设计了以人工智能计算平台Jetson NANO和YOLOv8算法为基础的生活垃圾智能分拣系统。首先,采用轻量化YOLOv8多目标检测算法将垃圾分为四大类并实现在有遮挡情况下的准确识别;其次,基于CoreXY坐标变换算法实现机械爪抓取的最佳角度判断与垃圾精准抓取;最后,基于TensorRT优化技术实现轻量化YOLOv8多目标检测算法在NANO硬件平台的部署,完成基于YOLOv8-NANO的四大类垃圾智能分拣系统整体构建。实验验证结果表明:在自制的包含四大类共13小类垃圾测试中,分拣系统可以有效实现单目标与多目标的检测,检测mAP0.5:0.95(当IoU阈值在0.5~0.95逐步增加时的平均精度)为97.3%,目标检测速度达到30.6 帧/s,垃圾分拣速度最快可达6件/min。

YOLOv8  /  Jetson NANO  /  TensorRT部署  /  多目标检测  /  识别抓取

In order to solve the problem of automatic sorting of garbage, an intelligent sorting method for household garbage was designed based on the artificial intelligence computing platform Jetson NANO and YOLOv8 algorithm. Firstly, the lightweight YOLOv8 multi-target detection algorithm was used to classify garbage into four categories and achieve accurate recognition in the presence of occlusion. Secondly, the CoreXY coordinate transformation algorithm was used to determine the optimal angle for mechanical claw grasping and achieve precise garbage grasping. Finally, based on TensorRT optimization technology, the lightweight YOLOv8 multi-objective detection algorithm was deployed on the NANO hardware platform, completing the overall construction of the four major categories of garbage intelligent sorting system based on YOLOv8-NANO. The results show that in the self-made garbage test consisting of four major categories and a total of 13 sub categories, the sorting system can effectively achieve detection of single and multiple targets. The detection mAP0.5:0.95 value is 97.3%, and the target detection speed reaches 30.6 frames/s. The fastest garbage sorting speed could reach 6 pieces/min.

YOLOv8  /  Jetson NANO  /  TensorRT deployment  /  multi-target detection  /  recognition and capture
竺春祥, 舒红, 杨波, 王斌锐. 一种低功耗嵌入式的YOLOv8-NANO智能垃圾快速分拣系统. 科学技术与工程, 2025 , 25 (9) : 3806 -3812 . DOI: 10.12404/j.issn.1671-1815.2403972
Chun-xiang ZHU, Hong SHU, Bo YANG, Bin-rui WANG. Low-power Embedded YOLOv8-NANO Intelligent Waste Rapid Sorting System[J]. Science Technology and Engineering, 2025 , 25 (9) : 3806 -3812 . DOI: 10.12404/j.issn.1671-1815.2403972
目前,中国城市生活垃圾分类回收主要依靠人工视觉分类。一些垃圾分拣站虽然也开始采用机械臂、传送带等机械设备进行分类,但是其识别率和处理效率并不高,并且系统在外部环境变化时缺乏稳定性,无法适用于复杂多变的现实环境[1-3]。基于机器视觉建立新的垃圾分类回收体系,提高垃圾资源化利用,构建“智能+垃圾分拣”模式具有重要意义[4]。YOLO(you only look once)算法在视频和图片中目标识别和定位领域应用广泛,因其速度快、准确率高而备受关注[5-7]。然而,传统的YOLO算法在处理复杂场景时仍存在一定的局限性,如对小目标的检测不够敏感,以及在垃圾分类等实时性要求较高的场景下处理速度仍有提升空间[8-10]。针对这些问题,现提出一种基于YOLOv8-NANO的改进算法,旨在进一步优化模型的性能,以适应智能生活垃圾分拣系统的需求。通过引入深度学习技术,结合嵌入式系统低功耗的特点,能够快速准确地识别和分类各种生活垃圾,从而提高分拣效率和资源回收率。
国内外学者针对智能+垃圾分拣模式的研究进行了大量的探索和实践。朱辰等[11]将树莓派和Haas100开发板结合,实现了对3种轻量级垃圾分类算法进行训练,并在树莓派上推理实现了分类。郑宜健等[12]针对景区场景下设计了智能垃圾分类装置原型样机,利用深度卷积神经网络建立了相关的可回收和不可回收的二分类结构模型,最后在OpenMV视觉平台上进行了验证。为了能够对垃圾进一步分拣到对应的垃圾箱内,张月文等[13]结合机械臂和深度相机,并训练了基于Yolov4和KCF的目标跟踪算法,实现了对可回收垃圾的精确分拣。随着YOLO系类算法的不断更新和发展,欧阳飞等[14]将YOLOv8算法与注意力机制结合,并在自制垃圾分类数据集训练YOLOv8网络,实现垃圾的检测和分类。Mao等[15]使用遗传算法优化了DenseNet121卷积神经网络的整体连接层,并通过测试优化后的DenseNet121分类精度达到99.60%。国际上,Ande等[16]利用深度卷积神经网络对垃圾进行分类,并集成了超声波传感器、湿度传感器和旋转执行器,实现了对放入智能垃圾箱的废物的有效分类和分离,提高了废物分类的有效性。Lilhore等[17]结合卷积神经网络和长短期记忆的混合模型,并使用迁移学习方法,该方法结合了ImageNet的优点,对废物类别进行分类和预测。Mishra等[18]提出使用针对卓越性能进行优化的高级深度学习模型DenseNet-121,从扩展的TrashNet数据集中提取有意义的特征,这些特征随后被输入支持向量机进行精确分类。然而,由于这些策略的网络结构过于庞大,并不适合部署到嵌入式设备上实施。为了将垃圾分类检测模型部署到嵌入式端,文献[19-20]利用树莓派和Arduino等硬件平台对垃圾分类模型进行部署推理,展示了其在垃圾分类自动化方面的潜力。但只停留在了垃圾类别的识别的层面,不能得出目标垃圾的准确夹取坐标,并且该方法难以获取垃圾的具体三维坐标位置,所以难以实现自动的垃圾精确抓取。
现提出一种基于改进YOLOv8算法与英伟达NANO嵌入式硬件平台结合的生活垃圾智能分拣实验系统,旨在解决传统垃圾分拣效率低下的问题。该系统通过集成YOLOv8算法,能够快速准确地识别和分类不同类型的垃圾。英伟达NANO嵌入式硬件平台的部署,则确保了系统的低功耗和高效率,使其更适合在资源有限的环境中部署,为智能垃圾分拣技术的发展提供新的思路和解决方案。
YOLOv8-NANO智能生活垃圾快速分拣系统由以下几部分组成:机械爪运动模块、目标检测模块、硬件机械平台、人机交互面板、待分拣对象和其他功能模块,其组成框图如图1所示。
机械爪运动模块由机动性强的CoreXY结构、WHEELTEC柔性机械爪、STM32单片机、PCB控制板等组成。在CoreXY结构与机械爪之间加入控制升降、旋转的舵机,用于调整机械爪抓取的最佳角度,根据上位机发送的指令,CoreXY机械手将待分类物体抓取并放置到指定分类区内,从而实现垃圾的抓取和分拣功能。本系统目标检测模块由软硬件两部分。硬件是英伟达NANO平台与4K高清摄像头作为目标检测的硬件模块,固定于装置的顶部框架上,同时在装置的入口处配备了传感器,当有垃圾投入到桶内时,触发传感器发送信号到NANO平台对垃圾进行在线识别。目标检测模块的软件部分为YOLOv8的多目标检测算法,同时借助官方支持的 TensorRT 加速技术在英伟达NANO硬件平台实现YOLOv8的加速推理,可以近乎实时检测垃圾而不需要稳定的互联网连接,更具有普适性。其他功能模块主要由自动打包、满载检测、语言提示和自动压缩等。当存放目标垃圾的桶装满后,将在显示操作板上显示,通过语音指令即可以实现垃圾的自动打包。当目标检测检测到的物品是易拉罐、矿泉水瓶、纸团等较大的物品时,会对其进行压缩以减小体积。人机交互面板设计是对装置进行基础性的操作和各项数据显示,在人机交互面板上可以对分类的数据进行显示,同时进行垃圾分类宣传视频播放。
YOLOv8-NANO智能生活垃圾快速分拣系统运行流程如下:上位机主要负责图像的采集与处理,下位机主要负责机械的操作和垃圾分类投放。当垃圾被投入装置内部时触发系统工作,NANO嵌入式平台运行YOLOv8算法对摄像头采集得到的图像进行识别生成类别标签和坐标数据,通过得到目标类别及像素坐标,由YOLOv8算法获取到的目标矩形框,截取出图像并判断最佳夹持角度,得到机械爪抓取的最佳角度信息,通过串口通信,将识别类别信息和抓取最佳角度发送到STM32单片机控制机械爪进行垃圾分拣。
YOLOv8目标识别算法采用了一种新的网络结构,引入更多的卷积层和特征融合模块来提高检测性能,同时使用了Anchor-Free等技术,使得自身的准确度也超过其他算法。受到此算法的启发,同时也对比了YOLOv3、YOLOv5和YOLOv8得到的训练结果,创新性地使用了YOLOv8和NANO嵌入式平台结合,并采用TensorRT优化技术实现算法的部署,总体算法框架如图2所示,从而使其能够达到精度与速度的平衡。另外,与传统的交并比(intersection over union,IoU)损失函数和通用交并比(generalized IoU,GIoU)损失函数相比,YOLOv8采用的完整交并比(complete-IoU,CIoU)损失函数在计算预测框和真实框的重合程度时,除了考虑它们的交集和并集面积外,还加入了中心点距离的惩罚项;中心点距离项可以更好地优化预测框的中心点,使预测框的中心点更接近真实框的中心点。因此在本实验中将上述损失函数采用CIoU损失函数,以降低回归误差,提升回归效果。CIoU 损失函数及其相关公式如式(1)~式(3)所示。
LCIoU=1-IoU+$\frac{{d}^{2}}{{c}^{2}}$+αμ
α=$\frac{\mu }{1-\mathrm{I}\mathrm{o}\mathrm{U}+\mu }$
μ=$\frac{4}{{\mathrm{\pi }}^{2}}{\left[\mathrm{t}\mathrm{a}{\mathrm{n}}^{-1}\left(\frac{{w}^{\mathrm{g}\mathrm{t}}}{{h}^{\mathrm{g}\mathrm{t}}}\right)-\mathrm{t}\mathrm{a}{\mathrm{n}}^{-1}\left(\frac{w}{h}\right)\right]}^{2}$
式中:d为目标框和预测框中心点的欧氏距离;c为包住目标框和预测框的最小方框的对角线距离;αμ为长宽比;whwgthgt分别为预测框的高宽和真实框的高宽。
为了使设备更具广泛性,收集了生活常见的可回收、有害、厨余以及其他垃圾共四大类,为了确保样本得多样性、随机性,采用视频录制每分钟60帧的方式,得到四大类垃圾不同姿态、遮挡情况下约30 000张。在自制数据集初期,使用可视化图像标定工具Labelimg手动标注所需样品,部分照片如图3所示。结合系统设备的制作材料和算法识别,采用了全黑的物料背景。在实际复杂多样的分拣当中,确保后期实际检测当中数据集建立的两大核心:①样品图像的选择,需要选择代表性且数量足够的样本,覆盖预计需要识别的各个类别:如样本质量的高低、属性的多样性(如光照、角度等)确保后续分类结果的正确性;②数据平衡与整理关系到数据集的质量和效果。只有选择代表的样本,并进行规范整理,才能为后续算法训练和评估提供有效支持,从而提升识别效果。
通过两个步进电机控制机械爪在平面内的运动,电机1旋转,机械臂会在角度为45°的直线上运动,电机2旋转,机械臂会在角度为135°的直线上运动,通过对期望的运动方向和速度构成的矢量进行分解,便可以控制机械臂在平面内任意角度运动。然而随着持续的垃圾分拣任务,由于步进电机的丢步问题,机械爪会逐渐偏移原来的位置,导致定位精度变差,需要实时进行校准。本设计中定位靠两个超声波测距模块进行定位,由于机械臂在运动过程中会发生震动,所以需要对信号进行滤波,滤波器使用2阶巴特沃斯低通滤波器,截止频率设置为4 Hz,如图4所示。数字截止频率为4 Hz,采样频率为100 Hz,阶数为2。滤波器公式如式(4)~式(6)所示。
H(p)=$\frac{1}{\stackrel{N}{\sum _{k}}{a}_{k}{p}^{k}}$
式(4)中:N为滤波器阶数,取值2;ak通过查表获得,a0=1,a1=1.414 2,a2=1。
p=$\frac{s}{4\mathrm{\pi }\mathrm{t}\mathrm{a}\mathrm{n}\left(\frac{\mathrm{\pi }{f}_{\mathrm{d}}}{{f}_{\mathrm{s}}}\right)}$
式(5)中:fd为数字截止频率;fs为采样频率。使用双线性变换可得
s=2fs$\frac{z-1}{z+1}$
s替换为z,得到H(z),最后将H(z)转化为差分方程用C语言实现,采用双精度地位的定位模式,大大提高了抓取目标垃圾的成功率,进一步增加了分拣的效率与速度。
通过YOLOv8算法得到矩形框,截断图像进行二值化。正常的二值化有一个预定的阈值,如果图像的亮度和颜色不同,会导致对整个外边界的识别不准确。因此,部署自适应阈值至关重要。在这种情况下,阈值是基于从图像上的中心点延伸到其相应阈值的小区域来计算的。因此对于不同图像上的不同阈值被用来使角度识别具有更好的鲁棒性。通过Sobel算子提取中心图形的外缘轮廓检测,并通过低通滤波器平滑曲线。对于一维情况,Sobel算子可以表示为两个3×3矩阵(分别用于计算x方向和y方向的导数),相关计算公式如式(7)、式(8)所示。
Gx=$\left[\begin{array}{lll}-1& 0& 1\\ -2& 0& 2\\ -1& 0& 1\end{array}\right]$*I
Gy=$\left[\begin{array}{lll}-1& -2& -1\\ 0& 0& 0\\ 1& 2& 1\end{array}\right]$*I
式中:I为原始图像矩阵;*表示卷积操作。为了避免截断后的图像中存在多类物体和封闭图形区域,二值化后的图像必须是形态学封闭的。随后,用拟合算法对剩余部分进行拟合,以找到其最小的外切矩形。通过求解矩形最短边与水平线之间的夹角,即可得到最佳夹持角。为避免截取出的图像中存在多类物体,对二值化后的图像进行了腐蚀和膨胀操作。图5为整个过程,依次为识别、二值化、腐蚀、膨胀、求出当前垃圾的抓取角度为122°。随后,通过最小二乘法,做出剩余部分的最小外切矩形。一旦有了最小外接矩形,可以通过计算短边与水平线之间的夹角来得到最佳夹持角。假设矩形的一个角为(x1, y1),另一个对角为(x2, y2),且假设(x1 < x2)且(y1 < y2),则短边与水平线的夹角β可以通过反正切函数计算相关计算公式如式(9)所示。
β=arctan$\left(\frac{{y}_{2}-{y}_{1}}{{x}_{2}-{x}_{1}}\right)$
当将深度学习模型部署到生产环境中时,TensorRT是一个非常有用的工具。TensorRT可以对深度学习模型进行优化,以提高推理性能和减少内存占用。在使用TensorRT之前,首先用自制的数据集训练YOLOv8模型,再将训练好的模型上传到在Jetson上,TensorRT部署框架如图6所示。训练过程中可以使用各种优化算法和技术来提高模型的性能和准确度。
同时使用Int8量化可以帮助提高模型的计算速度和节省存储空间,Int8卷积推理流程图-量化。假设CNN的输入和输出分别为YX,则YX之间的映射关系如式(10)所示。
Y=F(X$\left|\Theta \right.$)=fL{…f2[f1(X|θ1)|θ2]|θL}
式(10)中:F(X$\left|\Theta \right.$)为非线性映射和学习参数构成的函数;f1(X|θ1)为卷积层、池化层和全连接层的计算过程;Θ为学习参数集合。通过使用Int8量化,可以提高模型的计算速度和存储效率,同时保持一定的准确性。这对于在边缘设备等计算资源有限的环境中部署模型非常有帮助。TensorRT具有强大的优化功能,可以对模型进行层融合和内存优化。层融合可以将多个层合并为一个层,减少计算量和内存使用。内存优化可以通过重用中间计算结果和使用低精度数据表示等方法,降低内存占用并提高计算速度。构建TensorRT引擎是对优化后的模型进行编译和优化的结果。引擎构建的过程包括网络定义、层参数设置、内存分配等步骤。TensorRT会根据设备的硬件特性和运行时环境进行自动优化和调整,以获得最佳的推理性能。
通过对YOLOv8-NANO智能生活垃圾快速分拣系统研究并制作了如图7所示的设备,所涉及的实验数据皆是在该设备进行实验统计。一种低功耗嵌入式的YOLOv8-NANO智能生活垃圾快速分拣系统最重要的两大核心是分类识别垃圾夹取的准确性和分类标签的正确性:柔性机械爪夹取到指定的收集容器内的正确性等价于识别夹取的准确性;人机交互面板正确显示的分类标签等价于分类标签识别的正确性。对自动打包、自动压缩和语音操作功能的实现则代表了对其他功能模块的准确完成。
将实验设备分类投放到广场、校园垃圾分拣站和走廊等公共场所中进行测试,记录测试数据,进行了多类别垃圾实验测试,通过不同颜色、不同形状、不同大小获得多个类别的垃圾的识别,对橘子、白萝卜、胡萝卜、电池(#1、#5、#7)、鹅卵石、烟头、纸巾、矿泉水瓶、易拉罐(不同颜色)等13种垃圾进行识别和分拣,识别结果如图8所示,即使在有遮挡的情况下,也能准确识别图中的香烟。将多种类别的垃圾同时投入,在摄像头检测到垃圾后,会自动识别分析出垃圾种类、具体坐标和最佳夹持角度,通过机械臂进行自动拾取。
经过测试,可以有效实现多目标的检测,检测mAP0.5:0.95(当IoU阈值在0.5~0.95逐步增加时的平均精度)为97.7%,测试精度为97%,目标检测速度达到30.6帧/s。虽然无法与已经成熟的高精度平台相比,但是本平台的成本较低,且无需稳定的网络,易于普及,具有广泛的应用前景。实验结果如表1所示,在与YOLOv5和YOLOv3的在NANO上加速的效果对比表明,基于YOLOv8-NANO网络的垃圾分类检测模型检测准确率高,鲁棒性好。并且有助于垃圾智能分类大规模普及,有利于实现可持续发展。
在机械爪抓取精度方面,采用COREXY等多重坐标变换、重复定位精度减小物理误差,其精度可达到2 mm,足以做到抓取如7号电池、烟蒂等细小的垃圾,做到精准夹取。综上所述,提出的一种低功耗嵌入式的YOLOv8-NANO智能生活垃圾快速分拣系统装置,其mAP为97%,目标检测速度为30.6帧/s,垃圾分拣速度在无需压缩的情况下为6件/min。
为了实现垃圾自动按类处理,通过研究基于视觉与机械手配合的YOLOv8-NANO智能生活垃圾快速分拣系统,实现对垃圾的精准快速的分类与投放。视觉部分采用YOLOv8网络进行分类模型的训练,最后将模型在NANO嵌入式平台进行部署测试,从而验证了方法的有效性。该系统不仅提高了垃圾处理的效率,而且减少了人工分拣的劳动强度,具有显著的经济效益和社会效益。通过实验验证,该系统在处理各种生活垃圾时表现出色,准确率高,响应速度快,能够满足实际应用的需求。未来,随着技术的进一步优化和升级,该系统有望在更多领域得到广泛应用,为智能垃圾分类提供更加高效、可靠的解决方案。
  • 浙江省“尖兵”“领雁”研发攻关计划(2023C03186)
  • 浙江省高等教育“十四五”教学改革项目(jg20220284)
参考文献 引证文献
排序方式:
[1]
Dokl M, Van Fan Y, Vujanović A, et al. A waste separation system based on sensor technology and deep learning: a simple approach applied to a case study of plastic packaging waste[J]. Journal of Cleaner Production, 2024, 450: 141762.
[2]
Satav A G, Kubade S, Amrutkar C, et al. A state-of-the-art review on robotics in waste sorting: scope and challenges[J]. International Journal on Interactive Design and Manufacturing, 2023, 17(6): 2789-2806.
[3]
Fang B, Yu J, Chen Z, et al. Artificial intelligence for waste management in smart cities: a review[J]. Environmental Chemistry Letters, 2023, 21(4): 1959-1989.
[4]
Nwokediegwu Z Q S, Ugwuanyi E D, Dada M A, et al. AI-driven waste management systems: a comparative review of innovations in the USA and Africa[J]. Engineering Science & Technology Journal, 2024, 5(2): 507-516.
[5]
谢康康, 朱文忠, 肖顺兴, . 一种改进YOLOX_S的火焰烟雾检测算法[J]. 科学技术与工程, 2024, 24(8): 3298-3307.
Xie Kangkang, Zhu Wenzhong, Xiao Shunxing, et al. Flame smoke detection algorithm on improved YOLOX_S[J]. Science Technology and Engineering, 2024, 24(8): 3298-3307.
[6]
柏润泚, 王凯, 俞范文. 基于ResNet_Vd模型的生活垃圾智能分拣系统[J]. 环境工程, 2023, 41(S2): 1030-1033.
Bai Runci, Wang Kai, Yu Fanwen. ResNet_Vd based intelligent garbage sorting system[J]. Environmental Engineering, 2023, 41(S2): 1030-1033.
[7]
Varghese M P, Anooja V S, Akhila R, et al. IoT-based smart waste management system with level indicators for effective garbage waste segregation[C]//2024 Third International Conference on Distributed Computing and Electrical Circuits and Electronics (ICDCECE). Ballari: IEEE, 2024: 1-5.
[8]
Mangayarkarasi T, Sathyaraj C, Pradyumna G V, et al. Harnessing IOT for automated sorting system[C]// 2024 International Conference on Communication, Computing and Internet of Things (IC3IoT). Chennai: IEEE, 2024: 1-7.
[9]
Kahra M, Sridhar V, Breuß M. Fast morphological dilation and erosion for grey scale images using the Fourier transform[C]// International Conference on Scale Space and Variational Methods in Computer Vision. Cham: Springer International Publishing, 2021: 65-77.
[10]
Mortos Y C Y, Malbog M A, Nipas M, et al. SMARTSORT: YOLOv4-driven smart sorting for household waste management[C]// E3S Web of Conferences. Paris: EDP Sciences, 2024, 488: 03025.
[11]
朱辰, 张西晨, 李森杰, . 基于云边协同的智能垃圾分类系统设计[J]. 实验技术与管理, 2023, 40(9): 74-80, 186.
Zhu Chen, Zhang Xichen, Li Shenjie, et al. Design of intelligent waste sorting system based on cloud edge collaboration[J]. Experimental Technology and Management, 2023, 40(9): 74-80.
[12]
郑宜健, 孟小源, 申文元. 基于机器视觉的景区智能垃圾桶设计[J]. 机械设计, 2024, 41(S1): 43-50.
Zheng Yijian, Meng Xiaoyuan, Shen Wenyuan. Design of intelligent trash bin in scenic spot based on machine vision[J]. Journal of Machine Design, 2024, 41(S1): 43-50.
[13]
张月文, 李松恒, 张炜, . 基于机器视觉的可回收垃圾智能分拣系统设计[J]. 实验室研究与探索, 2022, 41(7): 98-103, 107.
Zhang Yuewen, Li Songheng, Zhang Wei, et al. Design of intelligent sorting system for recyclable waste based on machine vision[J]. Reserch and Exploration in Laboratory, 2022, 41(7): 98-103, 107.
[14]
欧阳飞, 吴旭, 向东升. 基于改进YOLOv8的垃圾分类检测方法[J]. 计算机与现代化, 2023(8): 68-73.
Ouyang Fei, Wu xu, Xiang Dongsheng. Garbage classification and detection method based on improved YOLOv8[J]. Computer and Modernization, 2023(8): 68-73.
[15]
Mao W L, Chen W C, Wang C T, et al. Recycling waste classification using optimized convolutional neural network[J]. Resources, Conservation and Recycling, 2021, 164: 105132.
[16]
Ande A, Sundar G N, Thomas R, et al. Smart waste bins using DCNN and Internet of Things[C]// 2024 Second International Conference on Emerging Trends in Information Technology and Engineering. Vellore: IEEE, 2024: 1-7.
[17]
Lilhore U K, Simaiya S, Dalal S, et al. A smart waste classification model using hybrid CNN-LSTM with transfer learning for sustainable environment[J]. Multimedia Tools and Applications, 2024, 83(10): 29505-29529.
[18]
Mishra S, Yaduvanshi R, Rajpoot P, et al. An integrated deep-learning model for smart waste classification[J]. Environmental Monitoring and Assessment, 2024, 196(3): 279.
[19]
Raman R, Rawandale C J, Meenakshi R, et al. Real-time video management system for robotic waste sorting and recycling using IoT and machine learning[C]// 2023 Second International Conference on Smart Technologies for Smart Nation. Singapore: IEEE, 2023: 227-232.
[20]
Mohankumar A, Gowtham R, Gokul B, et al. Optimizing urban sustainability: a smart waste management system with Arduino technology[J]. Asian Journal of Applied Science and Technology, 2024, 8(2): 182-192.
2025年第25卷第9期
PDF下载
541
234
引用本文
BibTeX
文章信息
doi: 10.12404/j.issn.1671-1815.2403972
  • 接收时间:2024-05-28
  • 首发时间:2025-07-09
  • 出版时间:2025-03-28
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-05-28
  • 修回日期:2024-12-26
基金
浙江省“尖兵”“领雁”研发攻关计划(2023C03186)
浙江省高等教育“十四五”教学改革项目(jg20220284)
作者信息
    1 中国计量大学工程训练中心、创新学院, 杭州 310018
    2 浙江省在线检测装备校准技术研究重点实验室, 杭州 310018
    3 运达能源科技集团股份有限公司, 杭州 310012

通讯作者:

* 王斌锐(1978—),男,汉族,江西人,博士,教授。研究方向:检测技术与工业机器人应用。E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/kxjsygc/CN/10.12404/j.issn.1671-1815.2403972
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

Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)

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