To solve the problems of matching omission and poor real-time performance of existing visual place recognition methods in scenes with changing viewpoints and environments, this paper proposed visual place recognition method based on one level feature fused with coordinate attention. Firstly, the relative place information of features was captured by coordinate attention. Secondly, an encoder for multi-scale feature fusion was constructed using dilated convolution and NetVLAD. Finally, the network was trained based on triplet loss. Validated by Pitts30k and Nordland datasets, the proposed method achieves the same recall accuracy and 19% faster retrieval speed compared with the state-of-the-art method Patch-NetVLAD of the same baseline in the test of position recognition. In the test of loop detection, the proposed method achieves a reasonable balance between robustness and retrieval speed.
| 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 |