TSD-Truncated Structurally Aware Distance for Small Pest Object Detection | |
Huang, Xiaowen1,2; Dong, Jun1,3; Zhu, Zhijia1,2; Ma, Dong1; Ma, Fan1; Lang, Luhong4 | |
刊名 | SENSORS |
2022-11-01 | |
卷号 | 22 |
关键词 | small object detection pest detection truncated structurally aware distance truncated structurally aware loss faster R-CNN |
DOI | 10.3390/s22228691 |
通讯作者 | Dong, Jun(dong.jun@iim.ac.cn) |
英文摘要 | As deep learning has been successfully applied in various domains, it has recently received considerable research attention for decades, making it possible to efficiently and intelligently detect crop pests. Nevertheless, the detection of pest objects is still challenging due to the lack of discriminative features and pests' aggregation behavior. Recently, intersection over union (IoU)-based object detection has attracted much attention and become the most widely used metric. However, it is sensitive to small-object localization bias; furthermore, IoU-based loss only works when ground truths and predicted bounding boxes are intersected, and it lacks an awareness of different geometrical structures. Therefore, we propose a simple and effective metric and a loss function based on this new metric, truncated structurally aware distance (TSD). Firstly, the distance between two bounding boxes is defined as the standardized Chebyshev distance. We also propose a new regression loss function, truncated structurally aware distance loss, which consider the different geometrical structure relationships between two bounding boxes and whose truncated function is designed to impose different penalties. To further test the effectiveness of our method, we apply it on the Pest24 small-object pest dataset, and the results show that the mAP is 5.0% higher than other detection methods. |
资助项目 | Central Government Guides Local Science and Technology Development Special Foundation Projects of China[S202107d08050071] ; Natural Science Research Project of Colleges and Universities in Anhui Province[KJ2020A0924] ; Natural Science Foundation of Anhui Province[2208085QC83] |
WOS研究方向 | Chemistry ; Engineering ; Instruments & Instrumentation |
语种 | 英语 |
出版者 | MDPI |
WOS记录号 | WOS:000887624300001 |
资助机构 | Central Government Guides Local Science and Technology Development Special Foundation Projects of China ; Natural Science Research Project of Colleges and Universities in Anhui Province ; Natural Science Foundation of Anhui Province |
内容类型 | 期刊论文 |
源URL | [http://ir.hfcas.ac.cn:8080/handle/334002/131621] |
专题 | 中国科学院合肥物质科学研究院 |
通讯作者 | Dong, Jun |
作者单位 | 1.Chinese Acad Sci, Hefei Inst Phys Sci, Hefei 230031, Peoples R China 2.Univ Sci & Technol China, Hefei 230026, Peoples R China 3.Anhui Zhongke Deji Intelligence Technol Co Ltd, Hefei 230045, Peoples R China 4.Wuhu Inst Technol, Wuhu 241006, Peoples R China |
推荐引用方式 GB/T 7714 | Huang, Xiaowen,Dong, Jun,Zhu, Zhijia,et al. TSD-Truncated Structurally Aware Distance for Small Pest Object Detection[J]. SENSORS,2022,22. |
APA | Huang, Xiaowen,Dong, Jun,Zhu, Zhijia,Ma, Dong,Ma, Fan,&Lang, Luhong.(2022).TSD-Truncated Structurally Aware Distance for Small Pest Object Detection.SENSORS,22. |
MLA | Huang, Xiaowen,et al."TSD-Truncated Structurally Aware Distance for Small Pest Object Detection".SENSORS 22(2022). |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论