CGT-YOLOv5n: A Precision Model for Detecting Mouse Holes Amid Complex Grassland Terrains
Source abstract
Abstract
This study employs unmanned aerial vehicles (UAVs) to detect mouse holes in grasslands, offering an effective tool for grassland ecological conservation. We introduce the specially designed CGT-YOLOv5n model, addressing long-standing challenges UAVs face, particularly the decreased detection accuracy in complex grassland environments due to shadows and obstructions. The model incorporates a Context Augmentation Module (CAM) focused on improving the detection of small mouse holes and mitigating the interference of shadows. Additionally, to enhance the model’s ability to recognize mouse holes of varied morphologies, we have integrated an omni-dimensional dynamic convolution (ODConv), thereby increasing the model’s adaptability to diverse image features. Furthermore, the model includes a Task-Specific Context Decoupling (TSCODE) module, independently refining the contextual semantics and spatial details for classification and regression tasks and significantly improving the detection accuracy. The empirical results show that when the intersection over union (IoU) threshold is set at 0.5, the model’s mean average precision (mAP_0.5) for detection accuracy reaches 92.8%. The mean average precision (mAP_0.5:0.95), calculated over different IoU thresholds ranging from 0.5 to 0.95 in increments of 0.05, is 46.2%. These represent improvements of 3.3% and 4.3%, respectively, compared to the original model. Thus, this model contributes significantly to grassland ecological conservation and provides an effective tool for grassland management and mouse pest control in pastoral areas.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Species Distribution and Climate Change · Wildlife Ecology and Conservation · Smart Agriculture and AI
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
College of Computer and Information, Inner Mongolia Agricultural University, Hohhot 010018, China
Funding
National Natural Science Foundation of China · Scientific Research Project of Higher Education Institutions in Inner Mongolia Autonomous Region · Inner Mongolia Natural Science Foundation joint fund project
References (41)
Show source-provided references
1. Yun, J.F., and Liu, D.F. (2005, January 20–23). Characteristics, status and construction of grassland resources in Inner Mongolia desert area. Chinese Forestry Society. Proceedings of the 2005 CCSA Annual Academic Conference 26 Session Proceedings (1), Xinjiang, China.
2. Chen, W., Liu, W., Zhao, Y., Lu, J., Lv, S., and Muyassar, S. (2023). Monitoring and Control Methods of Spatial Distribution of Pests and Rodents in Yili Grassland. J. Grassl. Forage Sci., 68–73.
3. He, D., Huang, X., Tian, Q., and Zhang, Z. (2020). Changes in vegetation growth dynamics and relations with climate in inner Mongolia under more strict multiple pre-processing. (2000–2018). Sustainability, 12.10.3390/su12062534 ↗
4. Seasonal pattern and dynamic mechanism of population survival of long-clawed gerbils in agro-pastoral ecotone in inner monglia
5. Sun, H.L. (2000). Encyclopedia of China’s Resource Sciences, China Encyclopedia Publishing House.
6. Monitoring Mouse-Hole Density by Rhombomys opimus in Desert Forests with UAV Remote Sensing Technology
7. Group coverage of burrowentrances and distribution characteristics of desert forest-dwelling Rhombomys opimus based on unmanned aerial vehicle (UAV) low-altitude remote sensing: A case study at the southern margin of the Gurbantunggut Desert in Xinjiang
8. Identification of rat holes in the typical area of “Three-River Headwaters” region By UAV remote sensing
9. Application of UAV low-altitude image on rathole monitoring of Eolagurus luteus
10. Comparative Analysis of Image Classification Algorithms Based on Traditional Machine Learning and Deep Learning
11. Weed classification of remote sensing by UAV in ecological irrigation areas based on deep learning
12. UAV real-time monitoring for forest pest based on deep learning
13. A Study of Rodent Monitoring in Ruoergai Grassland Based on Convolutional Neural Network
14. YOLOv3 rat hole recognition technology for UAV remote sensing images
15. Design of a Kiwifruit Orchard Pest and Disease Detection System Based on Aerial and Ground Multisource Information
16. Leng, R.X. (2023). Application of Transmission Line Foreign Object Recognition Algorithm Based on YOLOv8, Northeast Agricultural University.
17. Wang, Z., Yuan, G., Zhou, H., Ma, Y., and Ma, Y. (2023). Foreign-Object Detection in High-Voltage Transmission Line Based on Improved YOLOv8m. Appl. Sci., 13.10.3390/app132312775 ↗
18. Du, Y.F., Huang, L., Zhao, Z., and Li, G. (2023). Landslide Identification and Detection in High-Resolution Remote Sensing Images Based on DETR. Bull. Surv. Mapp., 16–20.
19. (2023, March 06). Glenn Jocher. Available online: https://github.com/ultralytics/yolov5.
20. Tiny object detection with context enhancement and feature purification10.1016/j.eswa.2022.118665 ↗
21. Lin, L.K. (2021). Full-Field High-Resolution Cell Morphology Analysis System and Its Application Research. [Master’s Thesis, Nanjing University].
22. Li, G.B. (2021). Research on Pedestrian Detection Technology Based on Deep Learning. [Master’s Thesis, Guizhou University].
23. Wang, H.M. (2022). Research on Railroad Traffic Safety Image Recognition Technology Based on Deep Learning. [Ph.D. Thesis, Lanzhou Jiaotong University].
24. Wu, N., Mu, C.G., He, Y., and Liu, T.H. (2023). Multi-scale infrared and visible image fusion based on nested connections. J. Beijing Univ. Aeronaut. Astronaut., 1–11.
25. Zhuang, J., Qin, Z., Yu, H., and Chen, X. (2023). Task-Specific Context Decoupling for Object Detection. arXiv.
26. Chen, Y., Dai, X., Liu, M., Chen, D., Yuan, L., and Liu, Z. (2020, January 13–19). Dynamic convolution: Attention over convolution kernels. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.10.1109/CVPR42600.2020.01104 ↗
27. Li, C., Zhou, A., and Yao, A. (2022). Omni-dimensional dynamic convolution. arXiv.
28. Diagnosis analysis and management of bird damage fault on transmission line of 330 kV Yushu networking project
29. Git Code, T.L. (2022, October 15). Available online: https://github.com/tzutalin/labelImg.
30. Receiver operating characteristics curves and related decision measures: A tutorial10.1016/j.chemolab.2005.05.004 ↗
31. Measures derived from a 2 × 2 table for an accuracy of a diagnostic test10.4172/2155-6180.1000128 ↗
32. Research on high-precision wafer defect detection method based on deep learning
33. Li, H. (2022). Research on Object Detection Algorithm Based on Lightweight Network, University of Chinese Academy of Sciences (Institute of Optoelectronic Technology, Chinese Academy of Sciences).
34. Faster r-cnn: Towards real-time object detection with region proposal networks
35. Redmon, J., and Farhadi, A. (2018). Yolov3: An incremental improvement. arXiv.
36. Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C.-Y., and Berg, A.C. (2016, January 11–14). Ssd: Single shot multibox detector. Proceedings of the Computer Vision–ECCV 2016: 14th European Conference, Amsterdam, The Netherlands. Proceedings, Part I 14.10.1007/978-3-319-46448-0_2 ↗
37. Reis, D., Kupec, J., Hong, J., and Daoudi, A. (2023). Real-Time Flying Object Detection with YOLOv8. arXiv.
38. Lv, W., Zhao, Y., Xu, S., Wei, J., Wang, G., Cui, C., Du, Y., Dang, Q., and Liu, Y. (2023). Detrs beat yolos on real-time object detection. arXiv.
39. Sum of ranking differences compares methods or models fairly10.1016/j.trac.2009.09.009 ↗
40. Method and model comparison by sum of ranking differences in cases of repeated observations (ties)10.1016/j.chemolab.2013.06.007 ↗
41. Sziklai, B.R., and Héberger, K. (2020). Apportionment and districting by Sum of Ranking Differences. PLoS ONE, 15.10.1371/journal.pone.0229209 ↗