TGC-YOLOv5: An Enhanced YOLOv5 Drone Detection Model Based on Transformer, GAM & CA Attention Mechanism
Drone detection is a significant research topic due to the potential security threats posed by the misuse of drones in both civilian and military domains. However, traditional drone detection methods are challenged by the drastic scale changes and complex ambiguity during drone flight, and it is dif...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-07-01
|
| Series: | Drones |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2504-446X/7/7/446 |
| _version_ | 1797589582932344832 |
|---|---|
| author | Yuliang Zhao Zhongjie Ju Tianang Sun Fanghecong Dong Jian Li Ruige Yang Qiang Fu Chao Lian Peng Shan |
| author_facet | Yuliang Zhao Zhongjie Ju Tianang Sun Fanghecong Dong Jian Li Ruige Yang Qiang Fu Chao Lian Peng Shan |
| author_sort | Yuliang Zhao |
| collection | DOAJ |
| description | Drone detection is a significant research topic due to the potential security threats posed by the misuse of drones in both civilian and military domains. However, traditional drone detection methods are challenged by the drastic scale changes and complex ambiguity during drone flight, and it is difficult to detect small target drones quickly and efficiently. We propose an information-enhanced model based on improved YOLOv5 (TGC-YOLOv5) for fast and accurate detection of small target drones in complex environments. The main contributions of this paper are as follows: First, the Transformer encoder module is incorporated into YOLOv5 to augment attention toward the regions of interest. Second, the Global Attention Mechanism (GAM) is embraced to mitigate information diffusion among distinct layers and amplify the global cross-dimensional interaction features. Finally, the Coordinate Attention Mechanism (CA) is incorporated into the bottleneck part of C3, enhancing the extraction capability of local information for small targets. To enhance and verify the robustness and generalization of the model, a small target drone dataset (SUAV-DATA) is constructed in all-weather, multi-scenario, and complex environments. The experimental results show that based on the SUAV-DATA dataset, the AP value of TGC-YOLOv5 reaches 0.848, which is 2.5% higher than the original YOLOv5, and the Recall value of TGC-YOLOv5 reaches 0.823, which is a 3.8% improvement over the original YOLOv5. The robustness of our proposed model is also verified on the Real-World open-source image dataset, achieving the best accuracy in light, fog, stain, and saturation pollution images. The findings and methods of this paper have important significance and value for improving the efficiency and precision of drone detection. |
| first_indexed | 2024-03-11T01:08:29Z |
| format | Article |
| id | doaj.art-24409de2c0d7435a94c091c9c5e51e88 |
| institution | Directory Open Access Journal |
| issn | 2504-446X |
| language | English |
| last_indexed | 2024-03-11T01:08:29Z |
| publishDate | 2023-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Drones |
| spelling | doaj.art-24409de2c0d7435a94c091c9c5e51e882023-11-18T19:01:12ZengMDPI AGDrones2504-446X2023-07-017744610.3390/drones7070446TGC-YOLOv5: An Enhanced YOLOv5 Drone Detection Model Based on Transformer, GAM & CA Attention MechanismYuliang Zhao0Zhongjie Ju1Tianang Sun2Fanghecong Dong3Jian Li4Ruige Yang5Qiang Fu6Chao Lian7Peng Shan8School of Information Science and Engineering, Northeastern University, Shenyang 110819, ChinaSchool of Information Science and Engineering, Northeastern University, Shenyang 110819, ChinaSchool of Information Science and Engineering, Northeastern University, Shenyang 110819, ChinaSchool of Information Science and Engineering, Northeastern University, Shenyang 110819, ChinaSchool of Information Science and Engineering, Northeastern University, Shenyang 110819, ChinaSchool of Information Science and Engineering, Northeastern University, Shenyang 110819, ChinaShijiazhuang Campus of Army Engineer University, Shijiazhuang 050003, ChinaSchool of Information Science and Engineering, Northeastern University, Shenyang 110819, ChinaSchool of Information Science and Engineering, Northeastern University, Shenyang 110819, ChinaDrone detection is a significant research topic due to the potential security threats posed by the misuse of drones in both civilian and military domains. However, traditional drone detection methods are challenged by the drastic scale changes and complex ambiguity during drone flight, and it is difficult to detect small target drones quickly and efficiently. We propose an information-enhanced model based on improved YOLOv5 (TGC-YOLOv5) for fast and accurate detection of small target drones in complex environments. The main contributions of this paper are as follows: First, the Transformer encoder module is incorporated into YOLOv5 to augment attention toward the regions of interest. Second, the Global Attention Mechanism (GAM) is embraced to mitigate information diffusion among distinct layers and amplify the global cross-dimensional interaction features. Finally, the Coordinate Attention Mechanism (CA) is incorporated into the bottleneck part of C3, enhancing the extraction capability of local information for small targets. To enhance and verify the robustness and generalization of the model, a small target drone dataset (SUAV-DATA) is constructed in all-weather, multi-scenario, and complex environments. The experimental results show that based on the SUAV-DATA dataset, the AP value of TGC-YOLOv5 reaches 0.848, which is 2.5% higher than the original YOLOv5, and the Recall value of TGC-YOLOv5 reaches 0.823, which is a 3.8% improvement over the original YOLOv5. The robustness of our proposed model is also verified on the Real-World open-source image dataset, achieving the best accuracy in light, fog, stain, and saturation pollution images. The findings and methods of this paper have important significance and value for improving the efficiency and precision of drone detection.https://www.mdpi.com/2504-446X/7/7/446drone detectionsmall target dronesYOLOv5TransformerGAMCA |
| spellingShingle | Yuliang Zhao Zhongjie Ju Tianang Sun Fanghecong Dong Jian Li Ruige Yang Qiang Fu Chao Lian Peng Shan TGC-YOLOv5: An Enhanced YOLOv5 Drone Detection Model Based on Transformer, GAM & CA Attention Mechanism Drones drone detection small target drones YOLOv5 Transformer GAM CA |
| title | TGC-YOLOv5: An Enhanced YOLOv5 Drone Detection Model Based on Transformer, GAM & CA Attention Mechanism |
| title_full | TGC-YOLOv5: An Enhanced YOLOv5 Drone Detection Model Based on Transformer, GAM & CA Attention Mechanism |
| title_fullStr | TGC-YOLOv5: An Enhanced YOLOv5 Drone Detection Model Based on Transformer, GAM & CA Attention Mechanism |
| title_full_unstemmed | TGC-YOLOv5: An Enhanced YOLOv5 Drone Detection Model Based on Transformer, GAM & CA Attention Mechanism |
| title_short | TGC-YOLOv5: An Enhanced YOLOv5 Drone Detection Model Based on Transformer, GAM & CA Attention Mechanism |
| title_sort | tgc yolov5 an enhanced yolov5 drone detection model based on transformer gam ca attention mechanism |
| topic | drone detection small target drones YOLOv5 Transformer GAM CA |
| url | https://www.mdpi.com/2504-446X/7/7/446 |
| work_keys_str_mv | AT yuliangzhao tgcyolov5anenhancedyolov5dronedetectionmodelbasedontransformergamcaattentionmechanism AT zhongjieju tgcyolov5anenhancedyolov5dronedetectionmodelbasedontransformergamcaattentionmechanism AT tianangsun tgcyolov5anenhancedyolov5dronedetectionmodelbasedontransformergamcaattentionmechanism AT fanghecongdong tgcyolov5anenhancedyolov5dronedetectionmodelbasedontransformergamcaattentionmechanism AT jianli tgcyolov5anenhancedyolov5dronedetectionmodelbasedontransformergamcaattentionmechanism AT ruigeyang tgcyolov5anenhancedyolov5dronedetectionmodelbasedontransformergamcaattentionmechanism AT qiangfu tgcyolov5anenhancedyolov5dronedetectionmodelbasedontransformergamcaattentionmechanism AT chaolian tgcyolov5anenhancedyolov5dronedetectionmodelbasedontransformergamcaattentionmechanism AT pengshan tgcyolov5anenhancedyolov5dronedetectionmodelbasedontransformergamcaattentionmechanism |