Infrared and Laser Engineering, Volume. 53, Issue 1, 20230472(2024)
Image target detection algorithm based on YOLOv7-tiny in complex background
Figures & Tables(22)
Fig. 8. Comparison chart of test results before and after adding attention mechanism
Fig. 10. Before and after the introduction of small target detection layer detection results map
Fig. 13. Comparison of detection performance of different algorithms
Fig. 14. Before and after the algorithm improvement CAM contrast chart
Table 1. Comparison of detection performance of different attention mechanism algorithms
View tableView in ArticleTable 1. Comparison of detection performance of different attention mechanism algorithms
Model Params/M mAP@0.5 FPS/frame·s-1 GFLOPS YOLOv7-tiny 6.02 84.3 74 13.2 +SE 6.05 84.5 75 13.2 +CBAM 6.02 84.9 73 13.3 +EMA 6.06 85.9 75 13.5 +SMSE 8.99 86.8 71 15.6
Table 2. The RFB structure algorithm is introduced to detect the performance comparison
View tableView in ArticleTable 2. The RFB structure algorithm is introduced to detect the performance comparison
Model Params/M mAP@0.5 FPS/frame·s-1 GFLOPS YOLOv7-tiny 6.02 84.3 74 13.2 +XMB 6.30 85.6 77 14.7
Table 3. The small target detection layer algorithm is introduced to detect the performance comparison
View tableView in ArticleTable 3. The small target detection layer algorithm is introduced to detect the performance comparison
Model Params/M mAP@0.5 FPS/frame·s-1 GFLOPS YOLOv7-tiny 6.02 84.3 74 13.2 +XMB 6.10 85.1 70 15.5
Table 4. Improved loss function algorithm detection performance comparison
View tableView in ArticleTable 4. Improved loss function algorithm detection performance comparison
Model Params/M mAP@0.5 FPS/frame·s-1 GFLOPS YOLOv7-tiny 6.02 84.3 74 13.2 +SIoU 6.02 85.0 78 13.2
Table 5. A comparison of detection performance before and after deformable convolution is introduced
View tableView in ArticleTable 5. A comparison of detection performance before and after deformable convolution is introduced
Model Params/M mAP@0.5 FPS/frame·s-1 GFLOPS YOLOv7-tiny 6.02 84.3 74 13.2 +DCN 6.08 86.6 71 14.8
Table 6. Gradually add each module algorithm detection performance comparison
View tableView in ArticleTable 6. Gradually add each module algorithm detection performance comparison
SMSE RFB XMB SIoU DCN Params/ M mAP@0.5 FPS/ frame·s-1 GFLOPS 6.02 84.3 74 13.2 √ 8.99 86.8 71 15.6 √ √ 9.29 87.4 73 17.1 √ √ √ 9.39 88.2 69 20.1 √ √ √ √ 9.39 88.7 73 20.1 √ √ √ √ √ 9.45 90.4 72 21.7
Table 7. Comparison of detection performance of different target detection algorithms
View tableView in ArticleTable 7. Comparison of detection performance of different target detection algorithms
Model Params/M mAP@0.5 FPS/frame·s-1 GFLOPS YOLOv7-tiny 6.02 84.3 74 13.2 YOLOv7 37.2 87.3 57 104.8 YOLOv5l 46.1 86.4 42 107.9 YOLOv7-drone 9.45 90.4 72 21.7
Table 8. PASCAL VOC dataset detection performance comparison
View tableView in ArticleTable 8. PASCAL VOC dataset detection performance comparison
Model Params/M mAP@0.5 FPS/frame·s-1 GFLOPS YOLOv7-tiny 6.02 65.0 74 13.2 YOLOv7-drone 9.45 71.0 72 21.7
Tools
Get Citation
Copy Citation Text
Shan Xue, Hongyu An, Qiongying Lv, Guohua Cao. Image target detection algorithm based on YOLOv7-tiny in complex background[J]. Infrared and Laser Engineering, 2024, 53(1): 20230472
Paper Information
Category:
Received: Jul. 30, 2023
Accepted: --
Published Online: Mar. 19, 2024
The Author Email:
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20