Ground infrared target detection method based on a parallel attention mechanism (<i>Invited</i>)

Table 1. Experimental results of different target detection models based on the Infrared-VOC dataset
View table
View in Article
Table 1. Experimental results of different target detection models based on the Infrared-VOC dataset
Model GFLOPS FPS/frame·s^-1 mAP@0.5
Faster R-CNN 224.5 6.5 77.1%
SSD500 137 38 72.6%
YOLOv3 155.1 48.2 75.3%
PAM-YOLO 50.6 46 82.2%

Table 2. Ablation experiment based on Infrared-VOC dataset

View table

View in Article

Table 2. Ablation experiment based on Infrared-VOC dataset

YOLOv3	CIoU loss and focal loss	PAConv	PAM-YOLO	GFLOPS	Training time/h	mAP@0.5	AP@0.5
YOLOv3	CIoU loss and focal loss	PAConv	PAM-YOLO	GFLOPS	Training time/h	mAP@0.5	Car	Bus
√				155.1	2.12	75.3%	72.9%	77.6%
√	√			155.1	2.10	77.1%	78.2%	76.0%
√		√		151.3	1.48	78.4%	80.1%	76.8%
√	√	√		151.3	1.48	80.4%	80.7%	80.1%
√	√	√	√	50.6	0.927	82.2%	81.4%	82.9%

Table 3. Comparison test of SE module and PAM module
View table
View in Article
Table 3. Comparison test of SE module and PAM module
YOLOv3 with loss SE module PAM module GFLOPS Training time/h Parameters/M mAP@0.5
√ 155.1 2.10 59.66 77.1%
√ √ 206.3 3.14 59.67 78.0%
√ √ 151.3 1.48 56.86 80.4%

Tools

Get Citation

Copy Citation Text

Xiaofeng Zhao, Yebin Xu, Fei Wu, Jiahui Niu, Wei Cai, Zhili Zhang. Ground infrared target detection method based on a parallel attention mechanism (Invited)[J]. Infrared and Laser Engineering, 2022, 51(4): 20210290

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites