Due to the low resolution, high noise, and strong spatial correlation of thermal infrared images, an improved YOLOv7 algorithm is proposed to solve the problem of low accuracy in road traffic infrared target detection algorithms. Introducing CBAM attention mechanism in the backbone network and Neck network to enhance the ability to locate small targets; Replace the original SPPCSPC module with an ASPP module in the Neck network to improve the detection accuracy of images with lower resolutions; Replace the original CIoU loss function with the WIoU loss function to enhance the detection accuracy of weak and small targets. Experiments on the Chinese thermal infrared data set CTIR show that compared with the YOLOv7 algorithm, the detection accuracy mAP value of the proposed algorithm is increased by 3.4%, and the detection performance is improved.
Object Detection Algorithm Based on YOLOv7 Traffic Infrared Image
11.10.2023
2510952 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Nighttime Vehicle Object Detection Based on Improved YOLOv7
TIBKAT | 2023
|MS-YOLOv7:YOLOv7 Based on Multi-Scale for Object Detection on UAV Aerial Photography
DOAJ | 2023
|