Improved YOLOv5 infrared tank target detection method under ground background

The detection precision of infrared seeker directly affects the guidance precision of infrared guidance system. To solve the problem of low target detection accuracy caused by the change of imaging scale, complex ground background and inconspicuous infrared target characteristics when infrared image seeker detects ground tank targets. In this paper, a You Only Look Once, Transform Head Squeeze-and-Excitation (YOLOv5s-THSE) model is proposed based on the YOLOv5s model. A multi-head attention mechanism is added to the backbone and neck of the network, and deeper target features are extracted using the multi-head attention mechanism. The Cross Stage Partial, Squeeze-and-Exclusion module is added to the neck of the network to suppress the complex background and make the model pay more attention to the target. A small object detection head is introduced into the head of the network, and the CIoU loss function is used in the model to improve the detection accuracy of small objects and obtain more stable training regression. Through these several improvement measures, the background of the infrared target is suppressed, and the detection ability of infrared tank targets is improved. Experiments on infrared tank target datasets show that our proposed model can effectively improve the detection performance of infrared tank targets under ground background compared with existing methods, such as YOLOv5s, YOLOv5s + SE, and YOLOV 5 s + Convective Block Attention Module.