Overview of pedestrian detection based on infrared image

Object detection is a research hotspot in theory and application in recent years. With the development of infrared imaging technology, it can be used normally in various harsh detection environments (strong light irradiation, dark environment, smoke occlusion), which makes the pedestrian detection algorithm based on infrared image have great research and application prospects. According to the development of infrared pedestrian detection technology, this paper will describe the traditional detection algorithm and the object detection algorithm based on deep learning. Firstly, the traditional detection algorithm flow is described, and the milestone detectors are summarized. Then, the foundation laying process of two main algorithm schools of infrared pedestrian detection based on deep learning is reviewed, including two-stage algorithm represented by R-CNN series and one-stage algorithm represented by YOLO and SSD, Finally, through the existing problems, the future research of infrared pedestrian detection technology is prospected.