A Study of YOLO Algorithm for Multi-target Detection

Abstract

With the development of deep learning, target detection has become one of the research directions of many scholars. As one of the more mature algorithms, the single-stage YOLO algorithms have been widely used in real life. Combining the development history of the YOLO algorithm, this article focuses on the main framework and main content of the current latest YOLOv5 algorithm, and uses the YOLOv5s model to identify and detect multi-target. The test results show that YOLOv5s algorithm has good detection effect and wide application meaning in real life.