Detection of vehicle line pressing on straight lane based on Cascade
Hough transform principle and improved YOLOV5 model
In this study, a detection method based on the Cascaded Hough Transform
(or CHT for short) and an improved YOLOV5 model is proposed for the
detection of vehicle pressing lines in straight lanes. This method
combines traditional image processing with depth learning by integrating
background modeling, cascade Hough transform principle and YOLOV5 depth
learning framework. The specific process of this method is as follows.
First, MOG2 background modeling is used to obtain the road background
image, then the CHT principle is used to extract the lane lines in the
road background image, and finally the improved YOLOV5 model is used to
recognize the position of vehicles and judge whether the vehicles are
pressing the line. The experimental results show that this method has
good recognition speed and accuracy, can meet the basic requirements of
road monitoring vehicle line detection, and also provides some reference
experience for the research of vehicle line detection.