Abstract: In order to meet the requirements of the accuracy and timeliness of lane detection in intelligent vehicle video forensics equipment for vehicle violation behavior forensics, a fast lane detection algorithm based on region of interest (ROI) model is proposed. Firstly, the edge detection operator is used to detect the lane edge information based on the characteristic that the lane and road have high contrast. Then the improved Hough transform is used to detect the possible lane based on the region of interest model. In order to extract the lane accurately, the algorithm is based on the vanishing point principle of the perspective image, and uses the inverse perspective transformation to eliminate the pseudo lane and complete the lane accurate positioning. Experiments show that the algorithm has the advantages of high positioning accuracy, high speed, and good robustness, which can meet the performance requirements of highway intelligent vehicle video equipment forensics.