Highway pavement detection method based on image processing

Abstract

The invention discloses a highway pavement detection method based on image processing, comprising the following steps: continuously collecting at least one video frame image of a video file, and obtaining a target video frame image according to the at least one video frame image; detecting the edge of the target video frame image to obtain an edge image containing road edge pixels; scanning the edge image to obtain a road region, transversely dividing the road region into sub-regions, and detecting each sub-region through probabilistic Hough transformation to obtain road edge line segments; calculating the vanishing point according to all the edge line segments in the sub-region on the top end of the road region, and determining the middle control points of each non-bottom sub-region and the boundary points of the bottommost sub-region according to whether there is an intersection point between the line with the maximum slope and the line with the minimum slope in each sub-region; anddrawing left and right lane edge lines according to the middle control points, the boundary points and the vanishing point. By applying the embodiment of the invention, the adaptability to a bend scene is improved.

Description

technical field [0001] The invention relates to the field of road surface detection, in particular to an image processing-based expressway road surface detection method. Background technique [0002] Intelligent video surveillance of expressways usually only focuses on the road surface area in the picture, but the surveillance images often contain backgrounds such as sky, trees, buildings, etc., which undoubtedly increases the computational overhead of the monitoring algorithm. At the same time, non-road surface areas are often accompanied by shaking leaves, light These interference factors also affect the accuracy of monitoring. Therefore, it is necessary to extract the highway road surface area in the video image as the preprocessing of video surveillance, filter out the background irrelevant to the road surface, reduce the redundant data in the image, improve the calculation speed, and avoid image information in irrelevant areas from being used for later image processing....

AI Technical Summary

Problems solved by technology

[0004] Based on the above ideas, Wang et al. published "Image and Vision computing" in "Image and Vision computing" in 2004. It is assumed that the roads in the monitoring scene are parallel, and the B-snake curve is used to fit the lane line. Compared with the spline Interpolation curve, this method makes the generated curve approach as much as possible instead of passing through the interpolation point, the fitted curve is flexible and smooth but requires multiple Hough transforms, which is not conducive to controlling the calculation amount of the algorithm; Jung et al. in 2005 in "Imageand "Lane Following and Lane Departure Using a Linear-parabolic Model" was published on Vision Computing, which uses sub-regional lane line fitting, the nearby road area uses a linear model to fit the lane line, and the far road area uses a parabolic model to fit Lane lines, but the far and near areas in this algorithm are divided in advance, the adaptability is poor, and the scope of use is limited; Lipski et al. published "AFast and Robust Approach to Lane Marking Detection and Lane Tracking》Calculate the local histogram of the road surface image, extract the color of the road surface image and road direction and other feature information. Although this method is less affected by the shape change of the road, it can monitor the changes in the lighting conditions and shadows of the road surface in the environment. Coverage and the decrease in the clarity of lane lines will affect the detection results; Lee et al. published "Effectivelane detection and tracking method using statistical modeling of color and lane edge" in "In Proceedings of 4th International Conference on Computer Sciences and Convergence Information Technology" in 2009 -orientation" uses lane line color and edge information to obtain road lane line pixels and calculates the histogram of edge information and HSV space color information, uses Bayesian criteria to classify each pixel in the image, extracts lane line pixels, and Hough Transformation for fitting; Kong et al. published "General Road Detecti on from A Signal Image》Use Gabor filtering to calculate the local texture of pixels, obtain edge information, and perform Hough transform to find and locate the road ROI area in the image. This method is less affected by noise but computationally complex too high

[0005] Among the above methods, the model-based method only needs to solve fewer model parameters, has a small amount of calculation, and has strong robustness to noise, but the accuracy of extracting the road surface area depends on the selection of the model and the solution method. The method of feature extraction is directly related to the selection of road surface features, and the expressway will have curves, and the lane line fitting of the entire picture directly affects the accuracy of left and right lane division in this scene

Images