Road surface detection method and device

Abstract

Disclosed are a road surface detection method and device. The road surface detection method includes: obtaining a V disparity map including the road surface; and extracting a straight line from the V disparity map as the road surface, wherein extracting the straight line from the V disparity map as the road surface includes: using Hough transform to extract the first straight line as the first road surface approximation; and based on the first straight line, use the least square method to fit the second straight line as the road surface. Utilize the road surface detection method and device of the present invention, can overcome the shortcoming that Hough transform may deviate from the road surface area because of focusing on the straight line with the most passing points, and can also overcome the shortcoming that the ordinary least square method is sensitive to noise, and can detect more accurately pavement.

Description

technical field [0001] The present invention relates to image processing, and more particularly to a road surface detection method and device. Background technique [0002] The application of driver assistance systems is becoming more and more popular. The road or lane warning system (Lane / Road detection warning, LDW / RDW) is a subsystem of the driving assistance system, which can avoid collisions and determine the driving direction more accurately. Road detection is very critical to the LDW / RDW system, and further processing, such as warning, is possible only on the basis of knowing the road information. [0003] Stereo cameras have the advantage of being able to obtain high-precision disparity maps and distance images, so they have been widely used. One of the tasks in understanding 3D road scenes is road surface plane detection. However, the existing methods are not suitable for complex environments and are easily disturbed by noise. [0004] In the non-patent literatu...

AI Technical Summary

Problems solved by technology

[0005] The current road detection method based on the disparity map usually only uses a simple Hough transform and some simple position features or shape features, so the robustness to complex environments is not strong

Images