Pavement flatness detection method and device based on image matching combined point cloud data

Abstract

The invention discloses a pavement flatness detection method and device based on image matching and point cloud data. The method comprises the following steps: acquiring three images of a pavement to be detected at the same moment through a trinocular camera; performing parallax comparison and matching pixel by pixel in the road surface areas in the three images; when the number of mismatched pixels is greater than a first preset threshold value, obtaining point cloud data of the to-be-detected road surface at the same moment through the laser radar; and calculating a distance difference of the same horizontal line of mismatched pixels corresponding to the point cloud data, and when the distance difference is greater than or equal to a second preset threshold value, performing road dangerous case early warning according to a mean value of point distance difference variances on each horizontal line. When the device is used for assisting manual driving, a driver can be reminded of the front road condition so that the driver can make adjustment in time, and vehicle bumping or distress caused by the fact that the driver is neglected to observe is prevented; when used for an automatic driving system, the system can give an early warning to the decision module, thereby facilitating the system to autonomously adjust the vehicle speed and guaranteeing the driving comfort and safety.

Description

technical field [0001] The invention relates to the fields of computer and artificial intelligence, in particular to a method and device for detecting road surface roughness based on image matching and point cloud data. Background technique [0002] The traffic system is composed of people, vehicles, roads and the environment. Among them, the road is an important factor, and its smoothness largely limits the upper limit of driving speed. Road surface roughness and driving speed jointly determine driving comfort and safety, which in turn affects the operating efficiency of the entire traffic system. [0003] Pavement roughness is so important that its evaluation value is one of the important indicators for road construction acceptance. However, because the road will be exposed to the sun or rain during long-term use, the smoothness of the road surface may decline over time, such as cracks, subsidence, and partial fracture, which will affect driving safety. [0004] In manua...

AI Technical Summary

Problems solved by technology

[0006] In view of the above problems, the road surface roughness detection method and device based on image matching combined with point cloud data of the present invention is proposed, aiming to solve the current problem that it is impossible to accurately identify small road surface defects and to better ensure the comfort and safety of driving

Images