Pavement defect detection model establishment method and pavement defect detection method and system

Abstract

The invention provides a pavement defect detection model establishment method, a pavement defect detection method and a pavement defect detection system. The pavement defect detection model establishment method comprises the following steps: obtaining pavement defects at all positions on a calibrated road section and vertical jitter parameters of a test vehicle at all positions when the test vehicle runs on the calibrated road section; and establishing a to-be-trained model based on machine learning, training the to-be-trained model with the up-down jitter parameters as a training set and the corresponding pavement defects as a test set, and obtaining a pavement defect detection model, wherein the pavement defect detection model comprises a Lasso model and a linear SVM model which are cascaded. According to the technical scheme, the defect detection accuracy and the defect detection speed can be greatly improved.

Description

technical field [0001] The present invention relates to the technical field of pavement detection, in particular to a method for establishing a pavement defect detection model, a pavement defect detection method and a system. Background technique [0002] Common pavement defects include asphalt cracks, cracks, concrete panel cracks, ruts, potholes, subsidence, etc. These defects can easily cause traffic accidents. In order to facilitate the timely processing of pavement defects, it is necessary to detect pavement defects in time. Currently, commonly used pavement defect detection methods include detection methods based on computer vision and methods based on physical property detection. [0003] Detection methods based on computer vision include: image algorithms for pixel feature analysis, such as extracting gradient edge features, analyzing histogram features extracted using HOG operators, etc.; algorithms based on machine learning and feature extraction, through feature ...

AI Technical Summary

Problems solved by technology

[0005] Due to the complex traffic conditions on the road surface, there are many pedestrians and vehicles on the road surface, the lens is easily blocked when collecting road surface images, which will seriously affect the detection accuracy of the analysis method based on computer vision

Moreover, the analysis method based on computer vision and the detection method based on physical properties are easily affected by features such as water stains, stains and wheel marks on the road surface, resulting in misjudgment

Images