Method and system for detecting road cracks based on unmanned aerial vehicle image

By fusing visible light and thermal infrared images and utilizing histogram analysis and Gaussian thermal diffusion rules, the problem of background noise interference in road crack detection was solved, achieving high-precision crack identification and damage assessment, and improving the accuracy and reliability of the detection results.

CN121982596BActive Publication Date: 2026-05-29GUIZHOU JIAOJIAN INFORMATION TECH CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN Β· China
Patent Type
Patents(China)
Current Assignee / Owner
GUIZHOU JIAOJIAN INFORMATION TECH CO LTD
Filing Date
2026-04-09
Publication Date
2026-05-29

AI Technical Summary

Technical Problem

Existing methods for detecting road surface cracks based on UAV imagery are susceptible to interference from complex background noise, leading to segmentation errors and failing to effectively quantify and assess the damage status of the internal structure of the cracks.

Method used

A dual-modal sensor is used to acquire visible light and thermal infrared images. Histogram analysis is used to locate abnormal regions in the thermal infrared images, seed points are extracted and mapped to the visible light images, and Gaussian thermal diffusion rules and image structure similarity analysis are combined to accurately detect and correct crack regions. The temperature distribution of the crack skeleton is quantified to assess the damage state.

Benefits of technology

It effectively suppresses interference from complex background noise, improves the accuracy of pavement crack detection, and realizes quantitative assessment from two-dimensional appearance morphology to internal damage state, providing a reliable basis for road maintenance decisions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121982596B_ABST
    Figure CN121982596B_ABST
Patent Text Reader

Abstract

The present application relates to the technical field of image processing, and particularly relates to a road surface crack detection method and system based on unmanned aerial vehicle images, the method comprising: acquiring visible light images and thermal infrared images synchronously collected by an unmanned aerial vehicle, locating original road surface thermal anomaly regions and seed points in the thermal infrared images, mapping to the visible light images to obtain mapped road surface thermal anomaly regions, and performing region growing to obtain first crack regions; performing Gaussian thermal diffusion and heat source superposition analysis on pixel points in the first crack regions to generate crack thermal influence maps, performing image structure similarity analysis on the crack thermal influence maps and the original road surface thermal anomaly regions, and correcting the first crack regions to second crack regions; extracting road surface crack skeletons based on the second crack regions and mapping back to the original road surface thermal anomaly regions, evaluating damage states of target road surfaces according to temperature distributions of the road surface crack skeletons, and completing crack detection. The method improves the accuracy of road surface crack detection results.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Airport pavement crack detection method based on infrared and visible light image data fusion

    CN110246130A

  • Crack detection method and system for expressway construction

    CN120404748A