A method for detecting subsidence using simple linear iterative clustering
By using a simple linear iterative clustering method to segment images from laser scanning data of the road surface, the subjectivity and accuracy issues of subsidence detection in existing technologies are resolved. This enables rapid and accurate identification and depth calculation of subsidence areas, and is highly adaptable to different road surface features and instrument conditions.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- WEI LE TECHNOLOGY GROUP CO LTD
- Filing Date
- 2026-01-09
- Publication Date
- 2026-05-26
AI Technical Summary
Existing technologies for detecting asphalt pavement subsidence suffer from problems such as high subjectivity, poor accuracy, low efficiency, lack of comprehensiveness, and difficulty in data quantification, resulting in the inability to accurately identify subsidence areas and depths, thus affecting road safety and lifespan.
A simple linear iterative clustering method is used to segment the laser scanning data of the road surface. The category of the pixel is determined by the comprehensive distance of gray value and spatial distance, which can quickly identify the subsidence area and calculate its relative height difference to eliminate the influence of slope and determine the subsidence depth.
It enables rapid and accurate identification and depth calculation of subsidence areas, improving the flexibility and accuracy of detection, adapting to different road surface features and changes in instrument status, and eliminating the need for a large amount of labeled data and predefined templates.
Smart Images

Figure CN122089649A_ABST