A method, medium, and system for enhancing infrastructure surface crack features
By combining a line structured light sensor with an inertial navigation device, pixel-level alignment of two-dimensional grayscale images and three-dimensional depth images is achieved, generating crack feature-enhanced images. This solves the problem of low crack detection accuracy in existing technologies and improves the accuracy of crack identification and the reliability of disease classification.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- LAN SHEN (BEI JING) KE JI YOU XIAN GONG SI
- Filing Date
- 2026-02-09
- Publication Date
- 2026-07-21
AI Technical Summary
Existing two-dimensional and three-dimensional image crack detection technologies struggle to achieve pixel-level precise fusion under complex road surface conditions, leading to difficulties in crack feature extraction, high false negative rates, and low recognition accuracy.
By combining a line structured light sensor with an inertial navigation device, precise registration of two-dimensional grayscale images and three-dimensional depth images is achieved through six-degree-of-freedom coordinate transformation correction and sub-pixel-level phase correlation algorithm. Combined with incremental encoder synchronous triggering of scanning and shooting, crack feature enhanced images are generated, and crack area identification and disease classification are performed through multi-scale crack identification model.
It achieves pixel-level alignment between two-dimensional grayscale images and three-dimensional depth images, improving the expressiveness and recognition accuracy of crack features. It can accurately identify cracks under complex road surface conditions and provide richer information on damage characteristics.
Smart Images

Figure CN122089795B_ABST