Method and system for detecting surface defects of wood based on image generation
By constructing a reference texture representation and reorganizing the pixel arrangement relationship, the problem of lacking a stable reference standard in wood surface detection is solved, thereby improving the accuracy and reliability of wood surface defect detection.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ZHONGSHAN HUASHENG FURNITURE MFG CO LTD
- Filing Date
- 2026-06-08
- Publication Date
- 2026-07-17
AI Technical Summary
Existing wood surface inspection methods lack stable reference benchmarks when faced with the interplay of natural texture gradients and local structural disturbances, leading to breaks or spread in defect boundary identification, which affects the consistency of inspection results and the reliability of sorting and positioning.
By constructing a reference texture representation corresponding to the original image, reorganizing the pixel arrangement relationship, and combining the correlation between the direction of brightness change and the direction of color change, the offset change is expressed in partitions to form an abnormal region representation with structural orientation. The defect contour range is defined by using the turning trajectory and the region closure relationship.
It improves the consistency of discrimination between different regions, enhances the stability of boundary expression, and improves the accuracy and reliability of wood surface defect detection.
Smart Images
Figure CN122409667A_ABST