A high light surface defect detection method based on light field correlation fusion

By calculating the temporal and frequency domain features of the light field video stream and performing weighted fusion, the efficiency and accuracy issues of high-brightness object surface detection are solved, and efficient and stable defect detection is achieved.

CN117392453BActive Publication Date: 2026-06-30TSINGHUA SHENZHEN INTERNATIONAL GRADUATE SCHOOL
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
TSINGHUA SHENZHEN INTERNATIONAL GRADUATE SCHOOL
Filing Date
2023-10-26
Publication Date
2026-06-30

Smart Images

  • Figure CN117392453B_ABST
    Figure CN117392453B_ABST
Patent Text Reader

Abstract

This invention discloses a method for detecting hyperbola surface defects based on optical field correlation fusion, comprising the following steps: S1, calculating temporal and frequency domain features of the real-time structured light field video stream captured by the structured light field defect detection system, and weighting and fusing the temporal and frequency domain features; S2, regressing the distribution of the fused feature values ​​on adjacent microlenses to predict the defect center and defect size; S3, generating a detection box and eliminating duplicate detections to obtain the defect detection result.
Need to check novelty before this filing date? Find Prior Art