Research Article

Inspection and Recognition of Generalized Surface Defect for Precise Optical Elements

Hongyu Chu, Zhijiang Xie, Xu Xu, Lidan Zhou and Qin Liu

According to the application of the current high-power solid-state laser devices with precision optical elements, it proposes the concept of the Generalized Surface Defects. Based on the machine vision, a non-contact detection scheme is proposed for the surface detection on defect features concerning the past low speed, smaller size in detection and less inconvenience in the testing or operation. A gantry platform for large-size three-dimensional motion and a clamping frame are designed, obtaining images with the linear array CCD, constructing partially closed spaces by the Graham method, achieving defective images through the adaptive threshold segmentation, extracting the defect parameters from the geometric space and the gray space and classifying the defects by the hierarchical support vector machine. Experiments show that the instrument detects the largest size of 700x600 mm and is capable of completing the scanning of the whole size of such elements within 25 min. Average error of area is 12% and average recognition accuracy is 92.2%.

PDF Fulltext XML References Citation

How to cite this article

Hongyu Chu, Zhijiang Xie, Xu Xu, Lidan Zhou and Qin Liu, 2011. Inspection and Recognition of Generalized Surface Defect for Precise Optical Elements. Information Technology Journal, 10: 1395-1401.

DOI: 10.3923/itj.2011.1395.1401

URL: https://scialert.net/abstract/?doi=itj.2011.1395.1401

Information Technology Journal

Article Trend

Total views 540

Search

Authors

Hongyu Chu

Zhijiang Xie

Xu Xu

Lidan Zhou

Qin Liu

Keywords

Research Article

Inspection and Recognition of Generalized Surface Defect for Precise Optical Elements

How to cite this article

Related Articles

Leave a Comment