Distortion correction in surface defects evaluating system of large fine optics

• Published in: Optics communications Vol. 312; pp. 110 - 116
• Main Authors: Wang, Shitong, Liu, Dong, Yang, Yongying, Chen, Xiaoyu, Cao, Pin, Li, Lu, Yan, Lu, Cheng, Zhongtao, Shen, Yibing
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01-02-2014
• Subjects: Apertures; Calibration; Computer vision; Defects; Distortion; Fittings; Image processing; Imaging; Optical inspection; Stitching; Surface defects; Computer vision; Calibration; Image processing; Optical inspection
• ISSN: 0030-4018; 1873-0310
• DOI: 10.1016/j.optcom.2013.08.083

It is challenging to quantitate micron-scale defects on the surface of large optics. Based on microscopic scattering dark-field imaging, sub-aperture scanning and stitching, the Surface Defects Evaluating System (SDES) can spot and measure scratch and digs on large optics. In order to evaluate defects of down to sub-micron over apertures of hundreds of millimeters, high magnification zoom microscope and large format CCD camera are employed to balance between the efficiency and resolution. Unfortunately, large optical distortion in the sub-aperture images can be found, and feature recognition of defect images as well as mismatches of the sub-apertures stitching are greatly affected. In this paper, a distortion correction procedure based on the Brown–Conrady model is proposed. Grayscale surface fitting and bilinear interpolation techniques are employed to obtain better correction accuracy. A grids-patterned fused silica standard plate has been specifically made to calibrate the distortion. Comparison experiments indicate that the distortion in the SDES can be corrected by the proposed method very well.