High precision Fast Line Detection of alignment and coupling for planar Optical Waveguide device

摘要

Machine vision is widely used in alignment between planar optical waveguide chip and optical fiber array. Optical component edge detection is one of the key steps of machine vision in alignment. This paper has proposed a line detection algorithm based on the progressive probabilistic Hough transform (PPHT) and iteratively reweighted least squares (IRLS) algorithm for alignment between planar optical waveguide chip and optical fiber array. The experiment results show that the detection angle error is less than 0.005 degrees and the time consumption is less than 0.5 s through the proposed algorithm. Besides, it also can accurately fit optical component edge with some non-random factors. Therefore, the proposed new algorithm has the advantages of high precision, fast computing speed and good robustness, and it can successfully realize the high-precision fast line detection of optical component edge. (C) 2017 Elsevier GmbH. All rights reserved.