The distortion can be formed as the magnifying power of image changes in the field of view in a practical optical system if the field of view is large enough. Optical distortion can be corrected by aberration correction theory, as well as digital image processing technology. The theory of distortion correction by digital image processing technology is discussed, the method of getting reference points by lattice templet is introduced, and the method of generating reference points automatically is proposed. The algorithm based on BP neural network which can approximate any continuous function is used to get coordinate transform function between ideal image and distorted image. The method of bilinear gray-level interpolation is adopted because the calculation of bilinear interpolation is less than others as strict requirements on calculation in engineering application. Finally, a sample of image correction is given, and the results show that it is very fast and convenient to pick up reference point, and the precision of results is high with the method of neural network.

Date Published: 20 May 2009
PDF: 5 pages
Proc. SPIE 7283, 4th International Symposium on Advanced Optical Manufacturing and Testing Technologies: Optical Test and Measurement Technology and Equipment, 72831R (20 May 2009); doi: 10.1117/12.828680

Published in SPIE Proceedings Vol. 7283:
4th International Symposium on Advanced Optical Manufacturing and Testing Technologies: Optical Test and Measurement Technology and Equipment
Yudong Zhang; James C. Wyant; Robert A. Smythe; Hexin Wang, Editor(s)