DIGITAL IMAGE-PROCESSING OF FRINGE PATTERNS IN PHOTOMECHANICS

A new methodology, developed for thinning fringes in photomechanics, splits fringe thinning into two distinct steps, namely, fringe edge detection and fringe skeletonization. A minimum intensity criterion is utilized to extract the fringe skeletons from broad bands. The algorithm is applicable to any orientations of the fringe. It is noniterative and requires less computations than the other reported methods. A novel scheme is developed to remove discontinuities and noise in the fringe skeleton. The various steps of the algorithm are explained by processing a test image free of noise. The applicability of the algorithm to actual experimental situations is verified by processing fringes obtained by photoelasticity and moire. Preprocessing of the fringes is discussed in detail. A method is also developed for determining the coordinates of any point chosen on the fringe contour. This step involves the use of the system characteristics of the PC-based image-processing system. Fringe ordering is kept interactive to utilize the experimentalist's expertise. A new approach to automating fringe ordering is proposed.