In this paper a new saving-deletion rule is applicable to process any orientation of the fringe quickly and precisely. The purpose of the fringe thinning method is to obtain the accurate location of the fringe. This means the width of the fringes in the image after processing is only one pixel. So despite advanced optical methods or filters which can obtain sharp images, the thinning procedure is necessary for a CCD auto-detecting system. The computer will automatically obtain the data of the processing result and does not depend on input data from measurements.

We mainly utilize an image processing technique{1-2} for dealing with the fringe pattern which the TV camera records, so that we can obtain the information that we want. We use a CCD camera to scan the fringe pattern, and this is connected to a frame grabber. The frame grabber is a plug-in card for an IBM PCXT or AT, and the resolution is 512 X 512 X 8 bits. For the image processing procedure, the first step is preprocessing, and this step consists of noise-suppression and thresholding, the noise-suppression is to filter the noise of the digitized image, and to enhance the connectivity of the broken segments, and the purpose of thresholds is to obtain the binary image and to classify the background and object. The second step is thinning, and in this step we use a new thinning method to extract the skeleton of the fringe, and the thinned pattern preserves the connectedness and the shape of the original pattern. The detailed steps of these procedures are explained as the follows ( Fig.1 ).

nextindexback