A novel technology-matching-based method for detecting edges of micro accessories

In order to improve the accuracy of measurement result when using visual inspection method measure the micro parts, the technology matching principle was promoted. It was pointed that optical characteristics of micro image had a relationship with the machining method of the micro parts, the edge detection method should be carried on according to its machining process when detecting the micro parts. A new edge detecting algorithm based on technology matching was proposed. The new method built the polynomial regression mathematics model of transition region image, and the accurate position was defined by mathematical derivation. The processing condition and statistics were added in the algorithm. In this foundation, evaluation model of edge image based on fractal theory in microscope measurement was proposed. Computing the fractal d imension of micro-machining images and analyzing the similarity of same micro -machining model, the new method for edge detection of micro accessory is universally applicable. Experimental results of four common micro-machining models shows the edge detection algorithm based on technology matching principle can accurately extract edge point, and the edge detection result can reach the inferior pixel level.