Adaptive thresholds edge detection for defective parts images based on wavelet transform

Image edge detection plays an important role in the system of computer vision. Wavelet is a powerful tool in image processing and has wide application to edge detection for its multiscale characteristic. Based on wavelet modulus maximum edge detection algorithm, an improved method is proposed in this paper, which gives an automatic determination function of eliminating noise threshold by using the clustering technique. Some experiments were made using B-spline wavelet and improved K-means clustering algorithm. The experimental results show that this method is correct and effective to defective parts, and the result was better than that using fixed thresholds.