Affiliation:
1. Xi'an University of Architecture and Technology
Abstract
Defect detection is necessary before the use of industrial castings. DR (Digital Radiograph) system has a broad application prospect because of high detection efficiency. The CV model is an image segmentation method with high accuracy, but its segmentation result is not perfect in the case of strong edge. As a common image segmentation method, region growing has simple calculation as a result of using the space character of the image. In this paper, the defect detection about DR synthetic color image of railway truck side frame was conducted by CV model and the region growing methods, and segmentation achieved good results.
Publisher
Trans Tech Publications, Ltd.
Reference9 articles.
1. SHEN Gong-Tian, LI Li-Fei. Progress of Non-Destructive Testing Technique on Cast Iron [J]. Non-Destructive Testing, 2011, 33(1): 62-68.
2. Bueno C, Barker M D. High-resolution digital radiography and three-dimentional computed tomography[C]. X-Ray Detector Physics Applications, 2009: 179-191.
3. CHAN TF, VESE LA. Active contours without edges [J]. IEEE Trans Image Processing, 2001, 10(2): 266-277.
4. MUMFORD D, SHAN J. Optimal Approximation by Piecwise Smooeh Functions and Associated Variational Problems [J]. Communications on Pure and Applied Mathematics, 1989, 42(4): 577-685.
5. CHAN TF, SANDBERG BY, VESE LA. Active Contours Without Edges for Vector-valued Images [J]. Journal of Visual Communication and Image Representation, 2000, 2(11): 130-141.