Abstract
A new algorithm based on local gradient is proposed to inspect steel rod surface defects. The local gradient method can weaken the effect of noise in steel rod surface image. Then the open operation of morphology theory is employed to remove the retaining noise. Experiments show that this algorithm is effective to inspect steel rod surface defects and is time-saving.
Publisher
Trans Tech Publications, Ltd.
Reference5 articles.
1. S.H. Choi, H.M. Bae, H.W. Hwang, C.H. Park, U.S. Patent 20100246974 A1. (2010).
2. T.S. Chang, D. Gutchess, Huang. U.S. Patent 20080063426 A1. (2008).
3. Y.C. Liu, Y.L. Hsu, Y.N. Sun, S.J. Tsai, C.Y. Ho and C.M. Chen. A Computer Vision System for Automatic Steel Surface Inspection[C] Proc. of 2010 5th IEEE Conference on Industrial Electronics and Applications. Taichung, Taiwan: IEEE, 2010: 1667-1670.
4. J.P. Yun, Y.S. Park, B. Seo and S.W. Kim. Development of Real-time Defect Detection Algorithm for High-speed Steel Bar in Coil(BIC)[C] Proc. of SICE-ICASE International Joint Conference 2006. Bexco, Busan, Korea: IEEE Computer Society, 2006: 2495-2498.
5. G.F. Wu, K. Xu and J.W. Xu. Computer Aplications. Vol. 25(2005), p.326(In Chinese).