Int J Performability Eng ›› 2019, Vol. 15 ›› Issue (5): 1472-1481.doi: 10.23940/ijpe.19.05.p24.14721481

Previous Articles     Next Articles

Edge Detection Method based on Lifting B-Spline Dyadic Wavelet

Zhibin Hu, Caixia Deng*, Yunhong Shao, and Cui Wang   

  1. Department of Mathematics, Harbin University of Science and Technology, Harbin, 150080, China
  • Submitted on ;
  • Contact: * E-mail address: dengcx2001@hotmail.com
  • Supported by:
    This work is supported by the National Natural Science Foundation of China (No. 51875142, 11871181).

Abstract: Aiming at the problem of discontinuous edge details in the wavelet transform modulus maxima edge detection algorithm, the shortcomings of the algorithm are addressed by improving the smoothness of the wavelet function and selecting appropriate wavelet filters. The order of vanishing moments of a wavelet function determines the ability of the wavelet to approximate smooth function. Therefore, this paper focuses on improving the order of vanishing moments of the B-spline dyadic wavelet, giving a new lifting scheme and lifting parameters, and realizing a B-spline dyadic wavelet filter with high-order vanishing moments, symmetry, and compact support. At the same time, an edge detection algorithm based on lifting the B-spline dyadic wavelet is proposed. The experimental results show that the algorithm can effectively suppress noise and display the continuous details of the image edges.

Key words: dyadic lifting scheme, high order vanishing moments, histogram equalization, edge detection