A fast edge detection model combining mixed L1 and L2 fidelity terms
2017 Joint 17th World Congress of International Fuzzy Systems Association and 9th International Conference on Soft Computing and Intelligent Systems (IFSA-SCIS)(2017)
摘要
Edge detection plays an immensely important role in image processing. In this paper, we propose a new model with the combination of the L
1
and L
2
fidelity terms on the basis of the well-known Mumford-Shah (MS) model. To solve this minimum model, we design an efficient algorithm based on a fixed-point iterative method and the Split-Bregman (SB) method. Experimental results show that the proposed model and algorithm can get better detected edges and have more advantages in efficiency and accuracy for different pure noisy images, even for mixed noisy images.
更多查看译文
关键词
edge detection,fixed-point method,Split-Bregman (SB) method,mixed noisy images
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要