An Edge Detection Algorithm Based on Edge-Preserving

Abstract:

The traditional first-order differential operator is under the influence of the Gaussian noise, therefore, it often implement boundary extraction after average filtering. But the filtering process would often smooth the details of some directions of image too much, so that the edge cannot be extracted correctly. To solve this problem, this paper puts forward the edge detection algorithm based on edges keep, to determine the keeping direction of the edge through matching different directions’ edge template. Instead of average filtering process, it can improve the performance of traditional operator, and provide the simulation results. Experimental results show that the algorithm can eliminate noise, and at the same time, keep more edge information of the image.

• Periodical: Key Engineering Materials (Volume 693)
• Pages: 1321-1325
• DOI: https://doi.org/10.4028/www.scientific.net/KEM.693.1321
• Online since: May 2016
• Authors: C.R. Tang*, A. Li
• Keywords: Edge Detection, Edge-Preserving, Noise

References

[1] Liu Ming-Yan, Zhao Jing-Xiu, Sun Ning. Opto Electronic Technology, 26, 4(2009).

[2] Wu Jie, Chen De-Zhi, Guo Cheng-Zhi. LASER & INFRARED, 38, 6(2008).

[3] Canny J. IEEE Transactions on Pattern Analysis and Machine Intelligence 8, 6(1998).

[4] Wen Shan, Li Bao-Qing,. Acta Automatic Sinica, 33, 5(2007).

[5] Huang Y R, Duan H, In Proceedings of the SPIE on the International Society for Op-tical Engineering, (2007).

[6] Wang Bao-Ping, Liu Sheng-Hu, Zhang Jia-Tian, Zhang Yan-Ning, Fan Jiu-lun, Chinese Journal Of Computers, 29, 4(2006).

[7] Faradji F, Rezaie A H, Ziaratban M, Proceedings of the 14th IEEE In-ternational Conference on Image Processing, (2007).

[8] Liu Wang-Jun, Jiang Qing-Ling, Zhang chuan, ACTA AUTOMATICA SINICA, 35, 12(2009).