Improved Median Filtering Algorithm Based on FPGA

Abstract:

Aiming at the loss of image boundary information and the blurred image details in the classical median filtering algorithm, the improved median filtering algorithm based on FPGA (Field Programmable Gate Array) is proposed. In the improved median filtering algorithm, the borders mirroring and threshold comparing are adopted, and the loss of the image boundary information is lowed and the detailed information of image is preserved. The improved median filtering algorithm based on FPGA consists of the filter template generating module, the median grayscale computing module and threshold comparing module. The experimental results show that that the NMSE (normalized mean square errors) of the image filtered with the improved median filtering algorithm is lowed by 25.6%, and the operating efficiency is improved by 16% comparing with the classical median filtering algorithm.

• Periodical: Applied Mechanics and Materials (Volumes 644-650)
• Pages: 3759-3762
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.644-650.3759
• Online since: September 2014
• Authors: Chang Bao Wen*, Cheng Fei Xue, Tiao Yang, Yan Ming Li
• Keywords: Borders Mirroring, FPGA (Field Programmable Gate Array), Threshold Comparing

References

[1] Ko Sung-Jea, Lee Yong Hoon: IEEE transactions on circuits and systems. 38 (1991) 984-993.

[2] Lin Sun, Yanxia Wu, Bowei Zhang, Guochang Gu: Journal of Convergence Information Technology. 7 (2012) 298-304.

[3] Benkrid K, Crookes D, Benkrid A: Proceeding-IEEE International Symposium on Circuits and Systems. 4 (2002) 425- 428.

[4] Zhonghu Yuan, Wentao Liu, Xiaowei Han: Applied Mechanic and Materials. 401-403 (2013) 1507-1513.