An adaptive weighted fuzzy mean (AWFM) filter with high stability is proposed for signal restoration in this paper. AWFM is an extension of weighted fuzzy mean (WFM) filter by linking a fuzzy detector and a dynamic selection procedure to WFM in order to overcome the drawback of WFM in fine signal preservation. The fuzzy detector detects the amplitude of impulse noise, which will be used as the argument of the dynamic selection procedure, by referring to two fuzzy intervals and the WFM-filtered outputs. Then, the dynamic selection procedure applies four heuristic decision rules to determine the final filtering output. AWFM not only preserves the high performance of WFM on removing heavy additive impulse noise, but also improves the performance of WFM on light additive impulse noise. Moreover, it results in a high stability on the full range of noise occurrence probability. Compared with the other filters, AWFM exhibits better performance in the criteria of mean absolute error (MAE) and mean square error (MSE). On the subjective evaluation of those filtered images, AWFM also results in a higher quality of global restoration. For the dedicated hardware implementation, the kernel of AWFM filter, WFM, is synthesized with the genetic LR fuzzy cells which realize high-speed fuzzy inference. The hardware complexity is much simpler than the conventional median filter, and simulation result exhibits that up to 6.6 M pixels per second can be filtered by a WFM filter with small chip area.
All Science Journal Classification (ASJC) codes
- Artificial Intelligence