A DATA-DEPENDENT FUZZY CENTER WEIGHTED MEDIAN FILTER

The authors have previously proposed a fuzzy stack filter that is defi ned by a fuzzy Boolean function. We have also presented a specific des ign method (called the fuzzy Boolean function determination method) fo r the most fundamental fuzzy stack filter, the FCWM (fuzzy center weig hted median). By varying the fuzzy Boolean function, we can achieve bo th an arbitrary CWM filter and a weighted average value filter in whic h only the weights at arbitrary processing points are varied. Hence, i n comparison with the CWM filter, our filter is better at eliminating Gaussian noise. In this paper, to improve the reconstruction accuracy of nonsteady state signals, such as the image signals, a data-dependen t FCWM (DD-FCWM) is proposed anew. The data-dependent filter is an ada ptive filter in which the filter parameters are varied in response to the nature of the data near the processing point (local information). In a DD-FCWM filter, the fuzzy Boolean function is varied by using loc al information. We demonstrate that the proposed DD-FCWM filter uses t he local signal information to appropriately vary, not only the weight s of the processing points, but also the filter mode (average value ty pe or median type). Furthermore, by use of image processing examples, we demonstrate that highly accurate signal reconstruction can be attai ned. (C) 1998 Scripta Technica.