Fuzzy Integral Filters: Properties and Parallel Implementation

Fuzzy integrals as image lters provide a standard representational form which generalize linear lters such as the averaging lter, morphological lters such as at dilations and erosions, and order statistic lters such as the median lter. However, fuzzy integral lters are computationally intensive. Computing the output value obtained by fuzzy integral ltering at a point involves sorting all the pixels in a neighborhood of the point according to their values and then computing the ordered weighted sum or maximum with respect to an appropriate fuzzy measure. In this paper, we discuss some properties of fuzzy integral lters and describe a method for enhancing the processing elements of SIMD mesh computers with comparators and counters to eeciently implement fuzzy integral lters.