The Effects of Preprocessing on Colorectal Polyp Detecting by Fuzzy Algorithm

In the following study the effects of two image preprocessing methods, namely Gaussian filtering and Wiener filtering, is studied on the results of a fuzzy inference method previously developed by the authors, for determining whether a colonoscopy picture segment contains any colorectal polyp. As earlier results show that less blurry, less compressed and less noisy images tend to be better classifiable, the effects of noise suppression with a Gaussian filter, which makes the images also blurrier, was beneficial on noisy, compressed images, and rather maleficent in good quality pictures. The effects of the Wiener filter, which both decreases noise and enhances edges, did not really manifest in classification improvement.