IMAGE TRANSFORMATION USING FUZZYBASED FILTERING FOR THE TEXTURE CLASS’S MEASUREMENT BASED ON THE DISTANCE OF THE FEATURES

A powerful feature extraction method has been developed widely in the field of image texture research. The uniform or different distribution of spatial intensity in an image area determines the number of texture classes formed. Therefore, to obtain quality features is not easy. In this study, the practical approach used to get texture classes is by applying the concept of image transformation where fuzzy logic is as an image filter. The gray level dynamic range is one of the factors that enlarge the texture class; therefore, the image needs to be converted into a new space based on the distance of the features. The image feature used is the distance and the correlation coefficient between the pixels of the image. Furthermore, it produces feature descriptions calculated as the distance between the texture class image and the original image. The results of this study indicate that the dominant background effect on the number of texture classes. So, the image with a closer visualization will represent significant information and vice versa. The results of this study can be used as a reference to determine how many texture classes of each original image