Study of Texture Images Classification Method Based on Fractal Dimension Calculation

This paper deals with the problem of recognizing textures in images. For this purpose we employ a technique based on the fractal dimension (FD) and a new fractal dimension estimating method is proposed by taking the area instead of the volume covering in box-counting to estimate the FD. Three FD features are based on the original image, the above average/high gray level image, the below average/low gray level image. To classify a scene into the desired number of classes, a support vector machine (SVM) approach is used. Experimental results demonstrate the higher recognition rates obtained using FD Features by the area covering method with orientation than using FD Features by the box-counting method for the texture images.