Skip to Main Content
Statistical texture features extraction algorithms can be classified into first, second, and higher order. The difference between these classes is that the first-order statistics estimate properties of individual image pixel values, while in the second and higher order statistics estimate properties of two or more image pixel values occurring at specific locations relative to each other. The most popular second order statistical texture features are derived from the co-occurrence matrix, which has been proposed by Haralick. However, the computation of both matrix and extracting texture features are very time consuming. In this paper we improve the performance of those algorithms using FPGA implementations. Our experimental results show that the computational time of co-occurrence matrix is 6.74 times faster than the computational time of extracting thirteen texture features for an image size 128×128 and 8bit gray level. Furthermore whole execution of both algorithms is almost 214x faster than the software implementations.