Computing textural features based on co-occurrence matrix for infrared images

We describe the computing of six textural features based on the co-occurrence matrix for infrared images. With regard to the fact that in the normal mode the warmest objects have the highest values of gray-level infrared images, used textural features should show which of them have influence in highlighting the difference between the warm objects and their background. For that purpose the MATLAB algorithm is used and eighteen infrared images are tested. Due to the computing intensity of the co-occurrence matrix the used infrared images with 256 gray-levels were reduced to 64 gray-levels. The results showed that variance and correlation contributed towards highlighting warm objects or their parts.