Texture analysis using spatial gray level dependence matrix and the logical operators for Brodatz images

The evaluation of texture features is important for several image processing applications. Texture analysis forms the basis of object recognition and classification in several domains. There is a range of texture extraction methods and their performance evaluation is an important part of understanding the feature extraction tools on the images. in this paper a new algorithm for texture classification based on logical operators and one of the statistical texture methods which is the spatial gray level dependence method. Ten digital Brodatz images used in this search with five samples taken from each image the logical operators have been convolved with texture images and (25) logical operators have been used, (7) textural feature have been obtained from the spatial gray level dependence matrix. These operators have been reduce to (10) logical operators and the features reduced to (3) features depending on the J-parameter, but this is not enough for texture analysis. So the minimum distance has been calculated between the images and the (5) samples belong to them taken into account the logical operators and the features after the minimum distance have been determined the features reduced to (2) and the logical operators to (4). From this result we concluded that the correlation and angular second moment gives the best features and the four logical operators.