A multi texture feature subset-based approach for Image classification

Abstract

In artificial intelligence, pattern recognition of images is important aspect of computer visioning. The main objective is to identify the texture patterns based on the availability of features or the information. In this article multiple feature subset based on image classification which further depends on GLCM, Contrast visibility, Spatial Frequency, Energy of Gradients and histogram bins of binary patterns. The proposed method also presents a fusion based decision criterion for the texture classification. The experiments were conducted on real time datasets and the performance is evaluated with single and multiple feature subsets with SVM classifier. The identification parts will be tough due to small joints and occlusions. Identification of Image features is described in this paper with the help of Box Model Based Estimation, Speed up robust features and finally with Optical flow tracking algorithm.