Abstract:

An improved method of basic image features (BIF) histogram is proposed. Firstly, corresponding BIF of every pixel in the image is respectively computed in four scales. Secondly, six histogram features and three high order statistical features are extracted respectively in every scale. Finally, the support vector machine (SVM) is applied for training and classifying the texture images. Experiment results show that the proposed method can reduce the computing complexity and consuming time, meanwhile, it is robust to noise.