A comparative study on the K-views classifier and Markov random fields for image texture classification

We compared two image classifiers which incorporate contextual information to classify each pixel in the raw images in this study: namely, the K-views classifier and the classifier using Markov Random Fields (MRF). These procedures incorporate contextual information by using spatial features. Preliminary experimental results are provided in this report.

[1]  Rama Chellappa,et al.  Multiresolution Gauss-Markov random field models for texture segmentation , 1997, IEEE Trans. Image Process..

[2]  Chih-Cheng Hung,et al.  Experiments on image texture classification with K-views classifier, Markov random fields and cooccurrence probabilities , 2004, IGARSS 2004. 2004 IEEE International Geoscience and Remote Sensing Symposium.

[3]  Chih-Cheng Hung,et al.  Image texture classification using datagrams and characteristic views , 2003, SAC '03.

[4]  John W. Woods,et al.  Two-dimensional discrete Markovian fields , 1972, IEEE Trans. Inf. Theory.

[5]  Robert M. Haralick,et al.  Textural Features for Image Classification , 1973, IEEE Trans. Syst. Man Cybern..