This paper aims to improve the accuracy of texture classification based onextracting texture features using five different texture methods andclassifying the patterns using a naive Bayesian classifier. Threestatistical-based and two model-based methods are used to extract texturefeatures from eight different texture images, then their accuracy is rankedafter using each method individually and in pairs. The accuracy improved up to97.01% when model based -Gaussian Markov random field (GMRF) and fractionalBrownian motion (fBm) - were used together for classification as compared tothe highest achieved using each of the five different methods alone; and provedto be better in classifying as compared to statistical methods. Also, usingGMRF with statistical based methods, such as Gray level co-occurrence (GLCM)and run-length (RLM) matrices, improved the overall accuracy to 96.94% and96.55%; respectively.
展开▼
