Combined statistical and model based texture features for improved image classification

TL;DR

This study combines statistical and model-based texture features to significantly enhance image classification accuracy, achieving up to 97.01% accuracy by integrating Gaussian Markov random fields and fractional Brownian motion.

Contribution

It introduces a combined approach using both statistical and model-based texture features, demonstrating improved classification performance over individual methods.

Findings

Combined features achieved up to 97.01% accuracy.

Model-based features outperformed statistical methods alone.

Integration of GMRF with statistical methods improved accuracy to over 96%.

Abstract

This paper aims to improve the accuracy of texture classification based on extracting texture features using five different texture methods and classifying the patterns using a naive Bayesian classifier. Three statistical-based and two model-based methods are used to extract texture features from eight different texture images, then their accuracy is ranked after using each method individually and in pairs. The accuracy improved up to 97.01% when model based -Gaussian Markov random field (GMRF) and fractional Brownian motion (fBm) - were used together for classification as compared to the highest achieved using each of the five different methods alone; and proved to be better in classifying as compared to statistical methods. Also, using GMRF with statistical based methods, such as Gray level co-occurrence (GLCM) and run-length (RLM) matrices, improved the overall accuracy to 96.94% and 96.55%; respectively.