Color Texture Analysis and Classification: An Agent Approach Based on Partially Self-avoiding Deterministic Walks

Abstract

Abstract. Recently, we have proposed a novel approach of texture analysis that has overcome most of the state-of-art methods. This method considers independent walkers, with a given memory, leaving from each pixel of an image. Each walker moves to one of its neighboring pixels according to the difference of intensity between these pixels, avoiding returning to recent visited pixels. Each generated trajectory, after a transient time, ends in a cycle of pixels (attractor) from where the walker cannot escape. The transient time (t) and cycle period (p) form a joint probability distribution, which contains image pixel organization characteristics. Here, we have generalized the texture based on the deterministic partially self avoiding walk to analyze and classify colored textures. The proposed method is confronted with other methods, and we show that it overcomes them in color texture classification.

Improved texture image classification through the use of a corrosion-inspired cellular automaton

“…Feature vectors for image texture are usually obtained from the analysis of groups of pixels and the way this analysis is performed is used to classify the different texture analysis methods. Five main categories can be distinguished: structural [1][2][3], statistical [4], model-based [5][6][7], spectral [8,9], and agent-based methods [10][11][12].…”