Texture classification is a fundamental task in computer vision and image analysis, widely used in applications such as medical imaging, material inspection, remote sensing, and object recognition. Among various texture descriptors, Local Binary Pattern (LBP) has gained significant attention due to its simplicity and computational efficiency. However, conventional LBP methods often ignore certain binary patterns with multiple transitions, which may lead to loss of important texture information. To overcome this limitation, this paper presents a machine learning based texture classification approach based on Sorted Consecutive Local Binary Pattern (scLBP). The proposed method extracts local texture features by analyzing binary relationships between neighboring pixels and sorting consecutive patterns to preserve detailed structural information. The extracted scLBP feature vectors are then used as input to a machine learning classifier for texture recognition. Experimental analysis shows that the proposed approach provides improved feature representation and achieves better classification performance compared with traditional LBP-based techniques. The results demonstrate that the scLBP-based framework can effectively enhance texture classification accuracy and robustness.
SriKrishna et al. (Tue,) studied this question.