Texture Classification Using Gray Level Cooccurrence Based on Edge Maxima.

摘要

Features based on the joint frequencies of occurrence of pairs of gray levels at given separations are often used for texture analysis. Recently, Davis et al. have suggested using joint occurrences of local maxima of a local property (e.g., edge maxima) to define texture features. This paper proposes a hybrid approach using joint gray level occurrences at and near edge maxima. This paper introduces a new class of texture features, based on the joint occurrences of gray levels at points defined relative to edge maxima. These features are compared with previous types of cooccurrence-based features, and experimental results are presented indicating that the new features should be useful for texture classification. Originator-supplied keywords include: Image processing, and Pattern recognition.