Study of Texture Images Extraction Based on Gray Level Co-Occurence Matrix

Abstract

The non-spectral characteristics play an important role in assisting image classification in remote sensing, as these characteristics can avoid mistake classification maken by the same object with different spectrum phenomenons. The texture ch aracteristic is one kind of non-spectral characteristics. It is also helpful to improve classification precision. In this paper, co-occurrence matrix applied to extract texture images based on pixel level is discussed and experiments are carried out to analyze this method.