Texture image description based on data compression
Nuo Zhang, Toshinori Watanabe · 2011
Texture analysis is important in many application fields in image processing. There are four domains, texture classification, texture segmentation, texture synthesis and shape from texture, in texture analysis. Generally, there are two phrases in texture classification process: the learning phase and the recognition phase. In this study, we introduce an approach for texture classification. Our approach is based on the consideration of searching the essential feature of frequent pattern in texture images, and the learning phase is not necessary. To find out the frequent pattern in a texture image, data compression is used in our approach. Data compression helps us to extract the longest and frequent features, without complicated computation, in out approach. The simulation results will show good performance of our approach.