Texture analysis from irregularly sampled data

This thesis presented a research on texture analysis for feature extraction, classification and segmentation of irregularly sampled images. To produce meaningful results within the field of image analysis two major approaches were revisited: statistical methods and multi frequency filtering techniques. Biologically inspired processing was progressively incorporated into the research to the point of proposing a biological paradigm of image segmentation. We can consider the contribution of this work in regard to two unifying factors through the research, namely, feature extraction and biologically inspired processing.