ABSTRACT
This chapter deals with the analytic methods that determine local properties of the image or spatially limited relations between these local properties, as used in characterizing textures. It explores the strongly nonlinear generalized morphological operators that belong to both image processing and analysis; however, they may significantly contribute to image examination and recognition and are thus usually considered analytic methods. Local features are the simplest description of image properties. The larger the local analytic areas, the coarser the spatial resolution of the parametric image because the uncertainty as to where exactly the derived parameter belongs is higher. Locally applied unitary transforms may also give relevant information on the local character of the image. In industrial applications, an important role belongs to corner detectors that enable the finding of image features in man-made scenes where rectangular and similar objects are common. Textures are frequent in natural images, including medical; consequently, they are well perceived and recognized by human observers.
