Research On Image Edge Detection Based On Mathematical Morphology

Edge detection Plays an important role in the digital image Processing. Based on conventional edge operators, such as Robert, Sobel, Laplace, the results of processing are somewhat missing. Furthermore they are sensitive to noise.Mathematical morphology, a methodology of nonlinear filters, has some characteristics such as simplifying image data, maintaining the basic shape of the image characteristics. In a word, the study object of mathematical morphology is morphological character of image. So it has used widely in many fields of image processing.The following research is mainly done, on the basis of studying the method of edge detection using mathematical morphology.(1) Existing methods of edge detection are summarized firstly.(2) The basic mathematical morphology theories - binary and grayscale mathematical morphology are presented. Based on the theories, the results of image processing are analyzed.(3) Using mathematical morphology methods, the methods of extracting edge from the close-range images of building and remote sensing images are mainly introduced. The methods of processing the close-range images of building and remote sensing images based on mathematical morphology theory are analyzed in feasibility and demonstrated in experiment.Based on the study above, the following conclusions are mainly summarized.(1) The mathematical morphology methods have simple basic thought, fast operation and good analytical capability. So the mathematical morphology in the image processing has some advantages.(2) Experimental results show that the method of extracting edge step by step, that is denoising by using morphology filter operators firstly and then detecting edge by using morphology gradient operators, obtains better extracting effect.The detected edge information of building can be used in deformation monitoring of building. The edge information can be used in object recognition and classification of remote sensing image with its spectrums, texture features. And the detected information can be used in the fields of GIS, agriculture and forest, meteorology, ocean, land resources and so on.