Digital Images Of Mathematical Morphology Algorithm Research And Applications

In recent years, image processing of morphology as a special image processing subject has already developed into a main research field of image processing. The application of morphology has covered nearly field of image processing, including characters recognition, medical image processing , digital image compression and coding , vision measuring , material science and robot vision ,etc. Some image analysis systems adopt operators of mathematics morphology as the basic system operators, The morphology method has already become engineers' and technicians' indispensable tool rapidly.Mathematics morphology is useful in nearly all respects of image processing, such as object recognition based on the hit-or-miss transformation , the segmentation of image based on current, the image skelecton and the image compress based on erosion and opening operator, the object classification based on morphologic filter, etc.. Mathematics morphology has not only a stability theoretical foundation and succinct , simple , unified thought, but also great application value. Mathematics morphology is rigorous in theory and simple , beautiful in basic idea.This paper will summarize and analyze the mathematics morphology algorithms and their applications, and then improve some algorithms. It is the master work:1, Study the basic theory and the applications of mathematics morphology in depth. Realize two basic operators of binary image, and introduce an algorithm for Image morphological processing based on breadth. While scanning the image, we get the pixels in boundary and save them. We do not process all the pixels but the boundary pixels which are unprocessed and their neighborhood, and it observably improves on memory efficiency and decreasing computation.2, Improve a bi-value image thinning algorithm. The traditional algorithm is time intensive on computation and converges slowly. It can only use one pair of structures element at one time and strips outer pixels in one direction. So we add a pair of diagonal structures to thinning operation, and improve the thinning speed greatly.3, Propose a improved algorithm of area morphology operator ( AOC ). It can maintain the same image result but shortens running time greatly.Traditional AOC operator deals with each layer of the image, there is many unnecessary operations on those layers which have no pixel. Considering the relativity of layers, we propose an improved AOC algorithm based on the relativity of neighbor layers...