Remote Sensing Image Object Extraction By Mathematical Morphology Research

In recent years, with the rapid development of computer, remote sensing and image processing techniques, extracting special targets from remote sensing images and getting military intelligence becomes an important part of the modern information war, and it also leads to various new methods of targets extraction.Mathematical morphology is a tool of the signal analysis in terms of shape, and it can carry on many kinds of processing by changing the shape of objects within the signal. Based on the theories and methods of mathematical morphology, in this paper it is focused on automatic military targets extraction in the remote sensing images, especially planes. The main work is presented as following:(1) The development, actuality and basic theories of mathematical morphology are summarized.(2) Various algorithms of mathematical morphology are researched, they include hit or miss Transformation, edge detection and morphological grads, etc.(3) The combination of traditional algorithms and morphological algorithms are used to preprocess the target image. Different target information is obtained by making use of different abutment pixel templates, and the extraction results are compared and analyzed.The experimental results show that it is very effective to apply the methods of mathematical morphology to special target extraction, and it will help to enhance the efficiency of work in getting military intelligence from the remote sensing image.