Threshold Image Segmentation Research And Analysis

On image processing has been a hot topic for many years in the computer information technology developing high-speed, image processing has been applied to various fields of society. Image segmentation is for a specific purpose, the adoption of technologies and methods related to image characteristics in accordance with their respective characteristics are divided into regions, the researchers extract the target of interest, thus further study of the target region. Image segmentation is the image processing, pattern recognition and artificial intelligence and other fields of the key issues to deal with computer vision technology is the most important and critical step. Application of image segmentation is very extensive, almost there on all areas of image processing. Therefore, from the 20th century, since the 60's, image segmentation has always been a hot research techniques.The paper first outlines the background of image segmentation tasks, the significance of the subject content and objectives of the study, pointed out that the image segmentation in the image of the key projects, image segmentation technology research, analysis image segmentation common technology, pointed out that from the theory, applications and application areas such as image segmentation to analyze the technology, for improving the image analysis and image processing technology are of great significance.Subsequently, studies were based on the bimodal histogram threshold method (mode method), the method is a diagram for bimodal histogram for segmentation of images, through in-depth analysis, analysis of the advantages and disadvantages of the method. Bimodal histogram method and improved, by improving the achievement of the threshold is automatically selected, and then, in summing up the results of previous studies on the basis of the focus on genetic algorithm based on Otsu segmentation algorithm, by contrast, can be seen Genetic Algorithm Otsu segmentation image segmentation algorithm is more obvious that a higher degree of automation to adapt to a wider range, and no prior measurement of gray-scale image histogram, more flexible, more precise. Subsequently, the operator by detecting genetic algorithm based on Otsu segmentation algorithm to improve and enhance the genetic algorithm based on Otsu segmentation algorithm to deal with the effects of picture.Finally, the image threshold segmentation research results are summarized and pointed out that the direction of future efforts. Through the thesis research, are familiar with image threshold segmentation of the basic methods and theory of knowledge, understanding of the field of new technologies and methods to improve their own teaching and research capacity.