Moment Description Of The Visual Characteristics Of The Target Image Segmentation Techniques, And Testing

Image segmentation is one important part in the image processing .It is widely used in digital video processing, pattern recognition, target recognition and image tracking. In addition, it is also used in video compressing based on content ,which is the main idea of advanced international standards, such as MPEG-4/7. It is vital to advance the efficiency in the multimedia communication.This dissertation discusses image segmentation technology based on the image visual features, the image neighborhood moment analysis and the related image detection and segmentation system. And the key functional modules'analyzed are as follows: the image neighborhood moment's analysis, the adaptive clustering after the visual moment transformation, the boundary detection of the object image and the extraction of the object image area of similar visual brightness. The background and research task of the dissertation and the development of the related techniques in domestic and abroad are introduced in the first chapter, and in the second one, the general theory and technologies of image segmentation are presented, including their disadvantage. In the third, focuses are put on the analysis for the human visual features and the mechanism of the human visual recognition of the image object, the description of the image visual features with image neighborhood moments and the theory and the model for the segmentation based on the visual feature are presented. Finally the system design based on visual features recognition is developed. In the fourth chapter the analysis for the visual features, the adaptive clustering and the extraction of the image area with the image similar visual brightness are analyzed. The analysis for the visual features of color image and the transformation between the color visual model and the industry color model, the system design for color image segmentation based on the human vision are presented in the fifth chapter. In the last chapter are discussed the design of the software system, and its flow chat and the data resulted from the PC simulation.