Research On Video Motion Object Segmentation For Content-based Application

With the development of the modern information society, more and more multimedia information is available. So the technology of multimedia processing is becoming the important task for the irrelevant area of scientist. Among of the multimedia, the visual information is more attractive due to its direct, vivid characteristic, but at the same time the huge amount of video data causes many challenges if the video storage, processing and transmission.Video compression is necessary for efficient storage, processing and transmission. Motion picture experts group (MPEG) published the secondary generation video coding standard, MPEG-4. Compared with the first generation standard, a significant character of MPEG-4 is object-based coding which means to code videos as a set of semantic video objects. Segmentation is necessary to get each video object. MPEG-7, which is based on video database querying, also defines a universal and normalized description of various multimedia objects, and the computer vision needs function based on object. Therefore the method of description based on object is becoming a huge requirement of multimedia information processing.Video object segmentation belongs to the most difficult and important problem. Though MPEG-4 introduces the concept of video object, it does not specify any concrete techniques for obtaining video objects from video sequence. The difficulty of video object segmentation lies on two aspects. One is the extreme complexity of video scenes, which means no uniform model for all video objects. Another one is the definition and description of semantic video object. Video object segmentation is carried out on low vision level. The semantic homogeneity of video object is hard to be modeled by any low level features, such as edge, texture, color and motion.Therefore, the dissertation focuses on the methodology and techniques for video object segmentation under the framework of MPEG-4. The main objective is the objects segmentation with the various of complexity motions, which not only include the single object segmentation, but also multiple objects. The key techniques were the emphases such as occlusion, objects combining and separating, and spatio-temporal segmentation. Major works of this dissertation are as follows:To solve the problem of inaccurate segmentation, an automatic video segmentation is presented. The algorithm is designed based on regions classification. Regions partition is simple and fast in respect to others methods, which just meets the requirements for automatic video object segmentation. To overcome the errors of regions classification between moving object and background, the candidate regions is introduced as one of three classes. Energy model based on the motion and the spatial constraints is build to perform the candidate regions selecting. A post processing is utilized to get the accurate object.To solve the occlusion problem between background and motion object, a novel video object...