Block-based Image Sequence Matching Motion Estimation Algorithm

As the key technique of computer vision and digital video processing, motion estimation of images sequence shows more and more important position and role not to be ignored in the various directions of production and life in society and has wide application value in industry, commerce, country defense, medicine, internet technology, etc. Recently, due to is slow computational complexity and high compression efficiency Block-Matching technique has been adopted by many video coding standards such as ISO MPEG and ITU-T H.26X, and gets great improvements and rapid developments in methodology and application and becomes a new research hot point of digital video technique.On the base of comprehensive summarization and deep analysis for current Block-Matching technique research work, a criteria for Block-Matching of restrictive conditional SAD and an adaptive motion estimation algorithm for real-time application based on image content have been proposed in this thesis. Experimental results show the new algorithms have obvious advantage in respect of enhancing accuracy and reducing the computational complexity, etc.At the beginning, a fully survey about motion estimation algorithms is given Matching algorithm based on block becomes to be considered, owing to its simplistic method, good estimation efficiency and potential to be implemented for the real-tune application using the available hardware. On the base of understanding for Block-Matching problem in nature, a systematic summarize is made for all current Block-Matching algorithm in the world. From the different search algorithm and search criteria adopted by different search algorithm based on block, we have analyzed their applicability, and compared them in terms of computational complexity.In the aspect of motion estimation criteria based on block-matching, the analysis is made to the current SAD matching criteria. Taking good advantaging of these ideas, a developing algorithm, restrictive conditional SAD based on Block-Matching, is proposed. At last, the computational complexity and accuracy of the solution are compared with existing algorithms. Experiment results show the good quality-performance, efficiency and robustness of the new criteria.An adaptive motion estimation algorithm for real-time application based on image content by the analysis of an adaptive initial search point method based on the correlation of motion vector is presented in the paper, which predict initial search point adaptively, uses different diamond search pattern based on image content and adopts effective stopping criteria. The experiment results reveal that with similar video quality as precondition, the proposed algorithm is two times faster than motion estimation algorithms in MPEG-4 standard and proposed algorithm possesses superiority on speed-up and search quality, so it fits real-time applications.