Motion Objects Detection Technology Research Based On Complex Background

With the development of hardware and software of computer, computer vision technology is attracted more and more attention by people. Especially in these fileds,which include military affairs,aviation spaceflight,CAD,smart android and so on, computer vision technology is used widely. The vision technology which is based on video frequency image sequences is the most important computer vision technology, so it has became hot points and difficult points in investigate works of people. Motion objects detection and tracking which is based on video frequency is also the basic and key of computer vision technology. Firstly, weather we can accurately detect motion objects from video frequency images sequence, is directly relation to other next processes weather can get success. For detecting motion objects, many scholars have doing interrelated investigations, and provide a lot of motion objects detecting arithmetic. This paper is based on other people's working, adequately consider the relativity of time and satirically of frequency image sequences, provide a moving objects detection algorithm colligated spatial-temporal information, and from the result of experiment proves validity and real-time character of the algorithm. For detecting motion objects from complex background, this paper uses a algorithm which is based on antilogy of statistical pattern recognition, and we have used the algorithm to detect swimmers from indoor natatorium, and acquire better detecting result. Secondly, based on motion detection, this paper uses a multiresolution tracking algorithm which is based on Hausdorff distance to track motion objects, from the result of experimentation, the tracking algorithm can accurately tracks motion objects, and if some motion objects shelter from other motion objects, the tracking algorithm can distinguish from different motion objects and respectively tracking, have definite robustness. At the same time, this paper is based on motion detection and tracking algorithm, make use of Visual C#.NET technology and VFW technology to achieve a motion detection and tracking system, and from actual testing to insure usability and stability of system.In order to organize this paper well the content, this paper uses the following chapter structure:First Chapter Introduced the computer vision technology survey, the research present situation, the application domain as well as the existence question, especially emphatically introduced the motion detection and tracking technology research present situation, the research value and the existence question.Second Chapter Introduced the video frequency picture processing related digital picture processing elementary knowledge, including image enhancement, picture filter, mathematics morphology, color model and so on.Third Chapter Introduced the commonly used motion detection algorithm, these methods may divide into two kinds: Frame difference algorithm and background difference algorithm. Works in the predecessor in the foundation, this paper proposed the auto-adapted motion detection algorithm(AMODA).In order to fully uses the video frequency picture sequence the space and time information, in the auto-adapted motion detection algorithm and in the symmetrical difference algorithm foundation, this paper proposed the comprehensive utilization space and time information motion detection algorithm(CSTMODA).Under the complex background motion detection algorithm technology aspect, this paper based on the most minor sort other divisible criterion thought, proposed one kind in view of the illumination condition quite stable situation in motion detection algorithm(CBMODA).Fourth chapter Introduced several kind of commonly used motion tracking algorithm, and is away from the measure algorithm in original Hausdorff in the foundation, proposed bidirectional Hausdorff is away from the measure the multi-resolution track algorithm (BHRMOTA),solved well has covered with the multi- goals situation in the motion tracking question.Fifth chapter Based on Visual C#.NET platform and VFW tool bag to achieve a motion detection and tracking system, and has carried on the actual test to the system.Sixth chapter Summarizes the full text, and proposed several ideas to the further research work-...