Infrared Image Sequence Analysis And Optimization Of The Moving Target Detection And Tracking System

Technology of video target detection and tracking is a very important field in modern science and technology. It involves image processing, automatic control and computer application, and so on. It is widely used in civil and military affairs, for example, advance warning, weapon control, video monitoring, video compression by contents segmentation and contents retrieval, and so on. This dissertation analyzes a infrared moving target detection and tracking system. The system can translate and store infrared image and target's data. Furthermore, the system can analyze the target feature and identify it. As a important terminal of intelligent information collection and analysis net, it can exchange information with other parts in the communication net.The dissertation analyzes the tasks stream and data stream, and optimizes the significant algorithm and its implementation. The hardware system is of typical high speed digital circuit which is the current hot research point. Some programmable chips such as DSP, FPGA, PCI9054 are integrated in the hardware system. Target can be captured and tracked in real time using appropriate program. This approach especially fits the image processing field where complex mass data operation is inevitable. With rational program, the hardware system can finish the work. Then, the dissertation develops the data stream to advance the performance of the board, and analyzes the image processing algorithm of the whole system. An improved iterative estimation algorithm with minimum error and orthogonal-sampled block matching algorithm are applied to estimate the motion between two frames. A pseudo-color presentation of 8-bits is used to display the image data of 14-bits.