Designed Dsp Processing Unit And Algorithm Transplant And Implementation Of Migration Of Millimeter Wave Imaging

Passive millimeter wave(PMMW) imaging systems without itself emitting electromagnetic signals, receiving millimeter wave energy that radiation scene and the object itself, and use of different objects in the scene of the millimeter-wave radiation energy intensity differences to achieve imaging system. Because millimeter wave has no radiation and can penetrate clothing, shelter, Even if in the dust, smoke, darkness and other harsh conditions that it can work around the clock, it has been widely used in security surveillance, military reconnaissance and other fields. In order to meet the needs of application scenarios, improve real-time imaging system, passive millimeter wave system uses to digital signal processor(DSP) core processing architecture. Where in the digital signal processing unit as a key part of the system hardware, primarily responsible for obtaining the original image from the front millimeter wave, and image processing, the processed image is transferred to the terminal display.In this paper, based on the multi-channel passive millimeter wave project, data-link design based on DSP unit data-link design based on DSP units, each algorithm of the system is analyzed, and transplanted to DSP unit execution, the main work includes:1. According to the principle of passive millimeter wave imaging and image features, combined with the project’s technical specifications drawn framework of the whole system. For DSP application development, discussed specific application development platform and hardware and software development process.2. Detailed analysis of the different projects were analyzed according to the demand for DSP selection, according to the actual processing platform configuration-related resources, design the DSP unit data interface with various other modules and debugging, finally demonstrate the effectiveness and stability of the entire system data link.3. According to the sytem of processing algorithms,analyzes the various algorithms.First use the different image preprocessing algorithm, After the use of a priori information on the gradient of the millimeter wave image super-resolution reconstruction algorithm to restore the details of the image, so that the image has a higher target information, based on background subtraction algorithm moving targets on the scene monitoring the object extraction, the final analysis of the fusion algorithm based on region segmentation and multi-scale decomposition. Through the analysis of system-related algorithms, understand the complexity of the algorithm and portability.4. Discussion of requiring major C/C++ programming problems in the algorithm transplant procedure, in which the principles of image pre-processing algorithms, image super-resolution algorithms and target detection algorithm have been achieved transplantation, combined with project performance indicators and C64+ hardware platform, algorithms are been optimization.Through experiments base on simulated and real data, verify the validity of the design of data transmission interface DSP unit and Analysis of the various algorithms portability, and basically complete the transplantion and optimization.