Passive Millimeter Wave Imaging Dim Moving Target Detection Algorithm

Passive millimeter-wave (PMMW) imaging technology forms images by detecting natural occurring millimeter-wave radiation from the scene and targets. Compared to optics imaging and infrared imaging, this imaging technology is less influenced by climatic conditions and could be operated with all-weather conditions. Compared to microwave, millimeter-wave has good directivity and its detecting and imaging system can have higher resolution. PMMW imaging technology has been applied widely in military affairs and civil applications such as battlefield reconnoitering, anti-terror detecting and security scanning. However, several key techniques of PMMW imaging technology need to research and perfect at present. Especially, the research of dim small moving targets detection of PMMW imaging is still at the beginning stage of development. This paper's research is relied on a specific research project. Focusing on the dim small moving targets detection of PMMW imaging technique, the main work of this thesis is shown as below.1) The ideal model of dim small target with background and noise has been analyzed. And the noise statistic of the real PMMW image has also been analyzed. Base on those analyses, the real model of dim small target with background and noise used by the detection algorithm has been presented.2) Four typical low SNR (Signal-to-Noise Ratio) image background suppression preprocessing techniques such as high-pass filter, median filter, self-adaptive filter and math-morphologic filter have been studied. And those four image preprocessing techniques have been emulated used a frame of PMMW image. By analyzing the emulation results, the advantages and disadvantages of the four techniques have been analyzed.3) The PMMW dim small moving targets detection algorithm base on DP (Dynamic Programming) has been studied. Based on the common theoretic analysis of DP, the principle of dim small moving targets detection used DP, the detecting steps and the algorithm flow have been summarized. The calculating method of the final detecting threshold has been improved. The validity of this technique has been proved by the simulation experiment.4) Based on the study of TBD (Tracking Before Detecting) algorithm used MHST (Multi-Stage Hypothesis Testing) and combining the advantages of DP and MHST, the improved algorithm of the DP dim small moving targets detection based on MHST has been developed. The validity of this technique has been proved by the simulation experiment. And the simulations show that this improved algorithm improves the detection performance,and reduces storage level compared with original algorithm.