Research On Infrared Dim Target Detection Under Complex Backgroud

With the rapid development of computer and digital image processing technology, infrared target detection technique has been used in many applications, such as astronomy prognosticates, remote sensing, target tracking, guiding etc. But with the characteristics of far distance, small area, weakness of shape of image, lose of detail of image and low SNR etc, the infrared weak and small target detection technique is one of the key and difficult point of infrared guidance system.Because of the characteristics of infrared weak and small targets, the research of infrared target detection of theories and methods is concentrated on rising of SNR, the detection ability of algorithm, coverage, efficiency of computation and reliability of detection algorithm. The research of infrared weak small target detection is still aissue with great challenge. In this regard, this thesis does a detail research on infrared dim target detection under complex background.The main contents of this thesis are as follows:(1) Firstly, some relevant concepts of small target and mathematical description of IR image are presented. Then the three elements (target, background and noise) of small target detection in the IR image are analyzed. Small target can be not the highest in the gray-scale, but compared with the local context it is more prominent. That is to say, it should have certain contrast in the local background and almostly no shape information.(2) Secondly, information entropy weighted by image variance is introduced to analyze the images above and describe image complexity, and substantial causes of the complexity features in classified regions are discussed, where based a novel image preprocessing method and a self-adaptive threshold acquisition method are constructed, so that the dim targets can be finally detected with self-adaptive threshold processing.(3) At last, in order to solve the target detectiong of image sequences, Multi-scale three-dimensional wavelet decomposition is introduced and then filter out the Low-frequency Component, Removing the detail coefficients, Retaining the Motion information and Edge information. Large number of background disturbance is removed by using Image morphology and candidate target points are obtained. Gray of the candidate target points is detected, threshold is determined, the false alarm points are eliminated and the target segmentation is carried out. The experimental results show that this method can detect infrared small targets in complex backgrounds.In summary, the infrared target detection problems are researched in this thesis, and new algorithms have been proposted.