Small And Dim Targets Detection In Infrared Image Under Complex Background

Small And Dim Targets Detection In Infrared Image With Complex Background is a old-line and attractive research topics in the field of infrared(IR) image processing. In recent years, With the quick development of computer and digital image processing technology, the infrared target detection technique start to be extensive apply in make to lead, follow, automatic-control, artificial-intelligence etc. many realm. Because the infrared moving weak and small target has the characteristics that the distance. the area is small, the shape characteristic is weak the detail characteristic part lose and the SNR is low etc.,to the infrared moving weak and small target detection is a important and difficult point of the target detection.As for the real weapon systems, how to make the best of IR target detection techniques to increase the distance of target detection and to obtain the related information about the invading target, have become important factors to decide the victory or defeat of modern warfare. The longer the distance of targets, the less the imaging area of targets and the larger the probabilities of targets influenced by backgrounds and clutter will be. Therefore, comparing with other topics in the field of IR target detection and tracking, how small targets can be robustly detected and tracked under complex backgrounds have become the more realistic and challenging research topics.The paper mainly researches the methods of the small and dim targets detection in infrared image with complex background. Firstly, the state of arts of infrared weak and small targets detection is reviewed in domestic and abroad, and the mechanism of infrared imaging is analyzed. Secondly, based on the analysis of real infrared images' features, the anther shows the model of infrared image, then proceed to the next step, analysis the imagery characteristics of noise, background and target. Thirdly, the paper systemically studies the theory and algorithm for the small and dim targets detection in infrared image with complex background, focus on the DBT target detection algorithm implementation process. Classic DBT target detection algorithm divide the process of targets detection into two parts clearly: single-frame targets detection and multi-frame targets detection. Therefore, from the macro-level, Different image preprocessing,the image segmentation algorithm of single-frame targets and multi-frame targets detection algorithm will be make up of the DBT target detection algorithm that is suitable for the different specific instance. By all appearances, such algorithm not only having a simple design, but also the program of the algorithm is easy to realize. From a micro perspective ,The DBT algorithm include 4 parts: image preprocessing,background suppression,single-frame detection targets and multi-frame targets detection. Among them the image preprocessing include digitizing images,non-uniformity correction,grayscale calibration and geometric correction; Background suppression use appropriate background suppression filter for filtering, in order to achieve the purpose of background suppression,noise suppression and targets enhancement, facilitating behind the Target Detection. Single-frame target detection module using traditional single-frame image of the target to complete the segmentation method. The mission is finding targets from the elimination of the background image. Multi-frame detection is using the characteristics of the movement(the continuity and the consistency of the Track),Radiation intensity,to identify the target trajectory from several pictures,as to identifying the target detection and outputting the results ultimately. This paper presents the algorithm are two types of algorithm DBT. Finally, in chapter 3 and 4, the thesis discusses the principles and the basis operations of the Wavelet-transform and the Mathematical Morphology separately, and shows two detection algorithms of small target based on the Wavelet-transform and Mathematical Morphology. Author discuss principle and structure of the two algorithms, and focus on the theoretical analysis and simulation experiment. There into, the detection algorithms of small target based on the Wavelet-transform aims at the one image, and it is based on the characteristic of small targets. Firstly, the Firstly, Wavelet image denoising could enhance the contrast of the image ,Secondly, by multi-resolution and pyramid decomposition, we can make small target be strong.Finally, wavelet image segmentation help us to get the target. the detection algorithms of small target based on Mathematical Morphology makes use of Top-Hat transform to finish background suppression firstly,then We will clean false targets and noise by opening operator,Finally, the infrared image is segmented with improved watershed,and we can get the target. The result of experiment the system is reliable and show that the arithmetic is reasonably, and effective.we can accurately get the target from infrared image With complex background. At last, the author summarizes achievements in the thesis and points out the future work.