Infrared And Low-light Fusion System For Detection Probability Of Human Targets

Infrared and LLL image fusion technology take full advantage of the correlation between infrared and low light level imaging system.It has improved the operating distance and the detection probability of the target effectively. It has become an important task in the field of night vision. Human Target is one of the most common targets in the field of computer vision and it is one of the most important and active elements too. How to evaluate human target detection performance in infrared and LLL image fusion system is a problem that is urgently needed to solve for night vision field. Target detection probability is one of the most important indicators in describing and evaluating the detection performance of systems on human target. Based on the analysis of target characteristics, this paper proposed a quantitative method to calculate the detection probability of human target on human target.This paper analyzes the radiation spectral characteristics of human target in the near NIR, MWIR and LWIR, and detailed analyzes the specific image characteristics of human target in the MWIR and LWIR. Secondly, starting from human skin and reflection characteristics of different fabrics to establish the reflectance spectral distribution equation of human target, and analyzes the impact of human target reflectance spectral characteristics to the horizon of image intensifier. Again, aimed at the critical point that target characterization analysis is an important part of the imaging system’s detection performance, analyze the human target characteristics in the infrared and shimmer situation, we propose a method to calculate human target detection probability of image fusion system quantitatively. The experimental results show:the results calculated with this method agree with the subjective, and can be used to assess the detection probability of human targets in the image fusion system. Finally, MATLAB simulation tool is used to design a software system to assess the target detection probability based the MATLABGUI environment, and verify the validity of the evaluation software.