Research On Infrared Image Enhancement And Reconstruction Based On Spectral Inversion

In the application of modern, the infrared imaging technology has many advantages, such as good concealment, good anti-interference ability, and being able to adapt to all kinds of weather and climate. In aspects of the military, it has been widely used. And it puts forward some higher requirements on corresponding performance because of many reasons. The resolution of the signal is low and the signal to noise ratio is also low because of the processing of imaging. What is worse, the contrast ratio is not high either, which brings a lot of difficulties to resolve.In many practical applications of infrared imaging distance is very far away, which leads to the perspective of target is very small. The very small target often does not have the shape of the information so they are often called small goals. In order to enhance the small target, this paper offered a new method. It borrows the thinking of the even component decomposition of seismic signal reflection coefficients improving the resolution greatly. It aims to solve the problem of low resolution of infrared image and hopes to increase its infrared energy refactoring through the research.(1) Research on the imaging principle of infrared imaging system, the external noise of image, some noise model of infrared image, infrared image and some characteristics of the small target. Two new evaluation mechanisms are introduced, namely, the local contrast and the significant evaluation index of small and weak targets, which make the evaluation of infrared imaging targets more reasonable.(2) Research on Several infrared image enhancement methods in spatial domain and transform domain are studied, including spatial filter, morphological filter, histogram equalization method, Fourier transform and wavelet transform. The infrared image data were tested, and the simulation results were analyzed and evaluated by using multi group parameters.(3) For the first time adding seismic spectrum inversion method to the image enhancement. On the same time it established a reflection model of the infrared energy. And we established the coefficient of wavelet domain decomposition model of odd and even components. It builds an objective function, and converted the problem of image enhancement into an optimization problem. The objective function was carried out by the particle swarm algorithm, and the results are analyzed and compared.(4) Evaluate a large number of actual data with the traditional evaluation parameter and the new proposed special evaluation of the evaluation of the small target.