All-Weather Long-Wave Infrared Image Modeling And Simulation For Out-door Scenes

Long-wave (8-12μm) infrared imaging and simulation for outdoor scences in all-weather condition focus on how weather, climate as well as capturing time affect on infrared image, and produce infrared images of out-door scenes under different weather conditions. The key techniques are to establish a LW infrared imaging model under different weather, climate, time and location, and simulate the infrared images under given weather conditions from the known infrared images with a known weather condition. From physical principle of infrared imaging, this thesis integrates the theories in image simulation, heat conduction, radiation transfer in atmosphere and solar radiation to analyze the effection of weather on infrared imaging from the radiation of object to image sensors, and establish weather imaging model, and proposes the LW infrared image simulation methods under different weather coniditions. One of the most important features of this study is that it can produce the LW infrared image of outdoor scences under different weather conditions from a known LW infrared image with the known weather condition. The LW infrared radiation from outdoor scences passes through the atmosphere, reaches the imager sensor, and is converted into image gray scales by infrared CCD camera. In this imaging procedure, the radiation of object and background propagated in atmosphere is sensitive to weather variation while the effect of weather conditions on the camera model is small. Thus, this thesis concentrates on the LW infrared image simulation of outdoor scene as well as atmospheric effect. Because imaging principal of infrared camera and weather related solar radiations are the basis of infrared imaging simulation, the problems such as the point spread fuction of infrared camera and solar radiation calculation are also addressed. Two aspects of infrared camera are studied. One is the relationship between image gray scale and the surface temperature of object, while the other is the measuremeant of point spread function of infrared CCD camera. First, the author calculated the infrared radiation emitted by the object and received by the camera after a reasonable simplification to the radiation of object in normal temperature, and the relationship between image gray scale and the surface temperature of object is derived. Secondly, the point spread function of an infrared camera used in our experiments was estimated using a modified Knife-Edge technique. The object temperature and atmospheric optical property are depended on sun radiation. This thesis discussed the calculation of sun position and radiation including sun orientation, relative time as well as the sun radiation received by object. Atmospheric effect has an important impact on the quality of infrared images. In this thesis, the author discussed atmospheric optical properity, analyzed the influence of atmospheric molecules and aerosol particles on the absorption and scattering of infrared radiation, which causes image contrast to descend and blur the image, and presented a method to simulate the atmospheric effect. The atmospheric effect on infrared image can be represented by atmospheric point spread function (PSF). First, atmospheric point spread functions under different weather conditions were calculated; then, the infrared image was obtained. Finally, the captured image was convoluted with the atmospheric point spread function and point spread function of the camera to obtain the image with the atmospheric effect. Two types of weather conditions were considered: 1) clear weather in a horizontal view; 2) rain and fog weather in a horizontal view. In the simulation process, the atmospheric point spread function was obtained by calculating the atmospheric modulation transfer function (MTF) including aerosal MTF and atmospheric turbulence MTF; the point-spread functions for rain and fog were calculated by Monte Carlo method. The point-spread function of the image was determined by using the modified Knife-Edge technique, followed by Wiener filter. After a long term observation to the infrared images of sky near ground scenes, we found that the mean gray level of image, the horizontal and vertical scanning line exposed sorts of distribution. Based on these distributions, an infrared image simulation method of sky near ground was proposed. In order to simulate infrared images of object, a thermal energy exchange equation about the object surface and its surrounding area was set up after simplification of the equation. The relationship between gray scale of object surface and the infrared characteritics of object surface, temperature of object surface, time, weather parameters, as well as geographical information of the scene are built. Using the parameters in the heat equation determined in a non-linear fitting of the infrared images with the known weather conditions, the gray scale of object under different weather conditions can be estimated.Furthermore, the simulation of infrared images of sky small object and a vertical plane surface was presented.