| Infrared detection system has been more and more widely used in urban security, fire monitoring, remote sensing, traffic safety and many other fields for its good observing capabilities during day and night. To test the target recognition capability of infrared detection system, numerous infrared images under different conditions are needed as the test data. However, if the infrared images were completely obtained with the experimental method under different conditions, it will be high cost, time and space constraints and so on. Therefore, it is necessary to get infrared images by simulating. In order to improve the fidelity of infrared simulation image, some key techiques are studied, including:modeling of infrared imaging link and method of infrared imaging simulation, modeling method of surface temperature for typical ground objects based on measurement data, emissivity extracting method based on infrared images and automatic data acquisition experimental platform.In this thesis, the infrared imaging process has been analyzed, a mid-infrared imaging link model has been built and the method of generating infrared images in different time from the real infrared images has been researched. The result shows that the advantage of this method is that the generated infrared images are of high accuracy.The modeling method of surface temperature for the typical ground objects has been researched based on measured data. In this method, some typical ground objects have been classified and their shared features have been found out. Base on the structure, material and radiative heat transfer features of the classified ground objects, their surface temperature models have been built. The coefficients of the models under different conditions have been achieved with multivariate fitting method based on the measured surface temperatures data of ground objects in different time and seasons. So, the surface temperatures of ground objects under different conditions can be calculated, which provides practical and effective surface temperatures for infrared characteristics simulation.According to the infrared imaging theory, the algorithm for extracting the emissivity of typical ground objects has been studied. In the method, the infrared radiance of one area has been calculated based on real infrared image and the imaging link model, the temperature of this area has been calculated based on the surface temperature model, the corresponding emissivity has been calculated based on Planck’s radiation law and the surface temperature. According to the imaging link model, emissivity and real infrared image, the infrared image in the other time can be simulated. Then, to correct the emissivity, the image gray values of simulation and real in the same time have been compared. The extracted emissivity, with this method, can be used for mid-infrared image simulation in different time and seasons.An automatic data acquisition experimental platform for infrared imaging simulation has been set up to get the data of selected ground objects in the interesting area synchronously, automatically and in real time. This platform can provide comprehensive and effective data to support the modeling and validating of infrared simulation in this thesis, which includes infrared images, meteorological conditions (e.g. air temperature, atmospheric pressure, wind speed, visibility, etc.), sureface temperatures of ground objects, location and distance information, etc. And the testing work of infrared characteristics simulation of ground objects has been done based on the measured data.By the above research work, high-accuracy infrared simulation images of typical ground objects have been achieved by simulation method, which provides an important technical support for testing the ability of target recognition for infrared detection system. |
PDF Full Text Request