| Atmosphere is an important part in space-based optical remote sensing chain. The cognition degree of atmospheric radiative characteristics directly affects the design and evaluation of photoelectric detection systems, and then the information quality and effectiveness of obtained data, thereby being a prerequisite for successfully executing remote sensing or detection missions. On the other hand, physical state changing information of the earth system over a wide range of spatial and temporal scales can be retrieved from the spectral radiative characteristics of the atmosphere, which is an important means to study earth sciences. At present, measurement and digital simulation are the two most proven ways to improve the knowledge level of the atmospheric background.This dissertation regards the mission requirement of China’s first atmospheric remote sensing satellite as the application background. The processing techniques and simulation methods of the infrared atmospheric image data are investigated deeply and systematically for the purpose of quantitative treatment and future application of the measured data. The research findings in data processing have already been applied to the satellite ground data processing system, while the formed models and methods in cloud image simulation are of great value to the specification determination and performance evaluation of space-based infrared imaging systems as well as to the inversion of atmospheric physical parameters both in theoretical and technical assistance. The main contents of this dissertation are organized as following:General design of the research scheme. The operation principles, modes and composition of the infrared camera are explained. Considering the practical measurement tasks and operating conditions of the camera, the radiometric calibration and geometric information acquirement are regarded as the core contents in measured data processing, and their corresponding research schemes are designed. According to the measured data and the practical application requirements, t wo simulation schemes are proposed, one relies on the statistic characteristics of the measured atmospheric radiance image data, and the other is on the basis of scene modeling and radiative transfer calculation.Study on the processing technologies of radiometric calibration. The calibration principle and the selection basis of calibration source are given from the radiative characteristics of measuring objects. The process and results of the pre-launch calibration test are presented. The data processing flow is formulated according to the process of the on-orbit radiometric calibration test. Three types of invalid calibration data are summarized and their filtering strategies are provided as well. A degradation model is proposed to characterize the camera response over time, and its applications are discussed in determining contamination tolerance and estimating calibration coefficients.Study on the processing technologies of acquiring the geometric information attached to the measured data. The imaging data are classified into earth background, limb background and space background according to the observing modes and measuring objects. The required geometric information for different background types is determined after analyzing, and their calculation models are provided and validated by comparing with STK results using the actual downlinked camera status data. Application examples of the solved geometric information are given at the same time.Study on the statistical characteristic analysis and simulation application of the measured atmospheric radiance data. The radiance distribution and power spectral density of the shortwave infrared(SWIR) images are analyzed. The feasibility and rationality of using fractal theory to generate cloud image are demonstrated. The two-dimensional(2D) rescale-and-add(RSA) fractal algorithm is introduced to construct the cloud texture structure. By numerical experiment, the relationship between the RSA algorithm parameters and the statistical characteristics of the simulated images is studied, and the method of parameter estimation is discussed. The measured images and the simulated images are compared to validate the effectiveness of the proposed simulation method.Study on the cloud image simulation method based on three-dimensional(3D) scene modeling and radiative transfer calculation. The RSA algorithm is extended to 3D to generate the spatial structure of cloud liquid/ice water content. The calculation methods of single scattering properties for spherical particles and randomly oriented non- spherical particles are given. The specular scattering phase function for horizontally oriented particles is modeled theoretically as a research focus, and is validated by the T-matrix and the Monte-Carlo method. A 3D radiative transfer model and a new optimization technique for the correlated k distribution parameters are proposed specially for absorption bands, and their accuracy is validated using the spherical harmonics discrete ordinate method(SHDOM). Finally the above simulation system is applied to 3D clouds with different structures, hei ghts and scattering properties. |
PDF Full Text Request