Research On Automatic Adaptive Exposure For Space-Borne Remote Sensing Camera

As the rapid development of space remote sensing technology, much attention was given to remote sensing information, and improving image quality is the key to obtain remote sensing information effectively. The exposure is the important parameter for the design and application of remote sensing camera, which is effect on the image quality of remote sensing camera. For the adaptive automatic exposure technology of remote sensing camera, the information of target scene should be obtained in advance, and it is used as the basis for regulating exposure parameter, which avoid the loss of image information caused by the underexposure and overexposure, hence the adaptive automatic exposure technology of remote sensing camera is a key technology for improving the image quality of remote sensing camera. Since its mode of push scan on image, the space sensing camera based on linear array TDICCD only imaging once to the same scene, so it cannot forecast the information of the target scene, which increase the difficulty of adaptive automatic exposure on space. In this paper, we design a focal plane structure of TDICCD based on the CMOS sensor to achieve the adaptive automatic exposure for the space sensing camera based on linear array TDICCD. Then we analyze and judge the target scene by using the CMOS image, and the adaptive automatic exposure on space is achieved. This paper finishes the job in five following aspects:(1)Deeply research on the principle and common methods of adaptive automatic exposure, and we analyze working principle and the imaging mode of TDICCD space sensing camera. We proposed the key point of adaptive automatic exposure of space camera by analyzing the influence to adaptive automatic exposure, and then we introduce the method of space sensing camera imaging quality evaluation in detail.(2) Aiming at the difficulty of obtain the target scene information, a focal plane structure of TDICCD based on the CMOS sensors is designed to obtain the target scene information. Combining with the imaging mode of CMOS, the irradiance at the pupil of space sensing camera can be calculated by use the image of target scene, which can be used for regulating exposure parameters. By analyzing the relationship between the power spectrum of image and the exposure time, a evaluation function of exposure time is established to ensure the exposure time.(3) Aiming at the effect of cloud on the calculation of the irradiance at the pupil and the power spectrum of image, we proposed a real time method of cloud judgment based on the CMOS image. By establishing the spectral irradiance model at the entrance pupil of the space camera, a theoretical value of the irradiance at the entrance pupil under cloudless day is obtained, and then by comparing the theoretical value and the measured value, we judge if the image has cloud quickly. After that, by using clustering method, the cloud recognition and elimination of the image which includes cloud can be finished, it improves the accuracy of irradiance calculation and power spectrum statistics.(4) Base on the result of exposure feature statistics, we finish the preliminary exposure based on the mid-grayvalue and the accurate exposure based on power spectrum, and then we derive the exposure formula of TDICCD by considering the integration series and gain. We proposed an image enhancement method based on self-adaptive Laplacian filter, and it improves the visual effect of original image and avoid the loss of Information of image caused by rejectting lower parts of data. We analyze the effect of the exposure parameters on the index of remote sensing cameras, and determined the constraint condition of the exposure parameters. Finally, we correct the exposure parameter by use of the image of TDICCD, which is to improve the stability and validity of adaptive exposure method.(5) We introduced the components of the TDICCD space camera, and design a hardware system for adaptive exposure based on FPGA, and verify the proposed adaptive exposure method.This paper research on the Adaptive Automatic Exposure algorithm, and design a hardware system for adaptive exposure, a lot of experiments are conducted to verify the proposed adaptive exposure method. The experimental results shows that the adaptive exposure algorithm can achieve the expectation of adaptive exposure, and the image quality is improved effectively, the instantaneity and validity can satisfied the engineering requirements.