Research And Simulation For Wide-Field High- Resolution Camera Based On Computational Imaging

Field angle and resolution of an optical system are always the major directions of the improvement of optical design, for the reason that they are two of the most important performance indicators. However, the traditional optical design faces a fundamental tradeoff between the field angle and resolution. An optical system can hardly meet these two demands at the same time. Besides, currently available wide-field high-resolution imaging systems are more or less limited by the difficulty of optical design, the cost of materials and manufacturing and the limitation of imaging. Wide-field high-resolution imaging is widely used in astronautics, aviation, safety monitoring and so on. And computational imaging has become an effective way to achieve wide-field high-resolution imaging because of its unique advantages.After a research on the background, requirement and technical difficulty, the development situation of the wide-field high-resolution camera at home and abroad is summarized, including monocentric multiscale theory, computational imaging technology, image restoration technology and image quality assessment.Then the theory of wide-field high-resolution camera based on computational imaging is introduced. The scaling laws for lens systems in traditional optical system are derived to describe the relation between optical system performance and construction. And the changes of these laws in computational imaging system are discussed next. Multiscale optical design and monocentric optical design are introduced. And the monocentric multiscale computational imaging system is focused, which combines the unique of multiscale design and monocentric design.The imaging model of optical system with space-variant point spread function is introduced. And the relation between PSF and geometrical aberrations of optical system is researched by calculation of pupil wave aberrations. The expressions for the pupil wave aberrations and the characters of geometrical aberrations and PSF are discussed. Restoration technology for space-variant PSF image is researched to solve the image degradation in the monocentric multiscale computational imaging system. Image processing technology based on color-separated correction for optical imaging system with aberrations is presented to correct the aberrations. In this method, color image is separated into red, green and blue channels. A non-linear image scaling method is used to harmonize the magnifications of different colors to correct the lateral chromatic aberration. The monochromatic aberrations are corrected by dividing the image to sub-frame sections and deburring the sub-sections using Lucy-Richardson algorithm. At last, the images in red, green and blue channels are fused to be the corrected image.After the research of the monocentric multiscale wide-field high-resolution camera based on computational imaging, a software simulation system is built. A single ball lens computational imaging system is built using optical design software and its imaging characters are analyzed. After that, a relay lens is added to complete the computational imaging system and its characters are also analyzed. Blurred image obtained by image simulation of the designed optical system and is corrected by the color-separated method. Corrected image is restored by fusing the red, green and blue channels after the correction for the lateral chromatic aberration and deburring. The simulation result shows the feasibility and effectiveness of the color-separated correcting method which is better than the classic image restoration method. Imaging process of the whole system with lens array is also simulated. Aberrations in every unit system are corrected and the whole image is obtained by image mosaic. Several simulations prove that the monocentric multiscale wide-field high-resolution camera based on computational imaging is feasible and effective.