Research On Degradation Effects Of System Errors On Image Quality And Compensation Methods In Integral Imaging

Integral Imaging(II) is one of the frontier research aspects in the field of autosteroscopic three-dimensional(3D) display, which is regarded as one of the most promising 3D display techniques. System errors are hard to be avoid in a complex 3D optical imaging system consisting of multiple devices, which can lead to the degradation of II display quality. However, there is few research on the degradation mechanism of system errors and the display quality compensation methods. Hence, it is necessary to study the degradation effects of system errors on the image quality and the display quality compensation methods in order to realize high quality 3D display in II system.Focusing on the degradation mechanism of system errors and the display quality compensation methods, this paper carries out the following research work:(1) The geometric optics and the ray tracing theory are utilized to build the image degradation theory model caused by system errors based on the integral imaging principle by considering the optical axis translation motion and the pixel information dislocation. Then, the effects of system errors on the image quality in II system are analyzed quantitatively on the basis of the above theory model. Finally, system tolerances of errors limitations are given corresponding to human vision system observable threshold for different system parameters, which will provide theoretical support for the optimization design of II system.(2)In order to evaluate the 3D performance of II system objectively, the 3D point spread function model of II system is derived based on wave optics theory by considering the diffraction effect of the pickup micro-lens and the display micro-lens, and the sampling effects of the image sensor and the display device. Furthermore, the above model is modified for II systems with different errors. Then, based on these models, the effects of system errors on the 3D point spread function are analyzed by theoretical calculation. Finally, by considering the resolution limitation of human vision, the system tolerances of errors limitations are given at a certain observation position. Under the given experimental conditions, the local error limitation of the system is 0.16 mm, the global error limitation ofthe system is 0.02 mm, the rotational assemble error limitation is 0.5 degree, the titled assemble error limitation is 0.5 degree, the defocus error limitation is 0.5mm and the pixel size limitation of the display device is 0.05 mm, respectivelly.(3)Focusing on the compensation of the 3D image degraded by the local positon error, the global position error, the rotation misalignment, the tilted misalignment and the defocus misalignment, homogeneous light field models with the above errors are derived based on the existing homogeneous light field conversion model, respectively. Based on the captured elemental image array(EIA), a new set of EIA that matches the display system is generated. Thus, the image degradation caused by system errors is compensated.(4) The sub-pixel coding method is proposed to compensate the image quality degraded by the unmatched relationship of the pixel size of the image sensor with that of the common display device. In this method, a pixel in common display device can be devided into many parts through placing a coded sub-pixel mask on it. Because the eyes experience the effect of the afterimages, we can display as many pixels of slightly different perspectives as possible by changing the coding mode of the coded sub-pixel mask and the displayed image within the time constant of the eyes’ response time. When the pixel size of the common display device is n′n times of that of the image sensor, the resolution of the II system can be improved n times in theoretical.