| Humans live in a three-dimensional world and obtain more than 80%of the external information through the visual system.However,the traditional two-dimensional display technology can only display the specific angle of the scene without the depth information,which limits people’s access to the whole information of the target scene.Different from two-dimensional display technology,three-dimensional display technology can truly bring the three-dimensional world to the audience and realize their high pursuit of three-dimensional sense,realism,and visual impact.Among three-dimensional display technologies,integral imaging 3D display technology can use all depth cues in human vision,including motion parallax,binocular parallax,and correct occlusion relationship,to reconstruct the stereo 3D image of the target scene.Therefore,it is considered to be the most promising 3D display technology.Since integral photography was first proposed by Lippmann in 1908,integral imaging technology has developed by leaps and bounds.Researchers have made a lot of efforts to improve and optimize the integral imaging display technology.However,the design and improvement of integral imaging system is always a time-consuming and labor-consuming process,especially when the display effect of the equipment cannot be obtained directly.This dissertation deeply studies the visualization simulation and optimization design of integral imaging,proposes effective solutions,and gives experimental results and analysis to prove the feasibility of the scheme.The main contents and innovations of this dissertation are as follows.(1)Backward ray tracing based modeling and high-speed visualization simulation method for integral imaging display systemTraditional commercial optical software focuses on the analysis of optical characteristics,rather than graphic visualization which can help to quantitatively analyze the integral imaging system.The commonly used computational integral imaging reconstruction algorithm can only reconstruct the specific viewpoint image by periodically synthesizing the fixed pixels through image processing technology,and the amount of calculation is too large to achieve real-time performance.To solve the problems above,a backward ray tracing based modeling and high-speed visualization simulation method for integral imaging display system is proposed.The geometric and optical models of each component of the integral imaging display system are established separately in the virtual scene,and the reconstructed images are obtained by tracing the scene based on backward ray tracing,and the field of view angle and depth of field of the visualization simulation results are estimated.The visualization simulation results are consistent with the theoretical and optical experimental results,which proves that the proposed method can predict the display images of integral imaging systems with different viewpoint positions in almost real-time.The experimental results show that the simulation time is 1 s when the number of sampling rays is 3840×2160 ×100.(2)Integral imaging display optimization method based on lens array modeling and simulationIntegral imaging 3D display needs optical media with specific curved surface structures such as lens to realize directional light control.Due to the different refractive index,the introduced optical medium is bound to lead to the existence of aberration,which will reduce the imaging quality,viewing angle,and depth of field.To solve the above problems,an integral imaging display optimization method based on lens array modeling and simulation is proposed.CSG modeling technology is used to model and simulate the lens in integral imaging display before scene acquisition,which is used to correct the starting point and direction of sampling ray.In addition,the assembly error of the lens is additionally considered.The corrected sampling light is parallelly calculated based on ray tracing in parallel,the path of the sampling light is highly consistent with the path of light propagation in the imaging process.In this way,the collected rendering results can be reconstructed into high-quality aberration free stereo images,so as to optimize the display effect of integral imaging.The Structural Similarity(SSIM)of the 3D image generated with our proposed method is 0.1 higher than that of the traditional pinhole method.The realtime frame rate also exceeds 25fps.(3)Automatic co-design of integral imaging light field display system based on simulated annealing algorithm and visualization simulationThe integral imaging light field system is complex from acquisition to display.There is still no effective method to co-design the global parameters of the integral imaging light field system.To solve this problem,an automatic co-design method of integral imaging light field system based on simulated annealing and visualization simulation is proposed in this dissertation.The process of integral imaging light field display system content acquisition and optical reconstruction are modeled and simulated,the simulation results of the system are obtained based on the backward ray tracing,and the objective function for evaluating the display effect of the system is established based on SSIM and Peak Signal-To-Noise Ratio(PSNR)according to the simulation results.The simulated annealing optimization method is used to find the optimal parameters of the integral imaging light field display system.According to the optimization results,a real integral imaging display device is established.The physical experiment shows that the integral imaging display system designed with the proposed method can produce clear and complete reconstructed 3D images,and can provide high-quality natural stereo visual effect for the observers.With our proposed method,the quality of the 3D image is higher than that of the traditional parameters. |
PDF Full Text Request