Hyun Joo-Bong, Hwang Dong-Choon, Shin Dong-Hak, Kim Eun-Soo
Department of Electronic Engineering, 3D Display Research Center, Kwangwoon University, 447-1 Wolge-Dong, Nowon-Gu, Seoul 139-701, South Korea.
Appl Opt. 2007 Nov 1;46(31):7697-708. doi: 10.1364/ao.46.007697.
A novel curved computational integral imaging reconstruction (C-CIIR) technique for the virtually curved integral imaging (VCII) system is proposed, and its performances are analyzed. In the C-CIIR model, an additional virtual large-aperture lens is included to provide a multidirectional curving effect in the reconstruction process, and its effect is analyzed in detail by using the ABCD matrix. With this method, resolution-enhanced 3D object images can be computationally reconstructed from the picked-up elemental images of the VCII system. To confirm the feasibility of the proposed model, some experiments are carried out. Experiments revealed that the sampling rate in the VCII system could be kept at a maximum value within some range of the distance z, whereas in the conventional integral imaging system it linearly decreased as the distance z increased. It is also shown that resolutions of the object images reconstructed by the C-CIIR method have been significantly improved compared with those of the conventional CIIR method.
提出了一种用于虚拟弯曲积分成像（VCII）系统的新型弯曲计算积分成像重建（C-CIIR）技术，并对其性能进行了分析。在C-CIIR模型中，增加了一个虚拟大孔径透镜，以在重建过程中提供多方向弯曲效果，并利用ABCD矩阵对其效果进行了详细分析。通过这种方法，可以从VCII系统采集的元素图像中计算重建出分辨率增强的三维物体图像。为了验证所提模型的可行性，进行了一些实验。实验表明，在距离z的一定范围内，VCII系统中的采样率可以保持在最大值，而在传统积分成像系统中，采样率随着距离z的增加而线性下降。还表明，与传统CIIR方法相比，采用C-CIIR方法重建的物体图像分辨率有了显著提高。
