体素
计算机科学
混叠
人工智能
计算机视觉
镜头(地质)
图像分辨率
点扩散函数
积分成像
还原(数学)
光学
物理
数学
滤波器(信号处理)
图像(数学)
几何学
作者
Yi‐Jian Liu,Xue‐Rui Wen,Xiao‐Shuai Hu,Weize Li,Yan Xing,Chao Liu,Han‐Le Zhang,Qiong‐Hua Wang
标识
DOI:10.1002/lpor.202400182
摘要
Abstract Integral imaging (InIm) 3D displays provide realistic 3D images with comfortable viewing experiences. However, existing approaches struggle to achieve high spatial resolution reproduction without relying on temporal or spatial multiplexing methods, which increases system complexity. Although precise control approaches for voxel formation have been explored to address this challenge, the existing methods lack systematic analysis theories for identifying the key factors affecting spatial resolution. In this study, a novel analysis framework is proposed that uses a point spread function (PSF) to characterize voxel formation. An analytical model based on the obtained PSF is established to achieve an aliasing‐free display. This framework identifies optimization directions for improving the spatial resolution, specifically by reducing the Gaussian blur caused by wavefront aberration in the lens unit. A targeted optimization of a compound lens array is conducted to achieve precise control of voxel formation. In these experiments, an InIm 3D display system prototype that substantially outperforms conventional InIm 3D displays is developed. The prototype achieves a 74.6% reduction in the system PSF and a 29.85% reduction in voxel size, enabling aliasing‐free voxels and a 2.82 times higher resolution in the InIm 3D display.
科研通智能强力驱动
Strongly Powered by AbleSci AI