摘要
阵列相机是一种采集光场的常用方法,可以对多路光场直接采集,但是对于每一路光场都需要一台相机设备进行采集,成本较高、体积庞大。本文提出一种基于压缩传感的光场采集与重建方法,这种方法使用少量相机组成稀疏相机阵列采集部分光场,利用超冗余字典,实现原始光场的重建。实验结果表明,稀疏相机阵列可以采集光场,并利用压缩传感技术训练合适的超冗余字典恢复出较高质量的原始光场。
Camera array which can capture multiple views of light filed directly is used as a common light field acquisition equip- ment, but camera array is difficult to maintain and high cost as for each view one camera is needed. This paper proposes a method of light field acquisition and reconstruction based on compressive sensing. Much fewer cameras are used to achieve the functions of full-populated camera array. The sparse camera array captures fewer views and reconstructs the original light field by an over- complete dictionary. The experiments demonstrated it is feasible to reconstruct the original light field at a high quality level.
出处
《计算机与现代化》
2015年第4期61-64,69,共5页
Computer and Modernization
基金
国家863计划项目(2012AA011903)
关键词
图像处理
光场
压缩传感
image processing
light field
compressive sensing