基于压缩传感的全息图压缩研究

Hologram Compression Based on Compressive Sensing

提出将压缩传感理论用于数字全息图的压缩研究.研究了2种不同变换域对全息图稀疏化的影响,针对全息图的特点选取合适的稀疏域,然后根据压缩传感理论直接获取图像的压缩表示,最后利用得到的压缩数据采用2种不同算法重构全息图并对比其重构效果.实验证实这种压缩方法是可行的,并且在傅里叶稀疏域下使用正交匹配追踪(OMP)重构算法能获得1.5%的压缩率.

Hologram is the image that contains the amplitude and phase information of an object, and the compres- sion of hologram is different from the nature image. This paper presents a new theory for digital hologram compres- sion using compressive sensing. A study of the impact of two transform domains to the hologram sparse is first pres- ented, which selects the appropriate transform domain according to the character of hologram, then it directly ac- quires the compression data of hologram using compressive sensing theory, recovery the hologram using these com- pression data by two reconstruction algorithms and make a comparison of their construction effects. The main pur- pose of this paper is to explore the feasibility of the CS framework for hologram compression. The experiments dem- onstrate that the method proposed is available and it can achieve the compression ratio of 5% when the OMP recon- struction algorithm is used in Fourier transform domain.