红外与毫米波学报, 2017, 36 (6): 799, 网络出版: 2018-01-04
多帧图像编码孔径光谱成像技术
Multi-frame image coded aperture snapshot spectral image technology
光谱成像 压缩感知 编码孔径光谱成像(CASSI) 重构算法 spectral image compressed sensing coded aperture snapshot spectral imaging (CASSI) reconstruct algorithm
摘要
在双色散结构编码孔径光谱成像系统(DD-CASSI)的基础上, 提出了利用多帧采样图方法(Multi Frame-DD-CASSI), 以提高信息采样率, 并提出新的数学模型, 实现了对三维数据立方体的光谱维压缩编码, 充分利用了光谱维数据的相关性, 利用32帧采样图恢复数据, 达到了0.99的光谱曲线相似度和40 db的图像峰值信噪比.
Abstract
Based on dual-disperser coded aperture snapshot spectral imaging(DD-CASSI)system, method using multiple sample image (Multi Frame-DD-CASSI ) was put forward to increase the sampling rate. And a new math model was put forward to realize spectral compression in 3-D data cube. The system makes full use of spectra’s correlation, and 0.99 spectral correlation fitting and 40db image peak signal noise were achieved with 32 frames of sample images.
刘世界, 张旭东, 张月, 李春来, 王建宇. 多帧图像编码孔径光谱成像技术[J]. 红外与毫米波学报, 2017, 36(6): 799. LIU Shi-Jie, ZHANG Xu-Dong, ZHANG Yue, LI Chun-Lai, WANG Jian-Yu. Multi-frame image coded aperture snapshot spectral image technology[J]. Journal of Infrared and Millimeter Waves, 2017, 36(6): 799.