1.Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun 130033, China 2.University of Chinese Academy of Sciences, Beijing 100049, China
Fund Project:Project supported by the National Natural Science Foundation of China (Grant No. 11703024).
Received Date:05 October 2018
Accepted Date:04 February 2019
Available Online:23 March 2019
Published Online:05 April 2019
Abstract:In order to obtain a new imaging strategy of the Fourier telescope (FT) with a better imaging quality and a less imaging time, we optimize and compare three down-sampling imaging strategies in this paper: the compressed sensing method (CS), the low-frequency full sampling method (LF) and the variable-density random sampling method (VD), which are different from the traditional Fourier telescope in both of the image quality and the imaging time. The analytical methods are as follows: based on the target’s spectral data obtained from the field experiment of traditional FT, three down-sampling methods (LF, VD and CS) are used to reconstruct the target’s images according to their own sampling modes and reconstruction methods, respectively; the differences between the three down-sampling methods and the traditional FT regarding the image quality are compared by the instinctive observation and the Strehl ratio; based on the analysis of the imaging time, the differences between the three down-sampling methods and the traditional FT regarding the imaging time are preliminarily compared. The analysis shows that: 1) the image quality of the compressed sensing method is better than that of the other two down-sampling methods (LF and VD), slightly lower than that of the traditional imaging; 2) although the image quality of the compressed sensing method is slightly lower than that of the traditional FT, its imaging time is much lower than that of the traditional FT; 3) the field data used in the analysis contain noises, which means that the reconstruction methods of the above three down-sampling strategies have a better robustness to the noises. Based on the above results, it can be seen that the Fourier telescope based on compressed sensing (CS-FT) is an excellent imaging strategy which can greatly reduce the imaging time in the condition with actual noises. Keywords:synthetic aperture imaging/ Fourier telescope/ image reconstruction techniques/ compressed sensing
令TSR = 0.2, LSR = 0.2. 当OPDF分别为3, 4和5时, 重构结果如图9所示. 因为OPDF只影响采用变密度采样模式的重构方法, 故只有VD-FT和CS-FT受其影响. 从图9可以看出, CS-FT重构性能随着OPDF的增大而略有下降, 故应尽量选择较小的OPDF. 根据前面对OPDF的含义和取值的介绍可知, OPDF越小表示随着频率的增加采样率下降越慢. 这里的分析说明, 实验所采用的目标频谱主要集中在中低频, 高频信息较少. 图 9 OPDF的影响 (a) 和 (b) 分别是OPDF = 3时的VD图像和CS图像; (c) 和 (d) 分别是OPDF = 4时的VD图像和CS图像; (e) 和 (f) 分别是OPDF = 5时的VD图像和CS图像 Figure9. The effect of the OPDF: (a) and (b) are the VD image and the CS image, respectively, when OPDF = 3; (c) and (d) are the VD image and the CS image, respectively, when OPDF = 4; (e) and (f) are the VD image and the CS image, respectively, when OPDF = 5
24.2.LSR影响的分析 -->
4.2.LSR影响的分析
令TSR = 0.4, OPDF = 2.6. 当LSR分别为0, 0.1, 0.3和0.4时, 重构结果如图10所示. 因为LSR只影响采用变密度采样模式的重构方法, 故只有VD-FT和CS-FT受其影响. 从图10可以看出, CS-FT重构性能变化的大致趋势是随着LSR的增大而略有下降, 故应尽量选择较小的LSR. 根据前面对LSR的含义和取值的介绍可知, LSR越小表示等间隔采样的比重越小. 这里的分析说明, 实验所采用的目标频谱的低频信息不多, 信息主要集中在中频部分. 图 10 LSR的影响 (a) 和 (b) 分别是LSR = 0时的VD图像和CS图像; (c) 和 (d) 分别是LSR = 0.1时的VD图像和CS图像; (e) 和 (f) 分别是LSR = 0.3时的VD图像和CS图像; (g) 和 (h) 分别是LSR = 0.4时的VD图像和CS图像 Figure10. The effect of the LSR: (a) and (b) are the VD image and the CS image, respectively, when LSR = 0; (c) and (d) are the VD image and the CS image, respectively, when LSR = 0.1; (e) and (f) are the VD image and the CS image, respectively, when LSR = 0.3; (g) and (h) are the VD image and the CS image, respectively, when LSR = 0.4
24.3.TSR影响的分析 -->
4.3.TSR影响的分析
令LSR = 0, 每次重构尽可能选择最小的OPDF. TSR的变化范围为0.05—0.8, 重构结果如图11所示. 从图11可以看出, CS重构性能随着TSR的增大而提升, 但LF-FT和VD-FT的重构性能也随TSR的增大而提升. 当TSR = 0.8时, 虽然从Strehl比来看, CS-FT仍然大于LF-FT和VD-FT(差别已经很小), 但从直观感受来说, LF-FT, VD-FT和CS-FT的重构效果与N-S-FT几乎无差别. 通过Strhel比和直观观察可以得出, CS-FT的优势主要体现在TSR较小 ( < 0.4) 的时候. 根据前面对TSR的含义和取值的介绍可知, 更小的TSR表示所需的频谱数更少, 成像时间也相应降低. 由此可见, 与其他降采样方法 (LF方法和VD方法) 相比, CS方法在较低成像时间时具有较好的成像质量. 图 11 TSR的影响 (a) 和 (b) 分别是标准图像和传统FT图像; (c), (d) 和 (e) 分别是TSR = 0.05时的LF图像, VD图像和CS图像; (f), (g) 和 (h) 分别是TSR = 0.1时的LF图像, VD图像和CS图像; (i), (j) 和 (k) 分别是TSR = 0.2时的LF图像, VD图像和CS图像; (l), (m) 和 (n) 分别是TSR = 0.4时的LF图像, VD图像和CS图像; (o), (p) 和 (q) 分别是TSR = 0.8时的LF图像, VD图像和CS图像 Figure11. The effect of the TSR: (a) and (b) are the standard image and the traditional FT image, respectively; (c), (d) and (e) are the LF image, the VD image and the CS image, respectively, when TSR = 0.05; (f), (g) and (h) are the LF image, the VD image and the CS image, respectively, when TSR = 0.1; (i), (j) and (k) are the LF image, the VD image and the CS image, respectively, when TSR = 0.2; (l), (m) and (n) are the LF image, the VD image and the CS image, respectively, when TSR = 0.4; (o), (p) and (q) are the LF image, the VD image and the CS image, respectively, when TSR = 0.8