苏州大学光电科学与工程学院,江苏 苏州 215006
通过测量矩阵获取Gram矩阵,梳理了Gram矩阵与系统点扩散函数的关系,进而基于点扩散函数提出最强旁瓣峰值大小、叠加旁瓣峰值大小、空间距离和频谱余弦相似度4个特征参量。在此基础上,构建了一种单像素压缩成像高质量图像重建的特征函数,建立了可重建的目标稀疏度与特征函数的关系,并通过数值模拟和实验验证了所提特征函数的有效性,该工作对于单像素成像系统测量矩阵的优化设计具有重要借鉴意义。
成像系统 单像素成像 压缩感知 测量矩阵 特征函数
图像压缩感知是一种在欠采样条件下尽可能重构原始图像的技术。为解决大部分基于卷积神经网络(CNN)框架的图像压缩感知方法容易受到卷积感受野的限制、对全局信息的关注较少的问题, 提出了基于Swin Transformer的图像压缩感知重构网络。网络使用卷积层对图像进行采样, 然后使用自注意力机制和残差结构结合的残差Swin Transformer组(RSTG)结构来关注图像的细节。实验结果表明, 基于Swin Transformer的图像压缩感知重构网络可以充分利用图像的先验信息, 进一步提高图像压缩感知的重构精度, 并获得比其他压缩感知方法更好的重构性能和视觉效果。
压缩感知 图像重构 自注意力机制 残差 compressive sensing image reconstruction self-attention mechanism residual
1 北京理工大学光电学院光电成像技术与系统教育部重点实验室,北京 100081
2 北京理工大学重庆创新中心,重庆 401120
3 北方自动控制技术研究所军种指控系统研发部,山西 太原 030006
4 北京印刷学院印刷与包装工程学院,北京 102600
压缩感知高光谱计算成像技术是当前高光谱计算成像领域的研究热点之一,其能够在保持系统元器件物理特性不变的前提下,有效地提升成像质量。本文概述了高光谱计算成像的研究背景和基本概念,详细介绍了压缩感知高光谱计算成像系统的发展现状,重点阐述了本团队提出的基于空-谱编码的压缩感知高光谱计算成像技术,并对其系统组成、数理模型以及最新进展进行了说明。通过总结压缩感知高光谱计算成像的背景知识以及空-谱编码压缩感知高光谱计算成像的研究工作,力求为科研人员探索压缩感知高光谱计算成像新体制带来新的思路,促进高光谱计算成像技术的发展。
成像系统 高光谱成像 计算成像 压缩感知 编码技术 光学学报
2023, 43(15): 1511003
1 重庆大学 微电子与通信工程学院,重庆400044
2 安徽大学 物质科学与信息技术研究院,安徽合肥30039
从极少量的测量值中有效且高概率高质量恢复出原始信号是压缩感知图像重建研究的核心问题,学者们相继提出了传统和基于深度学习的压缩感知图像重建算法,传统算法通常基于优化模型迭代求解,重建质量和重建速度都无法保证;基于深度学习的算法重建质量相对较高,但缺乏物理可解释性。受滤波流的启发,本文提出了联合全局与局部的深度压缩感知图像重建模型(G2LNet),其以卷积层执行压缩采样以及初始重建过程,利用快速傅里叶卷积与滤波流,同时考虑了图像全局上下文信息和图像像素局部邻域信息,联合学习优化测量矩阵与滤波流,建立了完整的端到端可训练的深度图像压缩感知重建网络。经实验验证,在压缩感知图像重建领域常用的Set5,Set11,BSD68测试集上取得了良好的重建效果,在采样率为20%的情况下,G2LNet的图像重建质量相比于经典的传统算法MH与基于深度学习的算法CSNet的平均PSNR分别提高了2.29 dB,0.51 dB,有效提升了重建图像质量。
压缩感知 图像重建 快速傅里叶卷积 滤波流 深度神经网络 compressive sensing image reconstruction fast Fourier convolution filter flow deep neural network 光学 精密工程
2023, 31(14): 2135
1 国防科技大学电子对抗学院脉冲功率激光技术国家重点实验室,安徽 合肥 230037
2 国防科技大学电子对抗学院电子制约技术安徽省重点实验室,安徽 合肥 230037
3 95438部队,四川 眉山 620000
变速运动目标的中频信号特征频谱具有连续集中的多频分量,且具有一定的多普勒展宽,在背景噪声和暗计数的影响下,光子回波外差信号的信噪比较低时,使用传统的信号处理方法得到的中频信号频谱以及时频分析特性效果较差。为提高信噪比,本文提出了将稀疏度自适应压缩感知和密度聚类相结合的信号处理方法,并采用该方法对变速目标的光子回波外差信号进行处理。该信号处理方法解决了变速目标频谱稀疏度K无法提前确定的问题,而且只需要较少的观测数据就可以重构信噪比较高的中频频谱。此外,该方法结合密度聚类算法对中频频谱进行了第二重去噪,大幅度减少了噪声分量。研究结果表明,该信号处理方法能够将信噪比提高一定幅度且多普勒展宽精度误差在10%以内,可以得到较为完整的重构中频信号频谱,同时较好地解决了信号时频分辨率较差的问题以及单光子探测等间隔时间序列造成的时频图中的倍频现象,得到了更好的时频特性描述。
光谱学 光子外差 压缩感知 聚类去噪 多普勒测速 时频分析 中国激光
2023, 50(10): 1011002
1 苏州市轨道交通集团有限公司, 江苏 苏州 215008
2 中国电科芯片技术研究院, 重庆 400060
针对光纤布拉格光栅(FBG)传感信号易受外界噪声干扰从而导致信号丢失的问题, 提出了一种改进型正交匹配追踪(OMP)算法。围绕FBG传感信号波长随应力漂移的本质特征, 在压缩感知理论的框架下, 通过去除稀疏系数中的虚部, 并利用指数饱和法对非零元素进行拟合与排序, 从而获取FBG信号的有效稀疏度。在此基础上, 通过改进经典OMP算法迭代过程中的原子选择策略与终止条件, 有效降低算法复杂度并提高信号的重构精度。对比实验结果表明, 所提出的算法在时间复杂度、信噪比与信号重构精度等方面均具有突出的优势。
光纤布拉格光栅 正交匹配追踪 压缩感知 稀疏度 fiber Bragg grating orthogonal matching pursuit compressive sensing sparsity
Author Affiliations
Abstract
1 East China Normal University, School of Physics and Electronic Science, State Key Laboratory of Precision Spectroscopy, Shanghai, China
2 Shenzhen University, Institute of Microscale Optoelectronics, Nanophotonics Research Center, Shenzhen Key Laboratory of Micro-Scale Optical Information Technology, Shenzhen, China
3 Peking University, School of Physics, Frontiers Science Center for Nanooptoelectronics, State Key Laboratory for Mesoscopic Physics, Beijing, China
4 Shanxi University, Collaborative Innovation Center of Extreme Optics, Taiyuan, China
Various super-resolution microscopy techniques have been presented to explore fine structures of biological specimens. However, the super-resolution capability is often achieved at the expense of reducing imaging speed by either point scanning or multiframe computation. The contradiction between spatial resolution and imaging speed seriously hampers the observation of high-speed dynamics of fine structures. To overcome this contradiction, here we propose and demonstrate a temporal compressive super-resolution microscopy (TCSRM) technique. This technique is to merge an enhanced temporal compressive microscopy and a deep-learning-based super-resolution image reconstruction, where the enhanced temporal compressive microscopy is utilized to improve the imaging speed, and the deep-learning-based super-resolution image reconstruction is used to realize the resolution enhancement. The high-speed super-resolution imaging ability of TCSRM with a frame rate of 1200 frames per second (fps) and spatial resolution of 100 nm is experimentally demonstrated by capturing the flowing fluorescent beads in microfluidic chip. Given the outstanding imaging performance with high-speed super-resolution, TCSRM provides a desired tool for the studies of high-speed dynamical behaviors in fine structures, especially in the biomedical field.
super-resolution microscopy high-speed imaging compressive sensing deep learning image reconstruction Advanced Photonics
2023, 5(2): 026003
Author Affiliations
Abstract
1 School of Engineering, Monash University Malaysia, Selangor 47500, Malaysia
2 College of Optoelectronic Engineering, Changchun University of Science and Technology, Jilin 130022, China
The detection and reconstruction of transparent objects have remained challenging due to the absence of their features and variations in the local features with variations in illumination. In this paper, both compressive sensing (CS) and super-resolution convolutional neural network (SRCNN) techniques are combined to capture transparent objects. With the proposed method, the transparent object’s details are extracted accurately using a single pixel detector during the surface reconstruction. The resultant images obtained from the experimental setup are low in quality due to speckles and deformations on the object. However, the implemented SRCNN algorithm has obviated the mentioned drawbacks and reconstructed images visually plausibly. The developed algorithm locates the deformities in the resultant images and improves the image quality. Additionally, the inclusion of compressive sensing minimizes the measurements required for reconstruction, thereby reducing image post-processing and hardware requirements during network training. The result obtained indicates that the visual quality of the reconstructed images has increased from a structural similarity index (SSIM) value of 0.2 to 0.53. In this work, we demonstrate the efficiency of the proposed method in imaging and reconstructing transparent objects with the application of a compressive single pixel imaging technique and improving the image quality to a satisfactory level using the SRCNN algorithm.
Transparent object imaging single-pixel imaging compressive sensing total-variation minimization SRCNN algorithm Photonic Sensors
2022, 12(4): 220413
光子学报
2022, 51(11): 1109001