1 北京工业大学信息学部计算智能与智能系统北京市重点实验室，北京 100124
2 先进信息网络北京实验室，北京 100124
Cherenkov激发的荧光扫描成像（CELSI）是一种新型的光学成像技术，为监测体内恶性肿瘤的生物学特性提供了一种手段。为提高CELSI图像重建质量，本文提出了一种基于迭代优化展开的深度学习图像重建算法——ADMM-Net。在该算法中，交替方向乘子法（ADMM）与卷积神经网络（CNN）相结合组成一个深度网络，网络中的所有参数通过端到端训练进行学习。实验结果表明：该算法可以有效提升重建图像的质量。当网络层数为5时，该算法重建的单荧光目标图像的平均峰值信噪比和结构相似性值分别可达到33.75 dB和0.86。该算法不仅可以分辨出边沿距离最小为2 mm的双荧光目标，而且在多荧光目标和不同荧光量子产额比率下表现出了良好的泛化能力。
医用光学 生物技术 Cherenkov激发的荧光扫描成像 图像重建技术 交替方向乘子法 深度学习 优化展开 中国激光
2023, 50(15): 1507106
1 华北电力大学电子与通信工程系，河北 保定 071003
2 华北电力大学河北省电力物联网技术重点实验室，河北 保定 071003
在光声内窥成像中，不均匀或不稳定的照明，以及生物组织的复杂光学特性均会导致成像平面内光通量分布不均匀，从而造成重建图像质量和成像深度的下降。提出了一种校正光通量变化的定量光声内窥成像方法，对光吸收能量分布进行稀疏分解，采用贪婪算法重构得到光吸收系数和光通量的稀疏表示，并通过稀疏矩阵分解实现光吸收系数与光通量分布的联合重建。仿真和体模实验结果表明，与一步法和基于模型的定量重建方法相比，采用所提方法估算光吸收系数的均方根误差（RMSE）可降低约48%，重建图像的归一化平均绝对距离（NMSAD）和结构相似度（SSIM）可分别降低约25%和提高约24%。与其他校正光通量的重建算法相比，所提方法估计光吸收系数的RMSE可降低约22%、NMSAD可降低约20%、SSIM可提高约10%。
成像系统 图像重建技术 光声成像 内窥成像 光吸收系数 光通量 定量成像
湖南大学信息科学与工程学院, 湖南 长沙 410082
风导致的信道气流是影响关联成像的重要因素,因此,对信道气流干扰下的关联成像研究进行了总结。首先,给出了近场气流影响的相位模型,并从光传输和关联成像两个角度进行了可靠性验证;然后,将该模型扩展到高风速区域,得到超声速气流下风速和边界层厚度对关联成像的影响规律,定量分析了成像质量的变化情况;最后,针对实际成像过程中的探测抖动问题,介绍了基于关联成像时间特性的抑制方法以及小样本成像算法。本研究结果不仅可以评估信道气流对关联成像的影响,还为关联成像在机载遥感等领域的应用提供了重要的参考价值。
量子光学 量子信息与处理 湍流 图像恢复技术 近场气流 关联成像 激光与光电子学进展
2021, 58(10): 1011017
1 北京航空航天大学机械工程及自动化学院, 北京 100191
2 中国工程物理研究院核物理与化学研究所, 四川 绵阳 621900
径向边缘的伪影是外部CT图像明显存在的缺点。为了抑制外部CT图像径向边缘的伪影,采用加权方向全变差(WDTV)算法计算沿径向和切向方向的局部方向差分,引入两个权重参数对这两个局部方向差分进行加权求和。WDTV算法可以更好地描述外部CT图像梯度模的稀疏性,有效提升重建图像的质量。针对高噪声中子外部CT检测的需求,在WDTV重建模型的框架下,对沿着径向和切向两个方向附近的一定角度范围内分别施加不同的全变差(TV)最小化约束。改进的WDTV算法中TV最小化的作用更明显,具有更强地抑制径向边缘伪影与抗噪声的性能。计算机仿真结果和齿轮冷中子实验结果表明,改进的WDTV重建模型能够更有效地抑制重建图像的噪声,提高重建图像的质量。
图像处理 图像重建技术 中子束 外部计算机断层成像 伪影抑制 光学学报
2020, 40(22): 2210001
1 北京工业大学信息学部计算智能与智能系统北京市重点实验室, 北京 100124
2 北京先进信息网络实验室, 北京 100124
切伦科夫激发的荧光扫描成像(CELSI)作为一种新兴分子成像技术,具有空间分辨率高和成像深度深的优点,在监测放疗过程中肿瘤的生理变化方面具有巨大潜力。前期工作基于Tikhonov方法成功实现了CELSI断层成像,但该方法无法对位置深度超过3 cm或低对比度的荧光目标进行准确重建。为克服这一问题,提出了一种基于近似信息传递算法的断层CELSI稀疏重建方法。为说明该算法的优点,将其与传统的Tikhonov正则化算法以及3种基于稀疏的重建算法进行比较。实验结果表明,就均方误差和对比噪声比而言,本文算法可以获得最优的重建结果。
医用光学 图像重建技术 切伦科夫激发的荧光扫描成像 断层成像 近似消息传递 稀疏重建
Author Affiliations
Abstract
1 Shanghai Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Shanghai 201800, China
2 Center of Materials Science and Optoelectronics Engineering, University of Chinese Academy of Sciences, Beijing 100049, China
In our Letter, two kinds of handwriting traces, colored and colorless, are studied by means of reflectance transformation imaging. The illumination direction and rendering mode can be changed alternatively to obtain two-dimensional and three-dimensional details of the traces that are not recognized easily by naked eyes. Furthermore, an objective evaluation method without reference is applied to evaluate the reconstructed images, which provides a basis for setting the illumination direction and rendering mode. Therefore, the handwriting trace information including the written content, the writing features, and the stroke order features can be obtained objectively and accurately.
110.3010 Image reconstruction techniques 120.6650 Surface measurements, figure 330.1715 Color, rendering and metamerism Chinese Optics Letters
2019, 17(11): 111101
1 中国工程物理研究院流体物理研究所, 四川 绵阳621900
2 哈尔滨工业大学 自动化测控系, 黑龙江 哈尔滨 150080
傅里叶叠层成像是一种能够同时实现大视场和高分辨的成像方法, 公开发表的文献表明其空间分辨率极限由照明数值孔径和物镜数值孔径决定。为了进一步提高其分辨率, 提出了频域和空间约束傅里叶叠层重建方法: 利用传统重建算法获得的空间频谱进行频域约束, 以传统重建算法获得的图像进行空间约束; 该方法基于一个假设: 图像具有稀疏特性; 从传统重建算法获得的图像中提取所需的频域和空间约束条件, 不需要额外采集数据和硬件改进。仿真和实验结果表明: 与传统无约束重建方法相比, 提出的算法能够提高分辨率和改善对比度, 空间分辨率提高幅度高达~26%。
图像重建技术 计算成像 显微 image reconstruction techniques computational imaging microscopy 红外与激光工程
2019, 48(4): 0422003
Author Affiliations
Abstract
1 School of Mechatronic Engineering and Automation, Shanghai University, Shanghai 200072, China
2 Shanghai Key Laboratory of Intelligent Manufacturing and Robotics, Shanghai University, Shanghai 200072, China
3 Mechanical and Electrical Engineering School, Shenzhen Polytechnic, Shenzhen 518055, China
Depth from focus (DFF) is a technique for estimating the depth and three-dimensional (3D) shape of an object from a multi-focus image sequence. At present, focus evaluation algorithms based on DFF technology will always cause inaccuracies in deep map recovery from image focus. There are two main reasons behind this issue. The first is that the window size of the focus evaluation operator has been fixed. Therefore, for some pixels, enough neighbor information cannot be covered in a fixed window and is easily disturbed by noise, which results in distortion of the model. For other pixels, the fixed window is too large, which increases the computational burden. The second is the level of difficulty to get the full focus pixels, even though the focus evaluation calculation in the actual calculation process has been completed. In order to overcome these problems, an adaptive window iteration algorithm is proposed to enhance image focus for accurate depth estimation. This algorithm will automatically adjust the window size based on gray differences in a window that aims to solve the fixed window problem. Besides that, it will also iterate evaluation values to enhance the focus evaluation of each pixel. Comparative analysis of the evaluation indicators and model quality has shown the effectiveness of the proposed adaptive window iteration algorithm.
100.6890 Three-dimensional image processing 100.3010 Image reconstruction techniques 100.2980 Image enhancement Chinese Optics Letters
2019, 17(6): 061001
Author Affiliations
Abstract
School of Mechatronic Systems Engineering, Simon Fraser University, 250–13450 102nd Avenue, Surrey, Canada, BC V3T 0A3
Diffuse optical spectroscopy is a relatively new, noninvasive and nonionizing technique for breast cancer diagnosis. In the present study, we have introduced a novel handheld diffuse optical breast scan (DOB-Scan) probe to measure optical properties of the breast in vivo and create functional and compositional images of the tissue. In addition, the probe gives more information about breast tissue’s constituents, which helps distinguish a healthy and cancerous tissue. Two symmetrical light sources, each including four different wavelengths, are used to illuminate the breast tissue. A high-resolution linear array detector measures the intensity of the back-scattered photons at different radial destinations from the illumination sources on the surface of the breast tissue, and a unique image reconstruction algorithm is used to create four cross-sectional images for four different wavelengths. Different from fiber optic-based illumination techniques, the proposed method in this paper integrates multi-wavelength light-emitting diodes to act as pencil beam sources into a scattering medium like breast tissue. This unique design and its compact structure reduce the complexity, size and cost of a potential probe. Although the introduced technique miniaturizes the probe, this study points to the reliability of this technique in the phantom study and clinical breast imaging. We have received ethical approval to test the DOB-Scan probe on patients and we are currently testing the DOB-Scan probe on subjects who are diagnosed with breast cancer.
Breast cancer diffuse optical spectroscopy image reconstruction techniques medical and biological imaging optical breast phantom Journal of Innovative Optical Health Sciences
2019, 12(2): 1950008
Author Affiliations
Abstract
1 University of Hong Kong, Department of Electrical and Electronic Engineering, Pokfulam, Hong Kong, China
2 Northwestern Polytechnical University, School of Natural and Applied Sciences, Xi’an, China
3 SharpSight Limited, Hong Kong, China
Digital holography records the entire wavefront of an object, including amplitude and phase. To reconstruct the object numerically, we can backpropagate the hologram with Fresnel–Kirchhoff integral-based algorithms such as the angular spectrum method and the convolution method. Although effective, these techniques require prior knowledge, such as the object distance, the incident angle between the two beams, and the source wavelength. Undesirable zero-order and twin images have to be removed by an additional filtering operation, which is usually manual and consumes more time in off-axis configuration. In addition, for phase imaging, the phase aberration has to be compensated, and subsequently an unwrapping step is needed to recover the true object thickness. The former either requires additional hardware or strong assumptions, whereas the phase unwrapping algorithms are often sensitive to noise and distortion. Furthermore, for a multisectional object, an all-in-focus image and depth map are desired for many applications, but current approaches tend to be computationally demanding. We propose an end-to-end deep learning framework, called a holographic reconstruction network, to tackle these holographic reconstruction problems. Through this data-driven approach, we show that it is possible to reconstruct a noise-free image that does not require any prior knowledge and can handle phase imaging as well as depth map generation.
digital holography computational imaging image reconstruction techniques machine learning deep learning Advanced Photonics
2019, 1(1): 016004
