显微热成像系统像素相关度插值微扫描图像重建技术

显微热成像技术根据温度变化对细微目标进行热成像,是国内外光电成像领域重要的研究方向.由于加工技术精度有限,前期研制的光学微扫描显微热成像系统存在扫描位置误差,该误差使基于微扫描图像直接融合形成的图像质量下降.提出一种结合图像预处理思想、微扫描原理和通过计算像素相关度进行微扫描图像插值重建的算法.仿真及实验表明:方法能改...     

光学微扫描显微热成像扫描零点定标方法研究

为改善已研制光学微扫描显微热成像系统的空间分辩力,微扫描零点需要确定.基于几何原理,研究提出了一种利用数字图像处理技术进行零点定标的方法.给出了微扫描零点的定义、详细分析了零点定标原理及方法,完成了实际显微热成像系统的微扫描零点定标.针对红外热图像,模拟零点定标前后的实际系统,采用不同重构方法进行了仿真研究,给出了评价参数;利用零点定标后的光学微扫描显微热成像系统采集低分辨力显微热图像序列进行过采样重构研究,仿真和实验结果表明了该方法的有效性,从而得到了高分辨力光学微扫描显微热成像系统,可应用于需要显微热分析的场合.     

基于局部梯度插值的显微热成像系统微扫描误差修正技术

显微热成像系统可观测、记录分析细微目标的温度变化过程,在需要细微热分析的诸多方面有着广泛的发展前景。由于设备加工及工作过程中存在误差,影响微扫描系统的精度,使得微扫描系统扫描过程中偏离标准位置,故采集得到的四幅低分辨力图像会存在误差,最终影响显微热成像系统高分辨力图像的重建质量。为尽可能降低微扫描误差,文章提出了基于局部梯度插值与预处理相结合的微扫描误差修正技术,通过进行模拟仿真和实验证实该技术可以降低系统微扫描误差,提高系统的空间分辨力。     

显微热成像系统自适应位置标定方法

利用光学微扫描技术,可在不改变探测器结构的情况下,提高显微热成像系统空间分辨率。但为了获得高质量过采样重构图像,微扫描位置需要标定。基于零点定标,提出一种各点自适应定标的微扫描位置标定方法。模拟实际系统在定标前后的欠采样图像,采用不同重构方法进行仿真对比验证;完成了实际采集的欠采样显微热图像序列的重构对比实验。实验结果表明该方法明显改善了显微热成像系统的过采样重构图像质量,提高了系统空间分辨率。此方法还可以应用在其他光电成像系统中。     

基于不可控微扫描的高分辨力图像重构方法

针对存在不可控微位移的序列低分辨力图像,提出了一种基于2×2不可控微扫描的高分辨力图像重构方法,采用投影法估算出低分辨力图像LR的微位移量,采用基于泰勒级数展开的重构算法由4帧图像重构高分辨力图像.模拟表明:该算法能精确地估计序列图像的帧间位移量并取得较好的重构效果,且算法简单有效可行,处理量小,有利于实现快速处理,在一定程度可满足同步轨道卫星、公安监视等领域对高分辨力成像处理的需求.     

基于不可控微扫描的高分辨力图像重构方法

针对存在不可控微位移的序列低分辨力图像,提出了一种基于2×2不可控微扫描的高分辨力图像重构方法,采用投影法估算出低分辨力图像LR的微位移量,采用基于泰勒级数展开的重构算法由4帧图像重构高分辨力图像.模拟表明：该算法能精确地估计序列图像的帧间位移量并取得较好的重构效果,且算法简单有效可行,处理量小,有利于实现快速处理,在一定程度可满足同步轨道卫星、公安监视等领域对高分辨力成像处理的需求.       

基于微球透镜的任选区高分辨光学显微成像新方法研究

提出和发展了基于毛细管-微球组合探针的任选区、 高分辨显微成像新方法. 建立了微球显微成像的物理模型, 利用成像理论,推导出微球成像的放大倍率; 采用3.0, 4.4, 5.6, 7.5, 10.0 μm等不同直径的SiO₂微球, 对未经刻录的DVD光盘进行了微球显微成像实验, 可以观察到DVD光盘的微纳米结构被明显放大且对比度显著提高, 与理论计算结果相符合; 采用毛细管微探针操纵微球的方法, 实现了基于微球透镜阵列的样品微纳米结构的高分辨显微成像; 在此基础上, 进一步将毛细管微探针与微球组合, 制备出毛细管-微球组合型探针, 首次实现了基于微球透镜的样品任意区域高分辨显微成像.     

扫描型亚像元成像处理对热成像系统性能改善的模拟分析

基于光学微扫描的亚像元成像技术是当前国内外迅速发展的一种高性能成像技术,它对探测器单元数较少且探测器单元尺寸较大的热成像系统更为有效。以扫描型焦平面探测器热成像系统FC和GP为研究对象,采用热成像系统性能模拟软件包CFLIR4.0,分析了光学系统的相对孔径、过采样和亚像元成像等对热成像系统性能的影响,包括传递函数MTF、噪声等效温差NETD、最小可分辨温差MRTD以及系统的探测和识别距离。模拟分析表明:过采样成像可提高热成像系统的性能,亚像元成像处理可进一步提高热成像系统的性能,可明显地提高系统的作用距离。研究结果对亚像元热成像技术的发展具有参考意义。     

光学微操纵过程的轴平面显微成像技术

光学俘获技术利用光与物质相互作用产生的光势阱效应来实现对微粒的操控,已经成功应用于生物医学、材料科学等交叉领域.在对微粒进行三维俘获时,传统的宽场光学显微技术只能观测到某一平面内微粒的横向运动,对微粒沿轴向运动的观测受到很大限制.本文将轴平面显微成像技术引入光学微粒操控研究中,利用45?倾斜的反射镜把微粒的轴向运动信息转换到横向平面进行观测,与传统宽场显微成像技术相结合,实现了对二氧化硅小球俘获过程横向和轴向运动的同步观测.该成像方法无需扫描和数据重构,具有实时快速等优点,在新型光束光镊、厚样品三维观测和成像等领域具有潜在的应用价值.     

基于等角度与等时间采样技术扫描成像系统

为满足扫描成像系统对图象配准精度,从理论上分析了在等时间采样方式下,遥感仪器对扫描系统速度稳定度的要求.探讨了等角度这一新型采样方式应用于未来遥感仪器的可行性.并将这两种采样方式成功应用于自行研制的短波红外成像仪.扫描图象表明等角度采样可以避免等时采样所产生的图象畸变.     

基于分块平滑投影二次重构算法的单像素成像系统

单像素成像系统是通过无空间分辨能力的单像元探测器来获取目标二维分布信息的计算光学成像技术,与传统直接成像技术相比具有高能量收集效率、高灵敏度等一系列优点,在高能物理诊断技术领域有着广阔的应用前景。针对实际单像素压缩感知成像系统在复杂诊断环境中存在的重建噪声较大的问题,提出并实现了基于分块平滑投影Landweber二次重构算法的单像素成像系统。根据算法观测矩阵分布特性以及数字微镜硬件输入要求实现了实际投影观测矩阵的变换,利用二次重构算法实现了单像素诊断的仿真分析与实验测试。仿真结果表明,在采样率为20%～30%的条件下,重建图像峰值信噪比大于20 dB,结构相似性高于0.8。进一步搭建单像素成像平台完成实验研究及验证,实验结果表明,利用二次重构算法模型对目标场景进行恢复的效果优于其余两种传统算法。二次重构单像素成像系统在采样率仅为20%的条件下能够重建出清晰的原始图像,具有较好的噪声抑制特性。     

High-resolution photoacoustic microscope for rat brain imaging in vivo

<正>A reflection-mode photoacoustic microscope(PAM) for rat brain imaging in vivo is constructed.A pulsed laser is used as an excitation source,and a focused ultrasound transducer is adopted to collect the photoacoustic signal.Raster scanning is applied to acquire three-dimensional(3D) data.The obtained measurements of the lateral and axial resolutions of the microscope are 45 and 15μm,respectively.The imaging depth in the chicken breast tissue is 3.1 mm at a signal-to-noise ratio(SNR) of 20 dB without any signal averaging.The imaging speed is 30 A-line/s.Experimental results in vivo demonstrate the capability of 3D imaging of the brain vessels of the rat after removing the skull.     

三维混合域重建算法在太赫兹全息成像中的应用

提出了一种三维混合域重建算法,该算法在垂直方位向采用空间波数域匹配滤波,结合距离水平方位域后向投影技术完成目标三维重建。与三维波数域算法相比,该算法不仅避免了复杂的三维Stolt插值,还继承了后向投影技术便于处理运动补偿的优点。相比传统的三维时域后向投影算法,该算法通过在垂直方位向的波数域处理,运算量显著减少。通过对0.2 THz波段仿真和实验数据的重建,验证了算法的正确性和有效性。     

三维混合域重建算法在太赫兹全息成像中的应用

提出了一种三维混合域重建算法,该算法在垂直方位向采用空间波数域匹配滤波,结合距离水平方位域后向投影技术完成目标三维重建。与三维波数域算法相比,该算法不仅避免了复杂的三维Stolt插值,还继承了后向投影技术便于处理运动补偿的优点。相比传统的三维时域后向投影算法,该算法通过在垂直方位向的波数域处理,运算量显著减少。通过对0.2 THz波段仿真和实验数据的重建,验证了算法的正确性和有效性。     

Evaluation of imaging protocol for ECT based on CS image reconstruction algorithm

Single-photon emission computerized tomography and positron emission tomography are essential med- ical imaging tools, for which the sampling angle number and scan time should be carefully chosen to give a good compromise between image quality and radiopharmaceutical dose. In this study, the image quality of different ac- quisition protocols was evaluated via varied angle number and count number per angle with Monte Carlo simulation data. It was shown that, when similar imaging counts were used, the factor of acquisition counts was more important than that of the sampling number in emission computerized tomography. To further reduce the activity requirement and the scan duration, an iterative image reconstruction algorithm for limited-view and low-dose tomography based on compressed sensing theory has been developed. The total variation regulation was added to the reconstruction process to improve the signal to noise Ratio and reduce artifacts caused by the limited angle sampling. Maximization of the maximum likelihood of the estimated image and the measured data and minimization of the total variation of the image are alternatively implemented. By using this advanced algorithm, the reconstruction process is able to achieve image quality matching or exceed that of normal scans with only half of the injection radiopharmaceutical dose.     

Evaluation of imaging protocol for ECT based on CS image reconstruction algorithm

Single-photon emission computerized tomography and positron emission tomography are essential medical imaging tools, for which the sampling angle number and scan time should be carefully chosen to give a good compromise between image quality and radiopharmaceutical dose. In this study, the image quality of different acquisition protocols was evaluated via varied angle number and count number per angle with Monte Carlo simulation data. It was shown that, when similar imaging counts were used, the factor of acquisition counts was more important than that of the sampling number in emission computerized tomography. To further reduce the activity requirement and the scan duration, an iterative image reconstruction algorithm for limited-view and low-dose tomography based on compressed sensing theory has been developed. The total variation regulation was added to the reconstruction process to improve the signal to noise Ratio and reduce artifacts caused by the limited angle sampling. Maximization of the maximum likelihood of the estimated image and the measured data and minimization of the total variation of the image are alternatively implemented. By using this advanced algorithm, the reconstruction process is able to achieve image quality matching or exceed that of normal scans with only half of the injection radiopharmaceutical dose.     

基于模型选择的模式波前重构算法研究

基于Zernike模式的波前重构算法通常忽略实际波前像差构成的差异, 而用一定数量的低阶Zernike模式进行波前重构, 导致模式混淆或耦合等问题, 进而影响波前重构的精度. 根据信息论中的最小描述长度准则对重构模型的阶数进行了选择, 在此基础上应用非线性优化算法计算重构系数, 并最终实现波前重构; 对不同信噪比条件下振幅均匀分布入射光束的波前进行了重构. 结果表明: 该算法不但能够实现相对于现有算法相对较高的波前重构精度, 并且具有优良的噪声适应性, 体现了模型选择在模式法波前重构算法中的意义.     

Iterative reconstruction algorithm for optoacoustic imaging

Optoacoustic imaging is based on the generation of thermoelastic stress waves by heating an object in an optically heterogeneous medium with a short laser pulse. The stress waves contain information about the distribution of structures with preferential optical absorption. Detection of the waves with an array of broadband ultrasound detectors at the surface of the medium and applying a backprojection algorithm is used to create a map of absorbed energy inside the medium. With conventional reconstruction methods a large number of detector elements and filtering of the signals are necessary to reduce backprojection artifacts. As an alternative this study proposes an iterative procedure. The algorithm is designed to minimize the error between measured signals and signals calculated from the reconstructed image. In experiments using broadband optical ultrasound detectors and in simulations the algorithm was used to obtain three-dimensional images of multiple optoacoustic sources. With signals from a planar array of 3x3 detector elements a significant improvement was observed after about 10 iterations compared to the simple radial backprojection. Compared to conventional methods using filtered backprojection, the iterative method is computationally more intensive but requires less time and instrumentation for signal acquisition.     

四光束剪切相干成像目标重构算法研究

针对传统剪切光束成像技术的准实时性问题,提出用口字形排布的四束光代替传统L形三束剪切光照射目标,研究了四光束剪切相干成像目标重构算法.只需单次测量就能同时重构出四幅目标图像,减少了用于降低散斑噪声、获取高质量图像所需的测量次数,同时大大减少了多组发射时的光束切换次数,提高了成像效率.在算法实现中,通过最小二乘法恢复出四组波前相位,利用散斑幅值的简单代数运算恢复波前幅值,从而重构出目标图像.仿真结果表明,与传统方法相比,在图像质量相同的前提下,本文方法所需的数据采集时间减少了至少1/2,不但提高了目标重构效率,还可为远程运动目标的成像识别提供更好的手段.