Compression of fluorescence microscopy images based on the signal-to-noise estimation

Modern microscopic techniques like high-content screening (HCS), high-throughput screening, 4D imaging, and multispectral imaging may involve collection of thousands of images per experiment. Efficient image-compression techniques are indispensable to manage these vast amounts of data. This goal is frequently achieved using lossy compression algorithms such as JPEG and JPEG2000. However, these algorithms are optimized to preserve visual quality but not necessarily the integrity of the scientific data, which are often analyzed in an automated manner. Here, we propose three observer-independent compression algorithms, designed to preserve information contained in the images. These algorithms were constructed using signal-to-noise ratio (SNR) computed from a single image as a quality measure to establish which image components may be discarded. The compression efficiency was measured as a function of image brightness and SNR. The alterations introduced by compression in biological images were estimated using brightness histograms (earth's mover distance (EMD) algorithm) and textures (Haralick parameters). Furthermore, a microscope test pattern was used to assess the effect of compression on the effective resolution of microscope images.     

二维小波变换在JPEG2000图像编码中的应用

为研究小波变换的性能,本文运用Matlab编程语言对基于二维小波变换的JPEG2000图像编码器进行分层压缩仿真,测试分解层数与量化步长对图像编码的影响,和JPEG2000中不同小波基的选择对图像压缩效果的影响。实验结果表明小波的分层压缩可以获得任意压缩比的图像,要根据图像类型和所需压缩比选用小波基。实验结果对小波变换在图像编码领域的应用有理论价值及实际意义。     

面向深空探测Bayer图像的高效编码

本文提出一种基于JPEG2000标准的Bayer图像高性能RBCR（Remove Bayer Component Relation, RBCR）压缩算法。在RBCR压缩算法中,根据Bayer图像相关性较高的特点,对Bayer图像进行颜色分量分离,得到处理单元子图,对各子图进行1*4整型离散余弦变换,降低Bayer图像中各颜色分量空间域内的相关性;对变换后的各分量DCT系数使用标准JPEG2000算法独立完成小波变换、Tier1编码、MQ编码和率失真斜率计算等,再基于率失真斜率联合截取方法完成各个分量的码流截取,即使用相同的率失真斜率门限值,按照率失真斜率值由高到低的顺序依次完成所有分量编码码块的码流截取,最后各个分量的截取结果再进行独立的码流组织输出。在RBCR算法中通过加入DCT变换降低Bayer图像相关性和对各个分量码流的率失真斜率联合截取,提高恢复图像质量且精确控制了码率。实验结果表明,RBCR算法与各个分量独立压缩方法相比,恢复图像质量得到提升,尤其在4倍的压缩倍数下效果最佳,峰值信噪比平均提高1.814dB,复杂峰值信噪比平均提高2.414dB,可以满足深空探测低复杂度和高质量图像的要求。     

数字图像压缩技术的研究及进展

数字图像压缩技术对于数字图像信息在网络上实现快速传输和实时处理具有重要的意义。本文介绍了当前几种最为重要的图像压缩算法：JPEG、JPEG2000、分形图像压缩和小波变换图像压缩，总结了它们的优缺点及发展前景。然后简介了任意形状可视对象编码算法的研究现状，并指出此算法是一种产生高压缩比的图像压缩算法。     

基于JPEG2000的图像加密算法

随着Internet技术与多媒体技术的飞速发展,对多媒体数据的存储及安全传输也提出了较高的要求.而传统的图像加密技术只是将其作为普通数据流进行加密,没有考虑多媒体数据的特点和数据压缩的重要性,因此有一定的局限性.提出了基于JPEG2000的图像加密算法,算法核心是将加密过程和解密过程嵌入到JPEG2000的编码和解码过程中,并且只选取小波分解后的低频和高频系数加密,加密算法采用混沌猫映射.因此其保密性较高,并且对部分数据加密也大大降低了算法的时间复杂度,提高了算法的系统的执行效率.     

星载遥感图像JPEG2000压缩算法去量化处理方法

随着空间技术的发展,空间遥感图像的数据量呈几何级数增加。为了满足JPEG2000星载遥感图像压缩系统的实时性要求,不仅需要设计高速处理硬件系统,还需要对压缩算法进行优化和改进。分析了离散小波变换后拉伸提升算法和JPEG2000量化算法,提出了一种基于离散小波变换后拉伸提升算法的去量化处理方法,该方法可去除JPEG2000压缩算法中的量化过程,减少压缩算法的运算量。介绍了该方法的基本原理和定点实现。实验结果表明,应用该方法可以使JPEG2000压缩算法中拉伸和量化过程实现的时间降低50％。     

无线传感网基于邻居簇的JPEG2000多节点协同实现

针对无线传感器网络能量、存储、处理能力严重受限的特点,基于"在网计算"的思想,结合JPEG2000算法流程以及无线传感器网络的网络拓扑结构特点,提出一种基于邻居簇的JPEG2000多节点协同图像压缩方法。首先,将相机节点采集的图像分片,并根据图像的梯度幅度值进行压缩比特率的自适应优化分配。然后,将压缩任务转移到多个邻居簇内,以最小化网络总能耗为目标,由多节点协作共同完成图像压缩和传输。仿真结果表明,该方法不仅使无线传感器网络中实现大尺寸图像的JPEG2000编码成为可能,且相机节点能耗仅为压缩图像后传输至基站方案的3.4%,极大地平衡了网络节点能耗,使网络生命周期提高了7倍以上。     

遥感图像感兴趣区域压缩算法的研究

遥感图像在人类生活和军事领域的应用日益广泛,适合各种要求的遥感图像编码技术具有重要的现实意义。本文针对遥感图像的特点,提出将分层数的集划分算法与JPEG2000提出的感兴趣提升算法相结合,用于遥感图像行感兴趣区域压缩。根据MAXSPIHT的不足,提出了GPBSHIFT感兴趣平面提升方法。实验表明．改进的方法与无ROI和MAXSPIHT比特面提升方法相比,在码率相同时,感兴趣区域和背景的PSNR有着明显的改善。     

无线传感器网络中基于邻居簇的JPEG2000多节点协同图像压缩方法

针对无线传感器网络能量、存储、处理能力严重受限的特点,一些学者提出了“在网计算”的思想,对实现无线传感器网络中大尺寸、高分辨率图像的压缩和传输极具指导意义。结合JPEG2000算法流程以及无线传感器网络的网络拓扑结构特点,本文提出一种基于邻居簇的JPEG2000多节点协同图像压缩方法,即将相机节点采集的图像分片,然后将压缩任务转移到多个邻居簇内,由多节点协作共同完成。仿真结果表明,该方法不仅使无线传感器网络中实现大尺寸图像的JPEG2000编码成为可能,且极大地缓解了相机节点的能耗压力,进而延长网络生命周期。     

Image analysis for denoising full-field frequency-domain fluorescence lifetime images

Video-rate fluorescence lifetime-resolved imaging microscopy (FLIM) is a quantitative imaging technique for measuring dynamic processes in biological specimens. FLIM offers valuable information in addition to simple fluorescence intensity imaging; for instance, the fluorescence lifetime is sensitive to the microenvironment of the fluorophore allowing reliable differentiation between concentration differences and dynamic quenching. Homodyne FLIM is a full-field frequency-domain technique for imaging fluorescence lifetimes at every pixel of a fluorescence image simultaneously. If a single modulation frequency is used, video-rate image acquisition is possible. Homodyne FLIM uses a gain-modulated image intensified charge-coupled device (ICCD) detector, which unfortunately is a major contribution to the noise of the measurement. Here we introduce image analysis for denoising homodyne FLIM data. The denoising routine is fast, improves the extraction of the fluorescence lifetime value(s) and increases the sensitivity and fluorescence lifetime resolving power of the FLIM instrument. The spatial resolution (especially the high spatial frequencies not related to noise) of the FLIM image is preserved, because the denoising routine does not blur or smooth the image. By eliminating the random noise known to be specific to photon noise and from the intensifier amplification, the fidelity of the spatial resolution is improved. The polar plot projection, a rapid FLIM analysis method, is used to demonstrate the effectiveness of the denoising routine with exemplary data from both physical and complex biological samples. We also suggest broader impacts of the image analysis for other fluorescence microscopy techniques (e.g. super-resolution imaging).     

JPEG2000核心算法的改进及实现

在对JPEG2000标准的基本原理、编码流程所使用的算法等进行了详细分析的基础上,研究了决定编码器性能的核心算法;对JPEG2000中EBCOT算法的位平面编码进行了两点改进,并将改进后的算法在DSP硬件平台上得以实现.实验结果表明改进后的算法具有良好的压缩性能.     

星载遥感图像JPEG2000压缩算法的去量化

随着空间技术的发展,空间遥感图像的数据量呈几何级数增加.为了满足JPEG2000星载遥感图像压缩系统的实时性要求,提出了一种对压缩算法进行优化和改进的方法.分析了离散小波变换后拉伸提升算法和JPEG2000量化算法.然后,通过一个新的拉伸量化因子将量化过程与后拉伸小波变换中的提升过程结合起来,减少了JPEG2000压缩算法的量化操作.最后,介绍了去量化处理方法的定点实现.实验结果表明:应用去量化处理方法可以使JPEG2000压缩算法中拉伸和量化过程所需要的时间降低50%,提升了JPEG2000星载遥感图像压缩系统的实时处理能力.     

基于小波变换的图像自适应盲水印算法

随着计算机和网络技术的飞速发展,数字图像、音频和视频产品越来越需要一种有效的版权保护方法,所谓数字水印就是一种嵌入到图像、视频或音频数据中的不可见标志,可以用于多媒体数据的版权保护、认证和标注等.本文提出了一个有效的静止图像的自适应盲水印算法,在水印检测过程中不需要原始图像.原始图像经小波变换后,选择内嵌水印所需要的子带,水印根据邻居特征平均值法和奇偶判决法内嵌到频域中所选择的子带上.实验结论和攻击测试表明,本文所提议的算法具有较好的透明性,对如JPEG有损压缩、中值滤波、附加噪声、伸缩、裁剪等各种图像处理的攻击有较强的顽健性.     

Linear filtering of 2-D wavelet coefficients for denoising ultrasound medical images

The paper presents a novel denoising method for ultrasound medical images, whose quality is degraded by the peculiar phenomenon of speckle noise. The method is constructed step-by-step on the basis of recent research on the topic, and consists in Gaussian filtering of proper wavelet coefficients of the image, corresponding to vertical and diagonal details. A comparison with other filtering techniques for ultrasound imaging, i.e. Wiener and median filter, is presented. The obtained results, combined with those reported in independent research, demonstrate that the proposed denoising scheme has very good performance and is very promising for actual medical application.     

视觉传感器网络协作块压缩感知图像传输方法

JPEG2000等现有图像压缩传输算法虽然压缩性能好,但复杂度高,资源受限的视觉传感器节点很难适应.基于压缩感知理论研究视觉传感器网络环境下的图像传输问题.首先设计主-从节点对模型,分块监测图像;然后研究协作块压缩感知理论,基于图像分块和节点协作来减少测量矩阵的维数,降低压缩感知的计算复杂度;对处在发送端的主从节点,提出尽力而为的压缩感知测量值传输方法,以最大化图像重构质量为目标,尽力传送最多的测量值;而在接收端,汇聚节点依据内积准则来判断重构质量,确定传输终止条件,避免多余的测量值传输,节省传输耗能.实验结果表明,和经典压缩感知相比较,4个视觉传感器节点协作的块压缩感知传输方法将测量矩阵维数减少了2个量级,且有更少的网络耗能.     

结合压缩感知和曲波的天文图像去噪

在天文图像去噪中,为了提高迭代曲波阈值算法的去噪重建性能,提出了基于循环平移和曲波维纳滤波的压缩感知迭代重构算法。首先,使用基于曲波阈值的循环平移方法对重构图像进行调整以抑制重构图像中的伪吉布斯效应;接着,用提出的曲波维纳滤波算子替代小波阈值在迭代过程中对图像曲波系数进行筛选以进一步提高重构图像的质量。通过对添加高斯白噪声的Lena图像和月球图像进行重构实验,分析本文算法和当前主流算法的性能。实验结果表明,与传统的压缩感知迭代曲波阈值算法相比,本文算法能够获得较优的去噪性能,有效地保护天文图像的细节信息,峰值信噪比大约提高了2.6~3.2dB。     

A comparison of single particle tracking and temporal image correlation spectroscopy for quantitative analysis of endosome motility

Single particle tracking (SPT) is becoming a standard method to extract transport parameters from time‐lapse image sequences of fluorescent vesicles in living cells. Another method to obtain these data is temporal image correlation spectroscopy (TICS), but this method is less often used for measurement of intracellular vesicle transport. Here, we present an extensive comparison of SPT and TICS. First we examine the effect of photobleaching, shading and noise on SPT and TICS analysis using simulated image sequences. To this end, we developed a simple photophysical model, which relates spatially varying illumination intensity to the bleaching propensity and fluorescence intensity of the moving particles. We found that neither SPT nor TICS are affected by photobleaching per se, but the transport parameters obtained by both methods are sensitive to the signal‐to‐noise ratio. In addition, the number of obtained trajectories in SPT is affected by noise. Diffusion constants determined by TICS are significantly overestimated when large immobile fluorescent structures are present in the image sequences, while the opposite is true for SPT. To improve the performance of both techniques, we compare three different methods for image denoising. Appropriate denoising significantly reduced the effect of noise and of immobile structures on both methods. Shape fluctuations of simulated particles had a more pronounced effect on TICS than on SPT analysis. In denoised images of fluorescent beads or cytosolic vesicles containing fluorescent protein NPC2 in human skin fibroblast cells, the transport parameters acquired by SPT and TICS were comparable emphasizing the value of both analysis methods.     

基于脉冲耦合神经网络模型的小波自适应斑点噪声滤除算法

分析了维纳滤波原理和脉冲耦合神经网络(PCNN)模型的特点,根据斑点噪声统计模型的特征,结合小波变换方法,提出了一种基于PCNN模型的小波自适应斑点噪声滤除算法(W-PCNN-WD)来改善超声图像质量.首先,对超声图像进行对数变换,使斑点噪声转换为加性噪声;对医学图像进行维纳滤波处理,计算其加性噪声的标准方差,并以此作为小波阈值.然后,利用小波变换对图像进行预处理,利用PCNN在小波域中对小波系数进行相应的修正.最后,进行小波逆变换和指数变换,获得滤除噪声的图像.结果表明:本文提出的滤波方法优于其他滤波方法,当噪声方差为0.01时,本文滤波算法获得的峰值信噪比(PSNR)比经Wiener滤波方法获得的高出9 dB.该滤波方法能在有效去除超声斑点噪声的基础上保留图像的边缘细节信息,极大地改善了图像的视觉质量.     

基于灰度差分不变量的快速局部特征描述算法

提出了一种基于灰度差分不变量区域统计直方图的快速局部特征描述算法来解决传统灰度差分不变量特征描述子计算复杂、稳定性较差且包含的信息量较少的问题.采用低阶且具有微分几何意义的灰度差分不变量描述特征点以降低特征描述子的计算复杂度,提高特征描述子的稳定性;利用特征点邻域的灰度信息和区域信息提高特征描述子的信息含量,增强特征描述子的鲁棒性.将该算法应用于图像匹配.实验结果表明,在图像尺度缩放、旋转、模糊、亮度变化、较小视角变化和JPEG压缩等多种变换条件下,该描述子不仅能够取得较好的匹配效果,而且处理速度比尺度不换特征变换( SIFT)提高约2倍.     

利用相位差异技术恢复宽带白光图像

研究了用于扩展目标成像的相位差异技术。该技术利用焦面和离焦面上的两幅图像之间的相位差异来估算波前相位畸变,同时对采集的图像进行恢复。设计了白光目标成像实验系统,用宽带白光作光源,在实验室搭建实验平台,模拟无穷远处目标和波前相位畸变。采用直角梯形分光棱镜将光路分为两个通道,在一台相机上同时采集两个通道像面上的图像,消除了相机不同产生的差异,极大地抑制了噪声对图像恢复的影响,分光后系统抗两个通道独立的高斯白噪声的能力为10%。实验结果显示恢复后的图像分辨率提高了19%,表明该技术是适合光电成像系统的图像恢复技术。