Determination of Optical Properties of Turbid Media from Spatially Resolved Diffuse Reflectance by Neural Network

A backpropagation (BP) network is applied to the inversion of spatially resolved diffuse reflectance from turbid media and then to determine its optical properties. A standard BP network may be trapped to the local minimum. A BP network with variable momentum and variable leaning rate can reduce this effect. After being trained, this network will produce reduced scattering coefficients and absorption coefficients when the spatially resolved diffuse reflectance are fed to its input.     

Influence of the phase function on determination of the optical properties of biological tissue by spatially resolved reflectance

Spatially resolved reflectance measurements are widely used for determination of the optical properties of biological media. However, the influence of the phase function on these measurements has not been quantified. We show that errors in the derived reduced scattering and absorption coefficients are as great as 100% for both absolute and relative spatially resolved reflectance measurements if a standard solution of the diffusion equation is used in the analysis. In addition, we investigated nonlinear regressions, using Monte Carlo simulations and an additional fitting parameter that characterizes the phase function, and found that the errors in the obtained optical coefficients were 相似文献   

近红外光谱在转基因玉米检测识别中的应用

随着转基因食品的推广应用,人们越来越关心其食用安全性。以转基因玉米及其亲本为实验材料,借助于近红外光谱仪对转基因玉米及其亲本进行了识别分析：扫描区间为12 000～4 000 cm-1,分辨率为4 cm-1,扫描次数为64次;以三层误差反向传播算法(简称BP算法)进行数据处理。结果显示通过扫描光谱及数学和计算机软件分析,非常准确、方便地识别了转基因农产品。所以通过近红外光谱所建立的转基因BP-网络识别模型完全可用于实际应用。近红外分析还具有无污染、成本低等优点,是一种极具前景的转基因食品安全检测识别技术。     

Method for noninvasive determination of hemoglobin content in biological tissues

We have developed a simple and efficient method for determining the hemoglobin concentration in biological tissues, based on spatially resolved measurements of the diffuse reflectance of the tissue at λ = 524, 578, and 662 nm (or 773 nm) using multiple regressions between the measured parameters and the hemoglobin concentrations. The method takes into account the presence in the blood of the major hemoglobin derivatives (oxyhemoglobin, deoxyhemoglobin, carboxyhemoglobin, and methemoglobin) and eliminates the effect of the parameters of the top layer of the biological tissue (for example, the epidermis of the skin) and the bulk scattering coefficient of the tissue on the measurement results. Based on numerical Monte Carlo experiments, we have estimated the uncertainties in determining the hemoglobin concentration in skin tissues when its structural and biochemical parameters are all variable.     

基于可见/近红外反射光谱的大豆叶绿素含量估算方法比较

作物叶绿素含量的估测可以为精准农业提供技术支持.该文利用 PROSAIL 模型模拟了不同叶绿素水平下的大豆冠层光谱反射率,而后针对多期实测高光谱及相应的叶绿素数据,在对响应波段进行小波能量系数提取的基础上,分别采用多元线性回归、BP 神经网络和 RBF 神经网络、以及偏最小二乘法进行估算,并进行了比较分析.研究结果表明...     

Estimating optical properties in layered tissues by use of the Born approximation of the radiative transport equation

We use the Born approximation of the radiative transport equation to recover simultaneously the absorption and scattering coefficients in a single layer of a two-layer tissue sample from reflectance data. This method reduces the estimation of both optical properties to a single linear, least-squares problem. It is valid over length scales smaller than a transport mean free path and hence is useful for epithelial tissue layers. We demonstrate the accuracy of this method by using spatially resolved reflectance data computed with Monte Carlo simulations.     

各向异性介质辐射特性参数联合反演

本文基于BP神经网络方法结合蒙特卡洛和BEER定律辐射传输模拟方法建立了联合反演各向异性散射介质的辐射特性参数模型。首先采用半球透射率结合半球反射率反演模型反演了各向同性介质的吸收系数和散射系数,在此基础上增加准直透射率,建立了联合反演各向异性介质的吸收系数、散射系数和散射不对称因子三参数联合反演模型。反演结果表明该模型能准确反演出介质辐射特性参数,具有实用意义。此外,为了检验测量误差对模型的反演准确性的影响,分别在不同程度测量误差情况下进行反演,结果显示测量误差对散射不对称因子反演值影响较大。     

优化光谱指数建立水稻叶片SPAD的高光谱反演模型

利用高光谱反射率光谱的特征波段构建光谱指数,建立叶绿素含量反演模型是实现水稻生产精准调控和科学管理的必要手段之一。为了建立适用于拔节孕穗期水稻叶片叶绿素相对含量（SPAD）的高光谱反演模型,分别获取了拔节孕穗期水稻叶片的高光谱和SPAD数据,利用小波分析法对原始光谱反射率曲线进行降噪处理,并对基于积分运算的光谱指数NAOC进行简化,获得了基于双波段简化运算的优化光谱指数。利用相关分析法计算由原始反射率光谱R和数学变换光谱LgR、1/R和R构建的优化光谱和变换光谱指数与水稻叶片SPAD的相关系数,获得了以积分限（a,b）为横、纵坐标的相关系数二维矩阵,并绘制相关性等势图,得到相关系数最高的3个波段组合：R（641,790）（0.872 6）,R（653,767）（0.871 7）和R（644,774）（0.871 6）,计算出20个原始样本中3个积分波段组合所对应的60个优化光谱指数值,按照2∶1的比例划分为建模集和验证集,建立了三种水稻叶片SPAD反演模型：偏最小二乘回归（PLSR）、支持向量机（SVM）和BP神经网络模型。结果显示：利用优化光谱和变换光谱指数建立的3种水稻叶片SPAD反演模型决定系数R2均大于0.79,归一化均方根误差NRMSE则小于5.4%。其中BP神经网络相对于其他两种模型具有较高的拟合度,预测精度也相对较高,建模集R2=0.842 6,NRMSE=5.152 7%;验证集R2=0.857,NRMSE=4.829 9%。总体来看,基于双波段简化运算后的优化光谱和变换光谱指数建立拔节孕穗期水稻叶片SPAD反演模型是可行的;对比分析3种模型反演结果发现,BP神经网络对水稻叶片SPAD的反演效果较好。该工作对提高拔节孕穗期水稻精准调控技术和建立水稻生产的科学管理体系具有一定的参考价值。     

基于BP神经网络的红外焦平面非均匀性校正技术

在传统BP神经网络非均匀性校正理论的基础上,结合红外焦平面阵列读出通道的时变特性,提出一种新的输出期望估计方法.该方法利用当前像素邻域和读出通道估计输出期望值.实验表明,算法能有效抑制焦平面固定图案噪音,提高场景目标的分辨能力,达到较好的视觉效果.     

Noninvasive spectral imaging of skin chromophores based on multiple regression analysis aided by Monte Carlo simulation

In order to visualize melanin and blood concentrations and oxygen saturation in human skin tissue, a simple imaging technique based on multispectral diffuse reflectance images acquired at six wavelengths (500, 520, 540, 560, 580 and 600 nm) was developed. The technique utilizes multiple regression analysis aided by Monte Carlo simulation for diffuse reflectance spectra. Using the absorbance spectrum as a response variable and the extinction coefficients of melanin, oxygenated hemoglobin, and deoxygenated hemoglobin as predictor variables, multiple regression analysis provides regression coefficients. Concentrations of melanin and total blood are then determined from the regression coefficients using conversion vectors that are deduced numerically in advance, while oxygen saturation is obtained directly from the regression coefficients. Experiments with a tissue-like agar gel phantom validated the method. In vivo experiments with human skin of the human hand during upper limb occlusion and of the inner forearm exposed to UV irradiation demonstrated the ability of the method to evaluate physiological reactions of human skin tissue.     

Identification of Rhubarb Samples by Using NIR Spectrometry and Takagi–Sugeno Fuzzy System

Abstract

The Takagi–Sugeno fuzzy system is implemented based on several back‐propagation neural networks (BP‐NNs) and has been applied to identification of official and unofficial rhubarb samples based on their near‐infrared spectra. Rhubarb is one of the most important Chinese medical herbs. It is of importance to identify official and unofficial rhubarb samples based on nondestructive near‐infrared spectrometry for quality control in Chinese herbal products. Near‐infrared diffuse reflectance spectrometry and the Takagi–Sugeno fuzzy system were used to classify 52 rhubarb samples, and the effects of the number of hidden neurons and of momentum parameters on prediction were investigated. The results obtained by using the Takagi–Sugeno fuzzy system were better than those by commonly used BP networks. With proper network training parameters, 100% correctness can be obtained by using the Takagi–Sugeno fuzzy system.     

用CCD测量生物组织的漫反射率和透射率

介绍CCD实验装置和用C语言开发的CCD图像分析系统 ,测量和分析了新鲜猪脂肪和猪肌肉空间分辨的漫反射率和透射率 ,将测量结果与MonteCarlo模拟结果以及基于漫射理论的两个模型的结果相比较 ,结果表明在距离光束中心 12mm的范围内理论结果与实验结果符合得较好。结论是用CCD测量装置以及作者所开发的图像分析系统可以用于测定生物组织的漫反射率和透射率 ,这项测量为生物组织光学参数的无损和快速测量提供了实验依据。     

生物组织光传播的时域特征分析

采用蒙特卡罗方法对脉冲光在组织中传播的时域特征作了研究，并把计算结果与漫射理论的解析解作了比较，文中详细分析了组织光学特性参数与组织表面时间分辨的漫反射光分布之间的关系。结果表明：散射系影响着漫反射光强到达峰值的时间，吸收系数影响着漫反射光强在峰值之后随时间下降的速度，而ｇ因子对响应曲线的前沿影响较大，在后沿阶段基本上没有影响。     

基于BP神经网络的弗兰克-赫兹实验曲线拟合

BP神经网络通过调整连接权重便可按预定精确度逼近非线性函数,利用这一特点可对非线性函数关系进行拟合.利用BP神经网络对弗兰克-赫兹实验数据进行处理,结果显示该方法处理结果精度高,拟合效果好.     

The Infrared Ray Transport in an N-Layered Matched Skin

Infrared ray (IR) has great potential in medical diagnosis and therapy. In order to detect tumor in skin, we set up the steady-state and time domain IR diffusion model of an n-layered matched medium with an infinitely thick. We utilize the diffuse equation to solve a five-layered infinite matched medium and obtain the accurate solution of a matched medium of the steady state and time domain in tissue. We compare the steady-state spatially resolved reflectance calculated with Monte-Carlo simulations. The Monte-Carlo simulation shows that the solution is valid. Our equation can be used to obtain the tumor information in medical diagnosis and therapy.     

基于P3近似的空间分辨漫反射研究

光源附近组织的空间分辨漫反射是近年来生物医学光子学领域的一个研究热点,其目的是发展一种能够测定活体生物组织光学参量的新技术。漫射近似理论研究光源附近组织的空间分辨漫反射具有很大局限性。P3近似理论考虑了相函数的三阶矩,能较准确地描述光源附近组织的光辐射分布。研究了基于P3近似的空间分辨漫反射,从输运理论的PN方程组出发,导出了P3近似方程组和P3近似的格林函数解;阐述了漫射近似与P1近似的关系,给出了外推边界条件下,准直光束近似后的P3近似漫反射率的完整表示,讨论了相函数二阶参量对P3近似漫反射的影响,并与漫射近似和蒙特卡罗模拟结果进行了比较,指出了P3近似的应用范围。     

Regression approach to non-invasive determination of bilirubin in neonatal blood

A statistical ensemble of structural and biophysical parameters of neonatal skin was modeled based on experimental data. Diffuse scattering coefficients of the skin in the visible and infrared regions were calculated by applying a Monte-Carlo method to each realization of the ensemble. The potential accuracy of recovering the bilirubin concentration in dermis (which correlates closely with that in blood) was estimated from spatially resolved spectrometric measurements of diffuse scattering. The possibility to determine noninvasively the bilirubin concentration was shown by measurements of diffuse scattering at λ = 460, 500, and 660 nm at three source–detector separations under conditions of total variability of the skin biophysical parameters.     

基于光谱技术的杨梅汁品种快速鉴别方法的研究

为了实现杨梅汁品种的快速无损鉴别,提出了一种用可见和近红外光谱技术快速鉴别杨梅汁品种的新方法。首先采用偏最小二乘法进行模式特征分析,经过交互验证法判别,确定最佳主成分数为9。完成特征提取后,将这9个主成分作为神经网络的输入变量,建立了三层BP神经网络,实现类别预测的同时也完成了数学建模与优化分析工作。3个品种的杨梅汁样本数均为20,共计60个样本。在神经网络学习中,将其分成训练集样本51个和预测集样本9个。对9个未知样本进行预测,准确率为100％。说明本文提出的基于光谱技术和模式识别的方法具有很好的分类和鉴别能力。     

基于BP神经网络的磁滞回线的曲线拟合

BP神经网络通过调节连接权重可以实现以任意精度逼近非线性函数,利用这一特点可对非线性函数关系进行拟合。本利用BP神经网络对测定磁滞回线实验中的数据进行处理,结果表明该方法处理结果精度高,拟合效果好,计算机处理程序通用。     

平稳状态下匹配介质三层漫射方程的空间分辨漫反射

如果应用漫射方程分析和确定生物表面组织的光学特性,就能够确定生物活体是否发生病变,实现生物活体的无损探测。目前,光在生物体内传播常用的基本模型是半无限厚模型和两层介质模型。而实际上,人体组织为多层介质。根据漫射方程理论,在平稳状态条件下,使用外推边界条件,精确给出了匹配介质中平稳状态下三层体系光的漫射方程格林函数的解。通过此解,可以算出空间分辨漫反射。使用蒙特卡罗方法来验证三层体系的漫射方程,发现我们给出的理论解,不仅可以包括与Kienle的结论一致的两层体系模型的空间分辨漫反射的解,而且三层体系漫射方程的空间分辨漫反射解和蒙特卡罗方法基本一致。