BP神经网络结合变量选择方法在牛奶蛋白质含量检测中的应用北大核心CSCD

牛奶中的蛋白质含量会影响牛奶的品质,利用高光谱图像的光谱特征信息研究对牛奶蛋白质含量预测的可行性。本文提出一种基于竞争性自适应重加权算法(competitive adaptive reweighted sampling, CARS)和连续投影算法(successive projections algorithm, SPA)结合多层前馈神经网络(back propagation, BP)的预测建模方法,实验以含有不同浓度蛋白质的牛奶为对象,利用可见光/近红外高光谱成像系统共采集到5种牛奶共计250组高光谱数据,通过实验对比选择采用标准化方法对获取到的吸收光谱预处理,然后采用CARS结合SPA筛选特征波长,得到18个特征波长,建立CARS-SPA-BP模型,经过试验,CARS-SPA-BP模型的训练集决定系数和测试集决定系数R;和R;分别达到0.971和0.968,训练集均方根误差(root mean square error of calibration,RMSEC)和测试集均方根误差(root mean square error of prediction,RMSEP)达到了0.033和0.034。研究发现,采用CARS结合SPA筛选的牛奶特征波长建立的多层前馈神经网络模型,其模型预测结果与全波长建模相比并没有明显降低,因此将CARS结合SPA用于波长筛选并且结合BP神经网络基本可以完成对牛奶蛋白质含量的预测。为验证CARS-SPA-BP模型的预测能力,在相同数据环境下,使用较为传统的偏最小二乘回归(partial least squares regression, PLSR)进行建模,实验结果表明,CARS-SPA-BP相较于PLSR,R;和RMSEP均有明显提升。研究表明,CARS-SPA-BP可充分利用牛奶光谱特征信息实现较高精度的牛奶蛋白质含量检测。     

Joint L1/Lp-regularized minimization in video recovery of remote sensing based on compressed sensing

L₁ regularization and L_p regularization are proposed for processing recovered images based on compressed sensing (CS). L₁ regularization can be solved as a convex optimization problem but is less sparse than L_p (0 < p < 1). L_p regularization is sparser than L₁ regularization but is more difficult to solve. This paper proposes joint L₁/L_p (0 < p < 1) regularization, which combines L_p regularization and L₁ regularization. This joint regularization is applied to recover video of remote sensing based on CS. Joint regularization is sparser than L₁ regularization but is as easy to solve as L₁ regularization. A linearized Bregman reweighted iteration algorithm is proposed to solve the joint L₁/L_p regularization problem. The performance and capabilities of the linearized Bregman algorithm and linearized Bregman reweighted algorithm for solving the joint L₁/L_p regularization model are analyzed and compared through numerical simulations.     

竞争性自适应重加权算法和相关系数法提取特征波长检测番茄叶片真菌病害

基于竞争性自适应重加权算法(CARS)和相关系数法(CA)特征波长选择方法,提出了利用可见-近红外高光谱成像技术检测番茄叶片灰霉病的方法。首先获取380～1 023 nm波段范围内80个染病和80个健康番茄叶片的高光谱图像,然后提取染病和健康叶片感兴趣区域(ROI)的光谱反射率值,作为番茄叶片灰霉病鉴别模型的输入来建立支持向量机(SVM)鉴别模型,训练集和验证集的鉴别率都是100%。研究进一步通过CARS和CA提取特征波长,分别得到5个(554, 694, 696, 738和880 nm)和4个(527, 555, 571和633 nm)特征波长,然后分别建立CARS-SVM和CA-SVM鉴别模型。结果显示,CARS-SVM模型中训练集和验证集的鉴别率都是100%,CA-SVM模型中训练集和验证集的鉴别率分别是91.59%和92.45%。以上结果说明了从可见-近红外高光谱图像中提取的光谱反射率值用于检测番茄叶片的灰霉病是可行的。     

Asymptotics of Estimating Equations under Natural Conditions

In a variety of statistical problems one needs to solve an equation in order to get an estimator. We consider the large sample properties of such estimators generated from samples that are not necessarily identically distributed. Very general assumptions that lead to the existence, strong consistency, and asymptotic normality of the estimators are given. A number of results that are useful in verifying the general assumptions are given and an example illustrates their use. General applications to maximum likelihood, iteratively reweighted least squares, and robust estimation are discussed briefly.     

用概率推导和加权迭代L1范数实现信号重构

在对信号稀疏性统计分析的基础上,将具有稀疏描述能力的拉普拉斯分布用于描述信号的先验分布,基于贝叶斯法,利用信号采样值、拉普拉斯先验分布和高斯似然模型,推导信号的后验概率密度估计;最后将最大后验概率(MAP)估计过程转化为加权迭代L1范数的最小化问题。在求解过程中,与非加权的L1范数法进行对比表明,信号重构性能明显提高;通过实验计算,详细讨论了其中一些参数的取值原则和范围;针对稀疏度不同的信号,随着信号非零点数的增加,本文算法重构结果明显优于基追踪(BP)和(OMP)法;与同类的IRL1算法相比较,本文算法更具普遍性和理论意义。     

Iteratively reweighted algorithm for signals recovery with coherent tight frame

We consider the problem of compressed sensing with a coherent tight frame and design an iteratively reweighted least squares algorithm to solve it. To analyze the problem, we propose a sufficient null space property under a tight frame (sufficient D‐NSP). We show that, if a measurement matrix A satisfies the sufficient D‐NSP of order s, then an s‐sparse signal under the tight frame can be exactly recovered. Furthermore, if A satisfies the restricted isometric property with tight frame D of order 2bs, then it also satisfies the sufficient D‐NSP of order as with a < b and b sufficiently large. We prove the convergence of the algorithm based on the sufficient D‐NSP and give the upper error bounds. In numerical experiments, we use the discrete cosine transform, discrete Fourier transform, and Haar wavelets to verify the effectiveness of this algorithm. With increasing measurement number, the signal‐to‐noise ratio increases monotonically.     

激光诱导击穿光谱结合CARS变量选择方法定量检测倍硫磷含量

利用双脉冲激光诱导击穿光谱(LIBS)技术对溶液中的倍硫磷含量进行定量检测。采用二通道高精度光谱仪采集不同浓度倍硫磷样品在206.28~481.77 nm波段的LIBS光谱,并对光谱进行多元散射校正(MSC)、标准正态变量变换(SNV)及3点平滑预处理,根据偏最小二乘(PLS)建模确定最优的预处理方法。在此基础上,利用竞争性自适应重加权算法(CARS)筛选与倍硫磷相关的重要变量,然后应用PLS回归建立溶液中倍硫磷含量的定量分析模型,并与单变量定量分析模型及未变量选择的PLS定量分析模型进行比较。结果表明,相比单变量定量分析模型及原始光谱PLS定量分析模型,CARS-PLS定量分析模型的性能更优,其模型的校正集和预测集的决定系数及平均相对误差分别为0.969 4、15.537%和0.995 9、5.016%。此外,与原始光谱PLS模型相比,CARS-PLS模型仅使用其中1.9%的波长变量,但预测集平均误差却由9.829%下降为5.016%。由此可见,LIBS技术检测溶液中的倍硫磷含量具有一定的可行性,且CARS方法能简化定量分析模型,提高模型的预测精度。     

基于近红外光谱的食用植物油中反式脂肪酸含量快速定量检测及模型优化研究

利用近红外光谱技术对食用植物油中反式脂肪酸(Trans fatty acids,TFA)含量进行快速定量检测,并通过波段选择、预处理方法、变量筛选及建模方法对TFA含量预测模型进行优化.采用AntarisⅡ傅里叶变换近红外光谱仪在4000~10000 cm-1光谱范围采集98个食用植物油样本的近红外透射光谱,然后采用气相色谱法测定TFA的真实含量.首先,对样本原始光谱进行波段、预处理方法优选;在此基础上,采用竞争自适应重加权法(Competitive adaptive reweighted sampling,CARS)筛选TFA相关的重要变量,最后应用主成分回归、偏最小二乘和最小二乘支持向量机方法分别建立食用植物油中TFA含量的预测模型.研究结果表明,近红外光谱技术检测食用植物油中的TFA含量是可行的,优化后的最佳预测模型的校正集和预测集R2分别为0.992和0.989,RMSEC和RMSEP分别为0.071%和0.075%.最佳预测模型所用的变量仅26个,占全波段变量的0.854%.此外,与全波段偏最小二乘预测模型相比,其预测集R2由0.904上升为0.989,RMSEP由0.230%下降为0.075%.由此表明,模型优化非常必要,CARS能有效筛选TFA相关的重要变量,极大减少建模变量数,从而简化预测模型,并较大提高预测模型的精度和稳定性.     

Key wavelengths screening using competitive adaptive reweighted sampling method for multivariate calibration

By employing the simple but effective principle ‘survival of the fittest’ on which Darwin's Evolution Theory is based, a novel strategy for selecting an optimal combination of key wavelengths of multi-component spectral data, named competitive adaptive reweighted sampling (CARS), is developed. Key wavelengths are defined as the wavelengths with large absolute coefficients in a multivariate linear regression model, such as partial least squares (PLS). In the present work, the absolute values of regression coefficients of PLS model are used as an index for evaluating the importance of each wavelength. Then, based on the importance level of each wavelength, CARS sequentially selects N subsets of wavelengths from N Monte Carlo (MC) sampling runs in an iterative and competitive manner. In each sampling run, a fixed ratio (e.g. 80%) of samples is first randomly selected to establish a calibration model. Next, based on the regression coefficients, a two-step procedure including exponentially decreasing function (EDF) based enforced wavelength selection and adaptive reweighted sampling (ARS) based competitive wavelength selection is adopted to select the key wavelengths. Finally, cross validation (CV) is applied to choose the subset with the lowest root mean square error of CV (RMSECV). The performance of the proposed procedure is evaluated using one simulated dataset together with one near infrared dataset of two properties. The results reveal an outstanding characteristic of CARS that it can usually locate an optimal combination of some key wavelengths which are interpretable to the chemical property of interest. Additionally, our study shows that better prediction is obtained by CARS when compared to full spectrum PLS modeling, Monte Carlo uninformative variable elimination (MC-UVE) and moving window partial least squares regression (MWPLSR).     

近红外光谱结合CARS变量筛选方法用于液态奶中蛋白质与脂肪含量的测定

采用CARS(Competitive adaptive reweighted sampling)变量筛选方法建模,显著提高了液态奶中蛋白质与脂肪近红外模型的预测精度。用蒙特卡罗采样(Monte-Carlo sampling)方法先剔除奇异样本,再对光谱进行中心化与Karl Norris滤波降噪处理,通过CARS方法筛选出与样本性质密切相关的变量,建立预测蛋白质与脂肪含量的偏最小二乘法(PLS)校正模型,并与未选变量的PLS模型进行比较。以定标集相关系数(r2)及交互验证均方残差(RMSECV)和预测误差均方根(RMSEP)作为判定依据,确定了蛋白质与脂肪的最佳建模条件。蛋白质与脂肪校正模型的相关系数分别为0.975 0、0.995 1,RMSECV分别为0.194 8、0.136 3,RMSEP分别为0.113 3、0.140 1,预测结果优于未选变量的PLS模型及其他选变量方法,有效简化了模型,适于液态奶中脂肪和蛋白质的快速、无损检测。