1 新疆大学信息科学与工程学院, 新疆 乌鲁木齐 830046
2 上海交通大学图像处理与模式识别研究所, 上海 200240
3 Knowledge Engineering and Discovery Research Institute, Auckland University of Technology, Auckland 1020, New Zealand
光谱分析是化学计量学的一个重要应用方向, 并已被广泛应用到各个领域, 其中光谱变量选择又是光谱分析的重要环节。 研究不同的变量选择方法客观地识别有用的信息变量和消除无关或干扰变量十分关键。 提出了一种新的变量选择方法, 命名选择比率的竞争性群体分析法(SRCMPA)。 该算法采用选择比率, 自适应加权采样和模型群体分析的思想, 并结合了变量排列和指数递减函数方法。 关键波长定义为多元线性回归模型中得分值较大的波长, 将线性模型PLS下的选择比率的得分值作为评价各波长重要性的指标, 然后, 根据每个波长的重要性, SRCMPA依次从蒙特卡罗采样中选择N个波长子集, 以迭代和竞争的方式运行。 在每一次采样运行中, 以固定比率的样品以建立校准的PLS模型并计算每个变量的选择比率值, 基于排序选择比率的得分值和作为权重的归一化的SR(选择比率)得分值, 采用指数递减函数的强制选择和自适应加权采样竞争选择的两步过程来选择关键变量。 最后, 应用交叉验证(CV)方法来选择具有最低交叉验证均方根(RMSECV)的子集作为最优子集。 该算法已在小麦蛋白数据集和啤酒数据集上进行了测试, 并使用三种高效算法作对比。 通过对实验结果来评估算法优越性, 该算法能够找到数据集的关键波长变量的最佳组合, 并能用于解释感兴趣的化学特性, 通过建模后的评价结果也是最佳的。 该算法在啤酒光谱数据集的运行结果, 相较于啤酒数据集的全光谱PLS模型, 变量个数由567个减少到42个左右。 并且模型的RMSECV由0.622下降到0.115, RMSEP由0.823减少到了0.263左右, 预测精度分别提高了81.5%和68.1%。 Q2_CV和Q2_test也分别由0.940, 0.852提高到了0.994和0.995。 在小麦蛋白数据集的运行结果, 相较于于小麦蛋白光谱数据集的全光谱PLS模型, 变量个数由175个减少到18个左右。 并且模型的RMSECV由0.607下降到0.292, RMSEP由0.519减少到了0.234左右, 预测精度分别提高了51.9%和54.9%。 Q2_CV和Q2_test也分别由0.748, 0.774提高到了0.931和0.839。
变量选择 选择比率 自适应加权采样 群体模型分析 蒙特卡罗采样 Variable selection Selection ratio Adaptive weighted sampling Population model analysis Monte Carlo sampling 光谱学与光谱分析
2020, 40(4): 1056
Author Affiliations
Abstract
1 School of Physical Science and Technology, Xinjiang University, Urumqi 830046, China
2 Affiliated Tumor Hospital of Xinjiang Medical University, Urumqi 830000, China
3 Department of Engineering, China University of Petroleum-Beijing, Keramayi 834000, China
4 School of Information Science and Engineering, Xinjiang University, Urumqi 830046, China
5 School of Electronic Engineering, Beijing University of Posts and Telecommunications, Beijing 100876, China
6 Quality of Products Supervision and Inspection Institute, Urumqi 830011, China
In this Letter, the surface-enhanced Raman scattering (SERS) signal of early breast cancer (BRC) patient serum is obtained by a composite silver nanoparticles (Ag NPs) PSi Bragg reflector SERS substrate. Based on these advantages, the serum SERS signals of 30 normal people and 30 early BRC patients were detected by this substrate. After a baseline correction of the experimental data, principal component analysis and linear discriminant analysis were used to complete the data processing. The results showed that the diagnostic accuracy, specificity, and sensitivity of the composite Ag NPs PSi Bragg reflector SERS substrate were 95%, 96.7%, and 93.3%, respectively. The results of this exploratory study prove that the detection of early BRC serum based on a composite Ag NPs PSi Bragg reflector SERS substrate is with a stable strong SERS signal, and an unmarked and noninvasive BRC diagnosis technology. In the future, this technology can serve as a noninvasive clinical tool to detect cancer diseases and have a considerable impact on clinical medical detection.
porous silicon Bragg reflector surface-enhanced Raman scattering breast cancer detection principal component analysis linear discriminant analysis Chinese Optics Letters
2020, 18(5): 051701
Author Affiliations
Abstract
1 College of Chemistry and Chemical Engineering, Xinjiang University, Urumqi 830046, China
2 College of Information Science and Engineering, Xinjiang University, Urumqi 830046, China
3 College of Physical Science and Technology, Xinjiang University, Urumqi 830046, China
We present a technique for fabricating a fluorescence enhancement device composed of metal nanoparticles (NPs) and porous silicon (PSi) diffraction grating. The fluorescence emission enhancement properties of the PSi and the fluorescence enhancement of the probe molecules are studied on PSi gratings. The fluorescence enhancement of the probe molecules on a fluorescence enhancement device is further improved through the deposition of metal NPs onto the PSi grating. In comparison to metal NP/PSi devices, metal NP periodic distributions can produce a stronger fluorescence enhancement that couples with the PSi grating fluorescence enhancement to achieve an overall three-fold enhancement of the fluorescence intensity.
050.0050 Diffraction and gratings 230.0230 Optical devices Chinese Optics Letters
2017, 15(11): 110501
Author Affiliations
Abstract
1 School of Physical Science and Technology, Xinjiang University, Urumqi 830046, China
2 School of Information Science and Engineering, Xinjiang University, Urumqi 830046, China
3 Department of Physics, Ningbo University, Ningbo 315211, China
The fluorescence of graphene oxide quantum dots (GOQDs) that are infiltrated into porous silicon (PSi) is investigated. By dropping activated GOQDs solution onto silanized PSi samples, GOQDs are successfully infiltrated into a PSi device. The results indicate that the intensity of the fluorescence of the GOQD-infiltrated multilayer with a high reflection band located at its fluorescence spectra scope is approximately double that of the single layer sample. This indicates that the multilayer GOQD-infiltrated PSi substrate is a suitable material for the preparation of sensitive photoluminescence biosensors.
050.5298 Photonic crystals 120.5700 Reflection 130.3990 Micro-optical devices 160.2540 Fluorescent and luminescent materials 160.4760 Optical properties Chinese Optics Letters
2016, 14(4): 041601
1 石河子大学 信息科学与技术学院，新疆 石河子 832003
2 新疆大学 信息科学与工程学院，乌鲁木齐 830046
采用双槽电化学腐蚀法制备了纳米多孔硅，主要研究了腐蚀时间和腐蚀电流对重掺杂p型（100）硅衬底上制备的多孔硅层有效光学厚度的影响,采用U4100光谱仪、场发射扫描电子显微镜(FESEM)技术对所制备的多孔硅光子晶体的结构和有效光学厚度进行了分析表征。研究结果表明，通过合理地选择腐蚀时间和腐蚀电流,可以比较精确地制备特定有效光学厚度的多孔硅薄膜，此方法可广泛应用于纳米多孔硅光子晶体的制备中。
纳米硅 双槽电化学腐蚀 腐蚀条件 有效光学厚度 porous silicon doubletank electrochemical corrosion method etching conditions effective optical thickness
1 上海交通大学图像处理与模式识别研究所， 上海200240
2 新疆大学电子系， 新疆 乌鲁木齐830046
3 奥克兰理工大学知识工程与发现研究所, 新西兰 奥克兰
由于遥感图像存在边缘混叠等问题， 经典的C-V模型会产生大量的冗余轮廓， 而且无法分割多个同质区域的目标。 为此， 提出了基于C-V模型的窄带多区域水平集图像分割方法， 采用N-1个水平集函数将图像分割成N(N＞1)个区域， 每个水平集函数表达一个区域。 该方法一方面通过建立独立多区域水平集模型可以消除多余的轮廓， 避免分割区域的重叠和漏分； 另一方面应用窄带技术来减小水平集方法中的计算量。 对遥感图像进行了实验， 结果表明该方法能快速有效地分割该图像。
图像分割 窄带水平集 多区域水平集 遥感图像 Image segmentation Narrow band level set Multi-region level set Remote sensing image 光谱学与光谱分析
2011, 31(11): 3001
新疆大学 信息科学与工程学院，乌鲁木齐 830046
均衡技术是用来解决通信系统中码间干扰的常用方法.针对传统恒模算法的收敛速度和稳态剩余码间干扰大的缺点，将双模式多模算法和改进型判决引导算法结合起来，并且应用于分数间隔盲均衡器中.该算法初期用双模式算法收敛均衡器，然后切换到改进型判决引导算法中进一步收敛，由于分数间隔均衡器解决了波特间隔均衡器因抽样率不高带来的频谱混叠问题，从而进一步地提高了均衡效果.蒙特卡罗仿真表明，该算法不仅收敛速度快，而且得到较低的剩余码间干扰.
恒模算法 盲均衡 分数间隔 双模式多模算法 Constant Modulus Agorithm(CMA) Blind equalization Fractionally-spaced Dual-mode blind equalization algorithm
Author Affiliations
Abstract
1 School of Physics, Xinjiang University, Urumqi 830046, China
2 School of Electronic and Information Engineering, Xi'an Jiaotong University, Xi'an 710049, China
3 Key Laboratory of Xinjiang Biological Resources and Gene Engineering, College of Life Sciences and Technology, Xinjiang University, Urumqi 830046, China
4 College of Information Science and Engineering, Xinjiang University, Urumqi 830046, China
Porous silicon (PS) suitable for optical detection of immunoreaction is fabricated. The structure of immunosensor is prepared by the following steps: oxidization, silanization, glutaraldehyde cross-linker, and covalent binding of antibody. When antigen is added into the immunosensor, the Raman intensity is estimated to be linearly reduced according to the concentration of the surface protective antigen protein A (spaA) of below 4.0 μg·ml?1. The ultimate detection limit is 1.412×10^2 pg.ml^{?1}. Controlled experiments are also presented with non-immune antigen of the spaA, and results show that the immunosensor has high specificity. Compared with the conventional enzyme-linked immuno sorbent assay (ELISA), this method is quick, inexpensive, and label-free.
多孔硅 免疫传感器 猪丹毒丝菌表面保护性抗原A 拉曼光谱 光学检测 280.1415 Biological sensing and sensors 070.0070 Fourier optics and signal processing 070.4790 Spectrum analysis 280.0280 Remote sensing and sensors Chinese Optics Letters
2011, 9(2): 022801
Author Affiliations
Abstract
1 School of Electronic and Information Engineering, Xi'an Jiaotong University, Xi'an 710049, China
2 School of Materials Science and Engineering, Tongji University, Shanghai 200092, China
3 School of Chemistry and Chemical Engineering, Xinjiang University, Urumqi 830046, China
4 College of Information Science and Engineering, Xinjiang University, Urumqi 830046, China
A TiO2/porous silicon (PS) composite system is prepared by chemical vapor deposition. The crystal form with anatase phase of the samples is evaluated by X-ray diffraction and ultraviolet-visible (UV-vis) absorbance spectra, and the morphology with microsphere of TiO2 particles is characterized by scanning electron microscopy. The composite system formed by this technique gives a broad blue luminescence and the mechanism of photoluminescence with TiO2/PS is also discussed.
多孔硅 二氧化钛(TiO2) 复合体系 光致发光 160.4670 Optical materials 310.6860 Thin films, optical properties 240.0310 Thin films 250.5230 Photoluminescence Chinese Optics Letters
2010, 8(6): 618
新疆大学信息科学与工程学院,新疆,乌鲁木齐,8300415
为了有效地滤除混合噪声,本文提出了一种基于人眼视觉特性的混合滤波算法.该方法首先采用基于人眼视觉特性的噪声敏感系数作为阈值来确定脉冲噪声点,对检测出脉冲噪声点采用自适应窗口大小的迭代中值滤波进行滤波,而对于含有高斯噪声的像素点则采用一种保护细节的改进的自适应模糊滤波器进行处理.该算法与标准滤波方法及其它改进混合滤波算法相比,具有更好的滤波性能.
混合噪声 模糊滤波器 人眼视觉特性(HVS)
