312 results about "Partial least squares regression" patented technology

The invention relates to a method for diagnosing the crop water deficit through a hyperspectral image technology, and especially relates to a method for diagnosing the Lycopersicon esculentum Mill. leaf area water based on hyperspectral images. The method comprises the following steps: 1, acquiring Lycopersicon esculentum Mill. leaf hyperspectral image data through a self-constructed hyperspectral imaging system; 2, selecting a characteristic wavelength by optimizing through an adaptive band selection process to realize multidimensional datum dimensionality reduction; 3, dividing the image ofeach sample at the characteristic wave, counter-rotating, carrying out form operation to obtain a target image, and extracting the leaf gray level and the leaf texture characteristic from the target image; and 4, selecting an optimal characteristic subclass through a GA-PLS (genetic algorithm-partial least square) process by fusing the gray scale and the texture characteristic and aiming at ten characteristic variables, and establishing a partial least square regression model based on the optimal characteristic, wherein the correlation coefficient R between a predicted value and a measured value of the model is 0.902. Compared with routine detection methods, the method of the invention has the advantages of rapid detection speed, and simple and convenient operation; and compared with a single near infrared spectroscopy or computer vision technical means, the method of the invention allows obtained information to be comprehensive, and the accuracy and the stability of the detection result to be improved.

An embodiment of the invention discloses a hyper-spectral estimation model constructing method of total nitrogen content of rice leaves. The method comprises steps as follows: multiple experimental plots are selected, and multiple sampling points are selected in each experimental plot; canopy spectral measurement is performed at the critical growing stage of rice; multiple sampling spectrums are recorded at each sampling point, and an average value is taken as a spectral measurement value of the sampling point; a hyper-spectral image of each experimental plot is acquired by an airborne imaging spectrometer; multiple function leaves at different parts are collected at each sampling point, and the total nitrogen content of the rice leaves is measured; the hyper-spectral estimation model of the total nitrogen content of the rice leaves is constructed with the adoption of spectral indexes or a partial least-squares regression method. The embodiment of the invention further discloses a hyper-spectral estimation method of the total nitrogen content of the rice leaves. The total nitrogen content of the rice leaves is estimated according to the model constructed with the method. The scientific and technical basis can be provided for space inversion of the nitrogen content of regional-scale rice and efficient implementation of precision agriculture.

The invention discloses a human face image face key point positioning method. The method includes the steps of marking to form a human face image training set with multiple face key points based on internet data; based on a human face detecting frame detected by a human face detector, aligning human face images, marked with the face key points, in the training set to obtain positions of the face key points of each human face image and average positions of the face key points of all the human face images after aligning; respectively learning distinguishing feature representation of each face key point and representing each face key point as a 10-dimesional feature vector; learning a regression positioning model of the face key points based on partial least squares regression; subjecting the inputted face images to be tested to human face detection and aligning, normalization, face key point position initialization and iterative regression processing to obtain final face key positions and output the final face key positions.

The invention relates to the field of electric energy quality monitoring, aiming to solve the problem that user-side harmonic pollution can not be correctly obtained in the prior art. The invention provides a method and a system for monitoring the user-side harmonic pollution. The method comprises the following steps of: collecting current data of a user-side electrifying apparatus by utilizing a current collecting device; collecting public connection point voltage data by utilizing a voltage collecting device; transmitting the current data and the public connection point voltage data to a system terminal; calculating the current data of emission harmonic of the user-side electrifying apparatus and an impedance value of the harmonic of the electrifying apparatus at the system terminal by utilizing a partial least-squares regression algorithm to acquire active power of the emission harmonic of the user-side electrifying apparatus; and sending alarming information when the active power of the emission harmonic exceeds the preset threshold. The invention has the advantages of conveniently monitoring the pollution to power grid by the user-side electrifying apparatus to reduce fluctuation of the power supply quality of the power grid.

Measurements of fluorescence spectra of fluid samples recovered downhole are processed to give the fluid composition. The processing may include a principal component analysis followed by a clustering method or a neutral network. Alternatively the processing may include a partial least squares regression. The latter can give the analysis of a mixture of three or more fluids.

A partial least squares (PLS) regression relates spin echo signals with samples having a known parameter such as bound water (BW), clay bound water (CBW), bound volume irreducible (BVI), porosity (PHI) and effective porosity (PHE). The regression defines a predictive model that is validated and can then be applied to spin echo signals of unknown samples to directly give an estimate of the parameter of interest. The unknown samples may include earth formations in which a NMR sensor assembly is conveyed in a borehole.

The invention relates to the field of detection of heavy metal pollution in soil , in particular to an unmanned-aerial-vehicle hyperspectral inversion based monitoring method of heavy metal pollution in soil. The method is characterized by including: field sampling; sample pretreatment; using an x-ray fluorescence analyzer to collect content of main research elements of heavy metal pollution sources of samples; using a field spectrometer to collect laboratory hyperspectral reflectance of the samples; performing data processing on original spectral reflectance data; performing correlation analysis on measured content of the main research elements respectively with laboratory hyperspectral original reflectance data, reciprocal, logarithms, first derivative data and second derivative data with a partial least squares regression algorithm and verification and modification on models to acquire an optimal transformation method, using an unmanned aerial vehicle equipped with a hyperspectral imaging spectrometer to collect research-area hyperspectral reflectance data as to-be-measured data, and performing extensive inversion of the heavy metal content. The unmanned-aerial-vehicle hyperspectral inversion based monitoring method has the advantages of large range detachable, non-invasive and non-contact quick sample detection and the like.

The invention discloses a method for fast predicting organic pollutant n-caprylic alcohol/air distribution coefficient based on molecular structure, belonging to the technical field of quantifying structure/active relationship (QSAR) facing to the environmental risk evaluation. The method is characterized of comprising the steps of: adopting the molecular structure of atomic center fragment characterization compound; and screening the atomic center fragment combination by means of stepwise regression and partial least-squares regression, to build a group contribution model for predicting KOA.The internal authentication and the external authentication improves that the built KOA group contribution model has stability and predicting capability, and a range and distance method and a probability density method express the application domain of the group contribution model, thereby defining the application range of the model and guaranteeing the predict accuracy. The method has the effectsand benefits of being capable of fast predicting the KOA of the high flux compound, obtaining the KOA with low cost, being helpful for obtaining the high flux KOA data, and having a significant meaning for the environment supervision and the risk evaluation of chemicals.

The invention discloses a method for optimizing wavelength variable in spectral analysis. The method comprises the steps as follows: obtained original spectrum is pretreated to obtain a spectral array with useless information eliminated; the purity value of each wavelength variable is calculated in the obtained spectral array to select the wavelength variable with maximum purity value as a first wavelength variable; the relative weighting function of no. j wavelength variable and selected (j-1) wavelength variables is calculated, and the purity value of each wavelength variable after the relative weighting function is added is calculated; the wavelength variable with the maximum purity value is selected as no. j wavelength variable, wherein, j is the integral more than or equal to 2; partial least square regression modeling is carried out by optimized different quantities of the wavelength variables, and predicted root mean square error is calculated; when the predicted root mean square error is minimum, the wavelength variable combination selected for modeling is the optimized wavelength variable combination. The quantity of the wavelength variables selected by the method is small, and the method can minimize redundant information and can improve modeling speed and efficiency obviously.

The invention belongs to the field of power quality managing, and discloses a method for applying the partial least squares regression to power distribution network harmonic source positioning and detecting. According to a model for carrying out equivalent transformation on the electrical power system side and the distortion user side and a harmonic voltage and a harmonic current signal which are obtained through synchronous measurement at the common coupling joint, a regression coefficient is dissolved through the partial least squares regression algorithm, an aggregative variable with the best explanatory capability for a dependent variable is extracted in the manner of decomposing and screening data information in the system, and information and noise in the system are identified. The method integrates the basic functions of multiple linear regression, the canonical correlation analysis and the principal component analysis, the modeling predicting type data analysis method and non-model type data recognition are organically combined, regression modeling, data structure simplifying and variable correlation analyzing can be achieved at the same time, and the method can be widely applied to power quality analyzing, monitoring, evaluating and controlling fields.

The invention discloses a Thevenin equivalent parameter identification method based on partial least squares regression. The Thevenin equivalent parameter identification method based on the partial least squares regression is based on the situation that part equivalent processing is conducted on a complex network below a certain power transmission section, conducts linear regression using current time section data and a plurality of groups of state data after part equivalent processing, and can solve the problem of unstable regression coefficients and the problem that because independent variables have strong relevance, regression parameters drift. The method is simple, practical, strong in identification result effectiveness, and fast in calculating speed, and does not have an initial value selection problem. The Thevenin equivalent parameter identification method based on the partial least squares regression comprises the following steps that first, a power transmission section is selected, a person looks at a system from any loading node i, and the system and a system with two nodes are equivalent; second, part equivalent processing is conducted on a power grid, and a Thevenin equivalent model of a loading node in an operation work place is acquired through the application of a partial least squares regression method.

The invention discloses a method for determining a radix notoginseng extract and contents of five types of ginsenosides in a preparation of the radix notoginseng extract by a Fourier transform near-infrared spectrograph. The method comprises the steps that a sample is detected by an ultra-high performance liquid chromatography, moving phase builds acetonitrile and water gradient elution parameters with separation degrees higher than 1.7 in chromatographic peaks of Rg1 and Re, and the spectrum preprocessing adopts the combination of 2 to 3 methods in Savitzky-Golay polynomial smoothing, second-order differential, Noriis derivative filtering and data normalization; and three preferred wave bands of the five types of ginsenosides are modeled, and a calibration model is built by any one of a partial least squares regression method, a principal component regression method and a multiple linear regression method. The method is used for determining the contents of the five types of ginsenosides in the sample to be detected, the determining result is consistent with a result determined by the ultra-high performance liquid chromatograph basically, the requirements of 'Chinese pharmacopoeia' are met, the accuracy is high, and the determination speed is increased substantially.

The invention discloses a method and a device for online detecting quality index of yellow rice wine in bottle. The online detection device comprises an illumination system, an air conditioning system, a conveying belt, a receiving lens fixing device, a receiving lens, an optical fiber, a spectrometer, a computer, a trigger circuit and a photoelectric switch. The method comprises the following steps of: collecting the near IR spectrum of an empty bottle of a yellow rice wine sample in the bottle as the background spectrum; reducing the background spectrum from the spectrum of an original yellow rice wine sample in bottle as the sample spectrum; constructing the correction model of the key quality index of yellow wine according to the spectra and a physical and chemical analysis result by adopting partial least-squares regression method; and online detecting the quality of the yellow rice wine according to corresponding model. Based on near IR spectral analysis technology, the invention can achieve online detection of the key quality indexes of the yellow rice wine sample, including total sugar, solids, alcohol degree, total acid, amino acid nitrogen as well as pH value, thus achieving online and rapid quality analysis of yellow rice wine product, reduced detection time and cost and ensured product quality.

The invention discloses a method for measuring effective component content in kuhuang injection on line by near-infrared spectroscopy quickly. Effective components in kuhuang injection mainly comprise aloe-emodin, rhein, emodin, chrysophanol, physcion, sophocarpine and matrine so on. Near-infrared absorption spectroscopy collecting kuhuang injection make data of effective component content relate with near-infrared spectroscopy by HPLC method. Calibration model is built by partial least-squares regression method. Quick on-line measurement for effective component in injection can be accomplished. Studying result shows that near-infrared spectroscopy analyzing method can measure effective component in kuhuang injection effectively. The method is provided with simple pretreatment for sample, quick measurement and measuring multiple components at the same time. It can be used in on-line analysis and on-line quality control in tcm manufacture process.

The invention relates to a method for detecting the content of a pesticide in cereal by utilizing a terahertz time-domain spectroscopy. The method comprises the following steps of grinding and then tabletting a to-be-detected cereal sample to obtain a to-be-detected cereal tablet sample, testing the to-be-detected cereal tablet sample by directly adopting a terahertz time-domain spectroscopy system, acquiring a terahertz area spectrum signal to be used as a sample signal, acquiring a terahertz time-domain spectrum signal under a nitrogen atmosphere as a reference signal, acquiring an absorptive coefficient spectrum of the to-be-detected cereal tablet sample, acquiring a characteristic wave band of a target pesticide, classifying the absorptive coefficient spectrum of the to-be-detected cereal tablet sample to a corrected set of sample absorptive coefficient spectrum and a verification set of sample absorptive coefficient spectrum, establishing a quantitative analysis model by utilizing a partial least-squares regression method, and acquiring a quantitative detection value of each to-be-detected cereal sample. The method can truly and effectively detect the pesticide residue in the cereal quantitatively, rapidly and accurately, and the prediction square correlation coefficient (R2) of the quantitative analysis model reaches up to 0.9903.

The invention discloses a spectrum wave number selection method. Correction samples are randomly sampled many times according to wave numbers of spectra, a partial least square regression model is built, a variable projection importance coefficient of each wave number is calculated, the variable projection importance coefficients are sequenced in a descending mode, wave number sets corresponding to the distribution sequence are obtained, the re-sequenced wave number sets are subjected to wave number screening step by step, a result of each time of wave number selection is calculated, and a wave number primary selection set is obtained; then, absolute values of a partial least square regression coefficient of each wave number in the wave number primary selection sets is counted and correspondingly processed, the processed partial least square regression coefficients are sequenced in a descending mode, a corresponding wave number distribution sequence is recorded, then a strategy for reversely eliminating wake-correlation wave numbers is adopted, and the optimal feature wave number set is obtained. The method can fully mine effective information in wave numbers, effectively solves the subjectivity problem of wave number selection, extracts feature wave numbers to the largest extent, eliminates influences of weak correlation factors, and remarkably improves robustness and precision of the model.

The invention discloses a dredging operation yield prediction model building method based on partial least squares regression. An advanced multivariate regression analysis method is adopted, the high dimensional data space of an independent variable and the high dimensional data space of a dependent variable are projected to corresponding low dimensional characteristic spaces, the mutually orthogonal feature vectors of the independent variable and the dependent variable are obtained respectively, and then the linear regression relation between the feature vectors of independent variable and the dependent variable is built. When the feature vectors are selected, the explanation and predication function of the independent variable on the dependent variable is emphasized, the influence of noise useless to regression is removed, the model comprises the minimum number of the variables, and therefore the model has good robustness and prediction stability. A theoretical foundation can be laid for the optimization study of the dredging operation yield, the aims of high efficiency, high yield and low energy consumption are achieved, and the method has great significance in production predication of a dredger.

The invention discloses a novel harmonic emission level assessment method. Harmonic equivalent circuits of a system and a user at a point of common coupling (PCC) are analyzed to obtain a relational expression between the harmonic impedance of the system and the user and the voltage and current at the PCC; a certain mathematical operation is conducted on voltage signals and current signals, wherein the voltage signals and the current signals are collected at the PCC; h subharmonic voltage signals and current signals are extracted; on the basis of data envelopment analysis of the signals, partial least square regression molding is conducted to obtain the estimated result of the harmonic impedance; lastly, the regression result is utilized for assessing the harmonic emission level of the user. According to the novel harmonic emission level assessment method, the data envelopment analysis is adopted for optimizing harmonic voltage and current values obtained through measurement at the PCC, invalid data caused by system operation fluctuation or a measuring instrument are eliminated, the defect that partial least square regression obtains a wrong model according to wrong data is overcome, the accuracy of the estimated result is greatly improved, and a basis is provided for duty allocation between the system and the user with respect to the power quality contamination status.

The invention relates to a hyperspectral-image-technology-based multi-quality nondestructive testing method for dried green soybeans. According to the technical scheme, the method comprises the steps of: a. selecting dried green soybeans; b. collecting hyperspectral images of the dried green soybeans by utilizing a hyperspectral image collecting system; c. extracting contour information of the dried green soybeans by utilizing a threshold segmenting method; d. extracting image entropy characteristic parameters by utilizing the obtained contour information; e. collecting the color, moisture rate, hardness and shrinkage rate indexes of the dried green soybeans by utilizing a destructive instrument; f. constructing an evaluation prediction model of the dried green soybeans by utilizing a partial least-squares regression algorithm; g. collecting the hyperspectral images of the dried green soybeans and inputting the hyperspectral images of the dried green soybeans to the evaluation prediction model to obtain the evaluation results on the color, moisture rate, hardness and shrinkage rate indexes of the dried green soybeans. Through the evaluation prediction model and the hyperspectral image collecting system, a multi-quality evaluation result can be obtained under the situation of being nondestructive for the majority dried green soybeans; the hyperspectral-image-technology-based multi-quality nondestructive testing method is easy to operate, good in real-time property and high in reliability.

The invention discloses a method for detecting the wine age of yellow rice wine. Its steps are as follows: 1) use the electrode array of the rice wine age detector to react with the rice wine to obtain a response signal; 2) detect the pH value, conductivity and total sugar content of the tested sample; 3) test the rice wine age detector Select and extract eigenvalues ​​from the original data; 4) use the eigenvalues ​​as the response signal of the electrode array, apply the eigenvalues ​​to establish the principal component analysis and cluster analysis pattern recognition model, and combine the pH value, conductivity and total sugar content to analyze the 5) Reduce the dimensionality of the eigenvalues ​​through principal component analysis, use the data after dimensionality reduction as the independent variable of partial least squares regression and artificial neural network to establish a prediction model, apply partial least squares regression and artificial neural network Quantitative prediction and analysis of rice wine age by network. The invention simplifies the experimental steps, reduces the interference, and greatly improves the sensitivity, reliability and repeatability.

The invention relates to a soft measurement method based on near infrared spectroscopy, and especially relates to a soft measurement method, based on near infrared spectroscopy, for relative density in a compound donkey-hide gelatin pulp medicinal material extract concentration process. According to the invention, a sample set is formed by production process samples and samples prepared in a laboratory; near infrared spectra of the sample set are collected; abnormal samples are rejected; characteristic spectral information of concentrated liquid is extracted by selecting appropriate spectral modeling bands and pre-processing methods; a relative density of the concentrated liquid determined by a pycnometer method is used as a reference value; a quantitative correction model of the relation between the concentrated liquid near infrared spectra and the relative densities is established by a partial least squares regression method; the near infrared spectrum of a sample to be determined is collected; and the relative density of the sample is obtained by using the established model. The invention provides a rapid and precise density soft measurement method which facilitates the improvement of the quality control level of the production process.

The invention belongs to the technical field of spectral nondestructive detection of crop biochemical components, and discloses a crop canopy leaf total nitrogen content estimation method. The methodcomprises steps: firstly, a successive projection transformation algorithm SPA is used to construct a sensitive spectral feature data set, a position feature data set and a vegetation index feature data set; then, a sensitive spectral variable set with the minimum collinearity is selected from the sensitive spectral feature data set, the position feature data set and the vegetation index feature data set; and finally, the sensitive spectral variable set is subjected to partial least squares regression PLS modeling to obtain a crop canopy leaf total nitrogen content LNC estimation model. The model can accurately estimate the crop canopy leaf total nitrogen content, which is not much different from the measured value.

The invention relates to multiband reflection spectrum noninvasive blood component measuring device and method. The device is provided with a scanning device for a tongue body by utilizing a light source and a spectrum receiving device for receiving a scanning signal of the scanning device, wherein the light source is a supercontinuum light source or wave band; the scanning device carries out scanning by adopting optical fibers, and the like; and the spectrum receiving device adopts a spectrograph or a hyperspectral imager. The method comprises the following steps of: scanning the reflection spectrum distribution of the whole tongue body surface by utilizing the light source through the optical fibers; obtaining a reflection spectrum generated by the tongue body by the spectrum receiving device; calculating a multiband normalized reflection spectrum of blood; establishing a mathematical model by utilizing methods of main component analysis, artificial neural network, partial least-square regression, support vector machine signal analysis and statistics to obtain normalized spectrum data; and calculating by applying the established mathematical model to obtain the blood component content. The invention resists interference, can obtain more spectrum information and the image information of a measured part, increases information amount and improves the analysis precision of the blood component content.

The invention provides a near infrared spectrum variable selection method based on LASSO. The method comprises the following concrete processes of collecting a near infrared spectrum of a sample, and using a conventional method for measuring a concentration vector of a tested ingredient; dividing a data set into a training set and a prediction set by adopting a certain grouping mode; determining the constraint value t of an LASSO method through crossed verification; using a minimum angle regression algorithm for calculating the regression coefficient beta; remaining the position of a wavelength point of the beta being not zero; building a partial least squares regression model between a training set spectrum and the concentration vector by utilizing the training set spectrum corresponding to the remained wavelength, and predicting the concentration of a tested ingredient of a prediction set sample. The method has the advantages that an effective wavelength can be extracted; a quantitative analysis model is simplified; the prediction precision of the model is improved. Compared with an existing variable selection method, the method has the advantages that the speed is high; the repeating performance can be realized; the higher prediction precision can be reached by using fewer variables. The near infrared spectrum variable selection method is applicable to the variable selection of complicated sample near infrared spectrums.

The invention belongs to the technical field of Wuxi Hao Tea quality detection, and discloses a detection method for rapid grade appraisal of Wuxi Hao Tea. Firstly, aroma of Wuxi Hao Tea is gathered through the headspace solid-phase microextraction (HS-SPME) technique, and the collected aroma is analyzed through gas chromatography-mass spectrometry (GC-MS), so as to construct a Wuxi Hao Tea aroma standard fingerprint spectrum with 35 common peaks; secondly, sensory scores are given to Wuxi Hao Tea samples through sensory evaluations; thirdly, a partial least squares (PLS) regression model for Wuxi Hao Tea aromatic quality is established in a manner that relative contents of the 35 common aromatic components in the Wuxi Hao Tea aroma standard fingerprint spectrum serve as an X matrix and the aromatic sensory scores serve as a Y matrix, the correlation coefficient of the model is 0.973, and the cross-validation correlation coefficient of the model is 0.947. The model has a high fitting degree, can better achieve the aromatic quality predication for Wuxi Hao Tea, and can be used for rapid grade appraisal of Wuxi Hao Tea.

The invention relates to a wavelength similarity consensus regression-based infrared spectrum quantitative analysis method and a wavelength similarity consensus regression-based infrared spectrum quantitative analysis device. The device comprises a spectrometer, a preprocessor, a wavelength filter and a partial least-squares regression analyzer which are connected through a data signal line; and the wavelength of the near-infrared spectrum ranges from 780 to 50,000nm. By performing cluster analysis on spectra along the wavelength direction, a spectrum is divided into different information blocks, multiple models are constructed, and the difficulty for a single model method to extract information of a spectrum signal is overcome; and by singly determining factor number for different submodels, effective information is fully extracted, and the prediction accuracy and robustness of an infrared spectrum analysis model are improved.

A volumetric segmentation method is disclosed for brain region analysis, in particular but not limited to, regions of the basal ganglia such as the subthalamic nucleus (STN). This serves for visualization and localization within the sub-cortical region of the basal ganglia, as an example of prediction of a region of interest for deep brain stimulation procedures. A statistical shape model is applied for variation modes of the STN, or the corresponding regions of interest, and its predictors on high-quality training sets obtained from high-field, e.g., 7T, MR imaging. The partial least squares regression (PLSR) method is applied to induce the spatial relationship between the region to be predicted, e.g., STN, and its predictors. The prediction accuracy for validating the invention is evaluated by measuring the shape similarity and the errors in position, size, and orientation between manually segmented STN and its predicted one.