25 results about "Calibration set" patented technology

The invention relates to the field of pipeline defect measurement, in particular to a method for measuring the position from a pipeline defect to a pipe orifice based on binocular vision. Internal and external parameter calibration of the binocular camera is completed through the calibration module, the binocular vision acquisition module enters a pipeline to obtain a binocular image sequence containing pipe orifices and defects, and precise solving of internal parameters and external parameters of the camera is achieved by adopting a checkerboard and multi-attitude calibration set and combining a Zhang Zhengyou calibration method. The image is preprocessed, the image quality is improved, dust noise in a large pipe diameter is removed, and edge features are enhanced. On the basis of Canny edge detection, a least square method and a YOLOv8 algorithm are matched to complete pipe orifice reference and defect feature collaborative identification, and efficient stereo matching and three-dimensional reconstruction of defect points are achieved in combination with SGBM stereo matching and triangulation. Finally, the problems that traditional pipeline defect positioning depends on manual work and is low in precision and poor in adaptability are solved, and automatic and high-precision measurement from the defects in the pipeline to the pipe opening position is achieved.

This invention provides a method and system for rapid online detection of fertilizer nutrients based on near-infrared spectroscopy. The method includes: acquiring the online near-infrared spectrum and corresponding nutrient concentration data of the fertilizer to be tested, constructing and preprocessing a calibration set; screening characteristic wavelengths through multiple rounds of iteration: in each round, a sample subset is generated by randomly sampling from the calibration set, and a partial least squares regression model is established; determining the importance weight of each wavelength, and determining the number of wavelengths to retain by combining an exponential decay function, and screening high-weight wavelengths to obtain a candidate subset; evaluating the accuracy of the candidate subset using optimized k-fold cross-validation: calculating the leverage value and Mahalanobis distance of all samples, mapping them to a two-dimensional space and dividing them into grids; sampling and grouping by grid as a stratum, and calculating the root mean square error of cross-validation; comparing all iteration results, selecting the candidate wavelength with the smallest error as the optimal characteristic spectrum; and using the optimal characteristic spectrum and nutrient concentration data, training a quantitative prediction model for fertilizer nutrients to achieve rapid detection.

The invention belongs to the technical field of vegetable quality detection, and particularly relates to an establishment method and a detection method of a tomato fructose content prediction model. The method comprises the following steps: firstly, collecting near infrared spectrums of tomato samples, dividing the tomato samples into a calibration set and a verification set, and removing abnormal values; secondly, comparing and screening the near infrared spectrums by using different pretreatment methods; carrying out dimension reduction treatment on the spectrum by adopting biPLS-CARS, and extracting a near infrared spectrum characteristic wavelength of the fructose content index; and finally, establishing a tomato fructose content prediction model by utilizing the selected characteristic wavelength through a partial least square method and combining with the absolute content of fructose. The established model can accurately predict the fructose content in the tomato, the operation process is simple, rapid and environment-friendly, and a new technical means is provided for detection of the fructose content in the tomato.

The invention discloses a method for rapidly detecting flavonoid components and moisture in persimmon leaves and products thereof. The method comprises the following steps: measuring the content of six flavonoid components and the content of moisture in a calibration set sample; scanning the persimmon leaf sample by using a near infrared spectrometer to obtain near infrared spectrum data; importing the obtained near infrared spectrum data of the sample into Brucker infrared spectrometer OPUS 6.5 software, analyzing the spectrum information, establishing a near infrared spectrum model by taking the spectrum information which is not preprocessed as a contrast and combining a partial least square method, and constructing a correction model of the relationship between the six flavonoid components and the moisture content of the persimmon leaf sample and the characteristic near infrared spectrum data; near infrared spectrum data of persimmon leaves to be detected are collected according to spectrum collection parameters the same as those of samples in the calibration set, the six flavonoid components and the moisture content in the persimmon leaves to be detected are obtained according to the constructed calibration model, the detection work efficiency is greatly improved, and comprehensive control over the product quality is facilitated.

The application provides a Dendrobium officinale powder release control method, relates to the technical field of drug quality control, and comprises the following steps: acquiring hyperspectral images of Dendrobium officinale powder under different temperature gradients and correcting; extracting average spectra and preprocessing by using an SNV algorithm; determining polysaccharide content; dividing a calibration set and a test set; extracting characteristic wavelengths by using a CARS algorithm; establishing a PLSR model under different temperature gradients; taking the characteristic wavelengths under all temperature gradients as input, and establishing a SO-PLS model; and determining whether to release or not by evaluating the probability risk of test set samples exceeding a control limit through bootstrap sampling. The application combines temperature control with hyperspectral technology, significantly improves prediction accuracy, introduces a probability risk evaluation mechanism, guarantees quality control, improves detection efficiency, and effectively solves the problems of long time consumption and low efficiency of traditional detection methods.

The invention provides a quantitative analysis method for simultaneously determining essential oil and total flavonoids in illicium verum and fennel based on a near infrared spectrum quantitative analysis model, and relates to the technical field of illicium verum and fennel determination, and the quantitative analysis method comprises the following steps: S1, collecting illicium verum and fennel samples with different attribute information, and constructing a sample archive; s2, acquiring near infrared spectrums of illicium verum and fennel samples in the sample archive, and dividing the illicium verum and fennel samples into a calibration set and a verification set according to a Kennard-Stone method; s3, calculating the coverage degree from the correction set to the verification set, judging whether the coverage degree exceeds a threshold value or not, if so, turning to step S4, and if not, turning to step S3; if not, supplementing the sample according to error analysis; s4, acquiring a universal model which is established on the basis of the calibration set and can simultaneously predict the contents of the essential oil and the total flavonoids in the star anise and fennel samples, inputting the near infrared spectrum of the verification set into the universal model, calculating an error between a prediction result and a corresponding reference content, and performing sample supplementation according to error analysis.

A calibration method is disclosed for use in an analytical method for detecting at least one analyte in a sample of bodily fluid by using a mobile device (110) having at least one camera (112) and at least one ambient light sensor (114), the calibration method comprising: a. Providing a set of illumination conditions; and b. Determining a calibration set comprising a calibration function (152) for each of the lighting conditions of the set of lighting conditions, each calibration function (152) provides a relationship between at least one optical information item derived from at least one image captured with the camera (112) from at least a portion of at least one test field (120) of an optical test element (118) under respective illumination conditions and a concentration of the analyte in a sample applied to the test field (120). Further disclosed is a computer-implemented analysis method as well as a mobile device (110), a computer program and a computer-readable storage medium for performing the analysis method.

This invention relates to a method and apparatus for quantitative analysis of cement raw meal components based on feature screening and fusion, belonging to the field of cement component analysis technology. The invention acquires and preprocesses near-infrared spectral data and laser-induced breakdown spectral data to eliminate light scattering effects, noise interference, and differences in data dimensions. A concentration gradient method is used to divide the preprocessed dataset into a calibration set and a validation set. A predefined feature screening method is used to screen target feature points affecting the quantitative analysis of each target component from the calibration sets of the near-infrared spectral data and the laser-induced breakdown spectral data, respectively. An equal-weighted fusion method is used to perform feature-level fusion of the target feature points of the near-infrared spectral data and the laser-induced breakdown spectral data to construct a fused feature spectrum. Based on the fused feature spectrum, a partial least squares regression algorithm is used to establish quantitative analysis models for each target component to be measured. Through complementary spectral advantages, the accuracy of quantitative analysis is significantly improved.

The application discloses a discrimination method and system for sorghum varieties based on decision fusion and near-infrared spectroscopy, and the method comprises the following steps: collecting and processing samples of sorghum grains; collecting near-infrared spectra of each sample to construct a calibration set; determining the tannin content and amylopectin content of each sample; constructing a first quantitative detection model between the near-infrared spectra of the samples and the tannin content, and constructing a second quantitative detection model between the near-infrared spectra of the samples and the amylopectin content; constructing a qualitative discrimination model between the near-infrared spectra of the samples and the category attribution; predicting the calibration set respectively to obtain three groups of prediction values, taking the three groups of prediction values as independent variables, and taking the category labels of the samples as dependent variables to construct a decision fusion model; for unknown sorghum samples, the tannin content and amylopectin content of the samples are predicted respectively, the qualitative discrimination model is used to predict the category attribution, and finally, the category prediction value is obtained; the method has the advantages of high precision and strong robustness.

The application relates to the field of Beidou system technology, and particularly discloses a radar detection method and system based on a Beidou system, which comprises the following steps: establishing a digital twin model of a ship and determining positions A and B; controlling the posture of the digital twin model to transform from an angle combination i to an angle combination f; determining a compensation set when the posture of the ship transforms from the angle combination i to the angle combination f based on the deviation of the positions A and B; obtaining a calibration set according to an angle combination Z1 of a current moment and an angle combination Z2 of a previous moment; and correcting Beidou positioning data and radar positioning data according to sub-angle combinations J1 and J2 in the calibration set. The application can eliminate the posture deviation of the Beidou and the radar caused by the installation position difference under the condition that the ship body is rocked by wind and waves, and significantly improves the precision and stability of offshore target positioning.

The wavelength selection method in near-infrared quantitative analysis provided by the application adopts a hyperspectral imaging system to obtain a hyperspectral image of a standard reference object, and adopts a micro spectral sensor to obtain a visible-near-infrared reflection spectrum of an analysis object. For the hyperspectral image of the standard reference object, second-order difference operation and smoothing denoising are performed on the spectral data to obtain a standard reference object data set; for the visible-near-infrared reflection spectrum of the analysis object, abnormal data is removed, a Monte Carlo sampling method is used for analysis, then a wavelet packet decomposition and reconstruction method is executed for data smoothing to obtain an analysis object data set. In the modeling process, the classification criteria of different data sets are fully mined, and the Euclidean distance similarity is judged, so that different classes of data sets are objectively obtained, and the largest and smallest data sets of the two classes are selected as the calibration set and the prediction set, respectively, so that better modeling effect can be achieved.

This invention relates to the field of optical thickness measurement technology, specifically a laser-based ice thickness measurement method, comprising the following steps: coaxially superimposing first and second band laser beams to lock the initial phase difference and generate a calibration set; analyzing the reflection and transmission at the beam surface and bottom interface and capturing the waveform using a detector; separating the time-of-flight difference and comparing the propagation delay of the two bands to generate a feature matrix; extracting the power attenuation after propagation and combining it with dynamic refractive index to correct the light speed and obtain the geometric thickness result. In this invention, by constructing a dual-band interface reflection analytical mechanism, the denoising capability of non-uniform beam tracking is enhanced, effectively solving the distortion interference caused by multipath scattering; separating inherent band time delay jitter and reducing signal attenuation caused by medium loss; and combining refraction compensation attenuation correction to correct the physical space solution accuracy, eliminating bubbles and tilt angle optical path errors, ensuring detection purity, and greatly improving the stability and monitoring accuracy of ice layer physical thickness reconstruction.

This invention discloses a calibration method for visual language models adapted during testing based on a single preparation. It utilizes a generative model to offline synthesize a small number of images for each category in the target domain, extracts features, and divides them into a pre-filled set and a calibration set. The pre-filled set is used to initialize positive and negative sample caches, achieving a "warm start." A lightweight probe network is trained based on the calibration set, which dynamically predicts the optimal fusion weights based on the original prediction scores and cached retrieval scores. During the testing phase, the network generates weights in real time and adaptively weights and fuses multiple predictions to obtain the calibrated prediction results. This invention requires only a one-time offline preparation, with extremely low online inference overhead. While maintaining efficient inference, it significantly reduces calibration errors, improving the reliability and security of model predictions, making it particularly suitable for real-time applications with strict confidence requirements.

The application discloses an anaerobic fermentation instability early warning method based on near-infrared transmission spectrum, which collects near-infrared transmission spectrum data of biogas slurry in an anaerobic fermentation impact test and corresponding biogas slurry samples online, determines the spectrum data corresponding anaerobic fermentation state according to the biogas slurry index determination results and the fermentation state threshold of the propionic acid / acetic acid ratio and the volatile fatty acid / total alkalinity ratio, performs wavelength variable primary selection on the preprocessed calibration set spectrum data by using linear classifier integrated wavelength selection fusion mutual information nonlinear feature supplement, simultaneously optimizes the characteristic wavelength and the extreme gradient boosting modeling hyperparameter by using a genetic annealing particle swarm algorithm, and constructs an anaerobic fermentation state intelligent diagnosis model, so that the anaerobic fermentation instability early warning can be completed in combination with host computer software and a 4G SMS cat. The method has unmanned online monitoring function, can realize more than 3-minute sampling interval full-automatic anaerobic fermentation state intelligent diagnosis and instability early warning, and further ensures efficient and stable operation of a biogas project.

This invention discloses a method for constructing a near-infrared identification model for Terminalia chebula medicinal materials. The method includes: collecting a first predetermined number of Terminalia chebula medicinal material samples from the original Terminalia chebula bud, Terminalia chebula velutipes bud, and Terminalia chebula fruit, and performing near-infrared spectral measurements to obtain an initial spectral set; establishing a calibration set and a validation set; preprocessing the sample spectra within the calibration set to obtain a preprocessed spectral set; selecting wavelengths from the preprocessed spectral set to obtain a first wavelength group or a second wavelength group, wherein the first wavelength group is used to identify Terminalia chebula medicinal materials at different maturity stages, and the second wavelength group is used to identify Terminalia chebula medicinal materials at different maturity stages and from different buds; and establishing a model within the spectral range of the first wavelength group and / or the second wavelength group. This invention utilizes near-infrared spectroscopy technology, and the sample preparation process is simple, non-destructive, and free from chemical reagent contamination.

The present application relates to the field of computer vision, and particularly relates to a method for obtaining an offline quantization calibration set of a face detection model, after obtaining a face detection dataset, each sample in the face detection dataset is classified according to different scenes, and the scene quality label, the face density label, the face average area label, the face average orientation label and the face average quality label corresponding to each sample in the face detection dataset are determined, then a plurality of samples meeting the requirements are screened out from the face detection dataset to form the offline quantization calibration set of the face detection model. By marking and storing each sample in the face detection dataset, the present application automatically selects a plurality of representative and diverse samples to form the offline quantization calibration set of the face detection model, which greatly improves the work efficiency compared with manually preparing the offline quantization calibration set of the face detection model, and is simple to operate and has strong scalability.

The invention discloses a method for determining the content of cyclo-hexogen and aluminum powder in a PBX explosive, and belongs to the technical field of explosive component measurement. The method comprises the following steps: dividing m to-be-tested PBX explosive samples into m1 calibration set samples and m2 verification set samples; collecting near infrared data of all samples; preprocessing all collected near-infrared data, and then screening variables in a full spectrum by adopting a joint interval partial least square method; establishing a correction model by adopting a cross test method, and selecting an optimal model variable according to a minimum value of a root-mean-square error of data cross validation of the samples in the correction set to obtain a hexogen recommendation model and an aluminum powder recommendation model; respectively predicting the content of the hexogen and the content of the aluminum powder in the verification set sample according to the hexogen recommendation model and the aluminum powder recommendation model, and selecting an optimal model according to the minimum value of the root-mean-square error predicted by the verification set. The method is suitable for determining the contents of cyclo-hexogen and aluminum powder in the PBX explosive, and is convenient to operate and rapid and accurate in analysis.

The application discloses a correction set sample adding method for near-infrared detection of distillate oil properties. The method is used for the case of production condition change in near-infrared spectrum modeling of oil properties, judges the change trend of oil properties based on a spectrum gradient angle, and automatically adds samples with the same trend change to a correction set, so that the situation that the prediction accuracy is reduced due to the deviation of new samples from the distribution area of original correction set samples after modeling for a period of time can be effectively avoided. The method can update the correction set in time, ensure the prediction accuracy of the model, is particularly suitable for the case of production condition change of a refining and chemical enterprise, and has important application value for real-time control and optimized production of an enterprise.

The invention relates to the technical field of optical fiber communication, in particular to a distributed sound wave sensing anomaly detection method based on self-supervised conformal inference, which comprises the following steps: firstly, carrying out preprocessing and robust normalization on a spatio-temporal data matrix demodulated by a DAS; constructing a causal time sequence prediction model to carry out quantile prediction, and carrying out self-supervised training through quantile regression by only using normal data; constructing a non-consistency score; a calibration set only containing normal data is used for calculating conformal calibration, and at the moment, it is judged that point anomalies exist; carrying out space-time connected domain analysis and identification on the abnormal points; extracting physical interpretable attributes of the event, namely physical characteristics (energy, dominant frequency and apparent velocity); and finally, outputting a detected abnormal event result. According to the method, the unknown type of anomaly can be detected without abnormal sample training, the adaptability to the dynamic environment is high, the false alarm rate is controllable, the interpretability is good, and the method is suitable for online anomaly detection and event interpretation under the strong non-stationary background.

This invention provides a method for detecting the boron isotope ratio and total boron concentration in solution based on microwave plasma emission spectroscopy. The method includes: directly exciting a boron-containing solution sample to be tested using a microwave plasma excitation source; obtaining the boron atom emission spectrum and BO2 molecule emission spectrum of the boron-containing solution sample after subtracting the blank spectrum; configuring a standard solution calibration set; acquiring the spectral data of the standard solution calibration set after subtracting the blank spectrum; establishing a calculation model based on the spectral data; inputting the boron atom emission spectrum or BO2 molecule emission spectrum of the boron-containing solution sample to the calculation model; and outputting the boron isotope ratio or total boron concentration of the boron-containing solution sample. This method for detecting the boron isotope ratio and total boron concentration in solution based on microwave plasma emission spectroscopy can meet the needs of online or real-time analysis.

The application discloses a grazing sheep chilled meat tenderness and water retention detection method based on hyperspectral imaging, relates to the field of nondestructive testing technology of agricultural and livestock products, and comprises the following steps: acquiring a hyperspectral image of grazing sheep chilled meat, establishing a calibration set in combination with shear force and water retention experimental data, extracting a region of interest and performing spectral pretreatment, adopting characteristic wavelength screening and full-waveband comparative analysis to construct a multi-index prediction model based on partial least squares regression, finally adopting a WT-PLSR model to predict and output tenderness, adopting an SG-PLSR model to predict and output cooking loss, and adopting a DT-PLSR model to predict and output drip loss, so that synchronous nondestructive quantitative detection of chilled meat tenderness and water retention indexes based on hyperspectral information is realized.

The application discloses a distillate oil near-infrared spectrum online property detection method for working condition switching, which selects samples similar to the to-be-detected distillate oil as a calibration set to improve the property detection accuracy of the to-be-detected distillate oil. Three similarity measurement methods are adopted to measure from three aspects of Euclidean distance, spectral information divergence and spectral gradient angle, the matching degrees of the measurement methods and the distillate oil spectrum are learned, spectral feature information is extracted in different directions, the to-be-detected spectrum is represented to the maximum extent, and the recognition accuracy of the online detection model is improved to cope with complex working condition changes.

The invention discloses a Wuyi narcissus baking color quality intelligent discrimination method and system based on a near infrared spectrum optimized by a genetic algorithm. The near infrared spectrum of a Wuyi narcissus raw tea sample at the wave band of 400-4000 cm <-1 > is collected, in combination with CIE-Lab chromaticity parameters, Savitzky-Golay smooth filtering and multiple scatter correction combined preprocessing is adopted to eliminate noise interference, key characteristic wavelengths are screened by using a genetic algorithm to construct a wavelength-chromaticity incidence matrix, and the wavelength-chromaticity incidence matrix is used for detecting the wavelength-chromaticity of the Wuyi narcissus raw tea. Partial least squares regression (PLS) and a support vector machine (SVM) are coupled to construct a GA-PLS-SVM hybrid prediction model, and accurate inversion of baking intensity and time parameters is realized. The model calibration set L * / a * / b * prediction R is larger than or equal to 0.90, the verification set R is larger than or equal to 0.85, 82.3% of baking process deviation can be recognized online, and the rock tea standardized production efficiency and the process control precision are remarkably improved.

The application discloses a fruit grading quality detection method based on near-infrared diffuse reflection spectrum characteristics, which can be applied to grading detection of partial internal quality of fruits with thin peels in combination with possibility theory. The method constructs reference core area and reference extension area and corresponding center value and boundary value of wavelength and light intensity of each grade standard partial internal quality corresponding wave trough or wave peak through fruit prediction set and fruit calibration set analysis, and then judges whether the data set composed of wavelength and light intensity of wave trough or wave peak of the fruit needing grading quality detection is a subset of the domain of a certain grade standard by possibility theory, and judges the belonging grade standard with the maximum possibility measure, so as to provide a basis for fruit grading from the perspective of partial internal quality.

The invention relates to the technical field of food detection, in particular to a method for detecting chemical pollutants in food based on an infrared spectrum. The method for detecting the chemical pollutants in the food comprises two stages of model establishment and sample detection. In the model building stage, target pollutant calibration set samples with different concentrations are prepared, infrared spectrums of the target pollutant calibration set samples are collected, and a quantitative calibration model is built in combination with a chemometrics method after pretreatment. A sample detection stage: performing the same pretreatment on a to-be-detected sample, collecting a spectrum, inputting the spectrum into the established model after the same pretreatment, and directly predicting the pollutant concentration of the to-be-detected sample; according to the method disclosed by the invention, by optimizing spectrum pretreatment and chemometrics modeling, the interference of a complex matrix of food is overcome, and rapid, lossless and accurate detection on chemical pollutants such as pesticide residues and illegal additives is realized; the method has the remarkable advantages of simplicity and convenience in operation, low cost and high efficiency, and is suitable for rapid screening and quantitative analysis of various foods.