1925 results about "Spectral data" patented technology

The invention relates to an infrared spectrum baseline drift dynamic compensation method and system based on multi-parameter fusion, and the method comprises the steps: continuously obtaining spectrum data, current environment parameters and light source state data of a target oil full wave band, carrying out the feature extraction, and obtaining a baseline drift sensitive feature set and a coupling feature set; constructing a prediction model taking the baseline drift amount as output, and taking the baseline drift sensitive feature set and the coupling feature set as input; the reverse baseline drift amount based on the baseline offset of each wave band is superposed in the spectral data as compensation; according to the invention, through multi-source data acquisition and analysis, monitoring of spectrum baseline drift influence factors is realized; accurate correction of different spectral characteristics is realized through a dynamic compensation strategy special for a wave band; and through a continuous self-adaptive updating mechanism, stable tracking of the long-term drift trend is realized, and the effect of prolonging the equipment maintenance period is achieved.

Techniques for determining a presence of an object, especially an object such as animal or debris, in a path of a vehicle, are discussed herein. For example, sensors of various modalities, which may include multispectral sensors, may capture data representing an environment the vehicle is traversing. In examples, one or more trained machine learned (ML) models, operating on a vehicle computing system, may detect and / or classify objects in the environment, based on input data of one or more modalities or spectral bands. The ML models may be pre-trained using training data including real sensor data, synthetic data, and / or augmented data, along with auto-generated annotations. In some examples, hyperspectral data may be used to identify materials associated with detected objects. A confidence score associated with the detection of the object may also be computed. The vehicle may be controlled based on detection of the object and its classification.

The invention relates to the technical field of data processing, and discloses an intelligent monitoring method and system for cyanobacterial bloom outbreak. The method comprises the following steps: collecting a water surface spectrum and underwater particle size data, carrying out atmospheric correction, calculating a normalized algae index and a blue-green wave band ratio, inputting the normalized algae index and the blue-green wave band ratio into a U-Net network to obtain a water bloom coverage area, carrying out integral interpolation on the particle size data to obtain a vertical section distribution curve, and calculating a surface layer enrichment degree and a floating trend index, and establishing a water surface-underwater association relationship through random forest regression training, and inputting the multi-dimensional features into a CNN-LSTM model to predict a water bloom outbreak probability and determine an early warning level. According to the method, the problem that the cyanobacterial bloom three-dimensional structure cannot be comprehensively described due to the lack of effective fusion of the water surface spectral data and the underwater vertical section data is solved, the problem that the early warning timeliness of cyanobacterial bloom outbreak is insufficient due to the lack of a multi-source data time sequence analysis model is solved, and the spatial integrity and early warning advance of cyanobacterial bloom monitoring are improved.

The invention relates to the technical field of boiler combustion control, in particular to a boiler combustion multi-target collaborative optimization control method and system. The method comprises the following steps: firstly, acquiring coal quality parameters in real time through a laser-induced breakdown spectrum analyzer, carrying out qualitative and quantitative analysis on spectral data by utilizing a machine learning algorithm, constructing a boiler combustion digital twinborn model fusing a combustion mechanism model and a data driving model, and predicting combustion characteristics, NOx generation trends and coking risks under different blending schemes; and then, dynamically generating an optimization instruction combination according to a prediction result of the digital twinborn model, establishing a safety interlocking mechanism while executing the optimization instruction, and automatically switching to a manual control mode when detecting that the working condition fluctuates severely or the model prediction deviation exceeds a preset threshold value. According to the method, the problems of coal quality information lag and untimely combustion adjustment caused by traditional manual sampling are effectively solved, and the economical efficiency, the safety and the environmental protection property of boiler operation are improved.

According to the spectrum-to-mass concentration imaging method and device based on physical mechanism deep learning provided by the invention, the actually measured spectrum and the reference spectrum of the pollution gas smoke plume are collected, the spectrum data set is constructed after differential processing, the meteorological data and the online mass concentration label are synchronously collected, and meanwhile, the spectrum-to-mass concentration imaging method and device based on physical mechanism deep learning are provided. A high-resolution gas absorption section is obtained and is convolved into a matrix; and constructing a deep learning model fusing a feature extraction module, an expanded least square module and a full connection module, taking the spectral data set, the meteorological data and the absorption cross section matrix as input, performing training in combination with labels to obtain an optimization model, and predicting the mass concentration of the target gas. According to the method, the problems of error accumulation, low calculation efficiency and poor interpretability caused by dependence on a complex physical model in a traditional method are solved, and high-precision, high-efficiency and interpretable real-time imaging of the mass concentration of the smoke plume of the pollution gas is realized.

The invention discloses a method and a system for reflecting a growth state by a spectral curve of a plant leaf under salt mist stress and a storage medium, and relates to the field of plant growth monitoring. Processing the collected leaf image, extracting a leaf main body area, removing an interference part, and matching the filtered original spectral data to obtain target spectral data of the leaf main body area; the method comprises the following steps: combining target spectral data of a leaf main body area with leaf moisture content data, constructing a spectral analysis model by adopting a partial least square method, taking the leaf moisture content as a covariable into the spectral analysis model, peeling off the interference of leaf moisture change on the spectral data, and calculating the leaf moisture content. And finally obtaining a target spectrum curve of the plant leaf under the salt mist stress. A characteristic wave band corresponding to the target spectrum curve is extracted, a plant growth index is constructed according to the characteristic wave band, and the plant growth index is used for reflecting the growth health state of the plant. The problems that traditional monitoring is high in subjectivity, insufficient in precision and low in efficiency are solved.

The present invention belongs to the technical field of coal gangue recognition and sorting, and in particular, relates to a coal gangue recognition method based on visible-near-infrared spectrum and image multi-modal information fusion. The method includes: S1 collecting spectral information and image information about a sample to be recognized; S2 preprocessing the spectral information and the image information respectively; S3 extracting spectral features from a spectral data set by using a spectral feature extraction neural network model; and extracting image features from an image data set by using an image feature extraction neural network model; S4 inputting the spectral features and the image features obtained from feature extraction into a two-stream fusion network; S5 inputting extracted spectral features, extracted image features and a comprehensive feature into a spectral branch classifier, an image branch classifier and a fusion branch classifier, respectively; S6 calculating importance weights of a spectral branch, an image branch and an image-spectrum fusion branch; and S7 subjecting the importance weights and corresponding confidence to multiply-accumulate operation to obtain a score matrix of coal gangue, and using the score matrix to achieve coal gangue recognition.

An LIBS-based bauxite component quantitative detection method and a related apparatus. The method comprises: controlling an LIBS system to collect actual spectral data of bauxite under test; preprocessing the actual spectral data to obtain preprocessed actual spectral data; on the basis of the preprocessed actual spectral data, extracting corresponding feature elements to be detected, said feature elements comprising one or more of information about feature wavelengths to be detected, information about element wavelengths to be detected, and information about remaining wavelengths to be detected; and, on the basis of said feature elements and a quantitative detection model, determining component information of said bauxite.

The invention discloses a rapid nondestructive testing method and system for the lipid content and deterioration degree of red pine nuts based on hyperspectral imaging and deep learning, and belongs to the technical field of nondestructive testing of the quality and safety of agricultural and forestry products. The method is provided for solving the problems that an existing method for detecting the lipid content and the oxidation degree in the red pine nut kernels is generally complex in operation process, high in large-batch detection cost, long in consumed time and difficult to achieve detection in the whole storage and transportation process. The method is characterized by comprising the following steps: acquiring original near infrared spectrum data of a pine nut sample through a collected hyperspectrum; determining the lipid reference content truth value and the oxidation deterioration reference degree of the pine nut samples at different sampling times, and establishing a database according to the values; the method comprises the following steps: preprocessing collected original near infrared spectrum data of red pine nuts, and dividing red pine nut sample data collected in different batches into a training set and a verification set; and designing an improved one-dimensional cavity convolutional network based on a dynamic weight distribution module to construct a deep learning model, wherein the deep learning model is used for constructing a deep learning structure suitable for spectral feature analysis of the red pine nuts. And a back propagation algorithm is adopted to train the constructed model, and reverse updating of network weight parameters is realized by minimizing a loss function. When the performance of the constructed model meets the rapid detection requirement, the method is used for efficient and lossless synchronous detection of the lipid content and the oxidative rancidity degree of the red pine nuts.

The invention relates to the technical field of Brassica rapa polysaccharide detection, and discloses a quantitative detection method of Brassica rapa polysaccharide based on near infrared spectroscopy. The method comprises the following steps: acquiring near infrared spectrum data of a brassica rapa sample, and dividing detection stages by combining spectral characteristics to obtain a plurality of detection stage identifiers; performing differential mapping on the spectrum load spectrum according to the identifier to obtain an optical parameter set of each stage; constructing a staged near infrared spectrum quantitative model by using the set; performing importance weighting on the original spectral data according to the identifier to generate optimized spectral data; and extracting polysaccharide related characteristics from the optimized spectral data through a staged model, and outputting a predicted value of the content of the Brassica rapa polysaccharide. According to the method, through stage division, differential mapping and data weighting, the spectral data utilization rate and the model suitability are improved, a targeted technical path is provided for quantitative detection of Brassica rapa polysaccharide, and the method can be used for quality control of Brassica rapa related products.

A regional soil moisture monitoring method based on airborne spectral reconstruction optical satellite remote sensing, the method comprising: acquiring satellite remote sensing data of a region to be monitored, and collecting unmanned aerial vehicle remote sensing data of a local region in the region to be monitored; on the basis of the unmanned aerial vehicle remote sensing data of the local region, performing spectral reconstruction on the satellite remote sensing data of the region to be monitored, so as to obtain reconstructed satellite spectral data of the satellite remote sensing data; and inputting the reconstructed satellite spectral data into a pre-trained soil moisture inversion model, so as to obtain the soil moisture content of the region to be monitored. Therefore, soil moisture is accurately and quickly monitored.

The invention relates to the field of gas concentration detection, and particularly discloses a real-time feedback and adaptive learning method for natural gas infrared spectrum measurement, which comprises the following steps: S1, acquiring infrared spectrum information of a natural gas body by using an infrared spectrometer, and constructing a historical sample set; s2, preprocessing the spectral data of the historical sample set; s3, selecting an optimal algorithm and a hyper-parameter by adopting XGBoost and Bayesian optimization; s4, constructing a qualitative model to identify gas types and match data; s5, calculating the similarity between a field sample and a historical sample through a Siamese network, and setting a threshold value to screen local data; s6, improving the KNN to construct a local dynamic quantitative model to predict the concentration; s7, processing low-similarity abnormal data by the global dynamic model, and improving the reliability by combining moving average and abnormal calibration; and S8, introducing reinforcement learning and online gradient descent to adjust parameters in real time to optimize the precision. According to the technical scheme, high-accuracy natural gas detection can be carried out in a complex environment.

The invention provides a spectral data optimization method and device, computer equipment and a storage medium, and relates to the technical field of semiconductor detection.The method comprises the steps that original spectral data output by a spectrograph is obtained, and the central wavelength value of the original spectral data of each frame is calculated; querying from a preset database according to the central wavelength value to obtain a corresponding instrument linear function; and processing each frame of original spectral data and the corresponding instrument linear function by adopting a deconvolution algorithm to obtain optimized real spectral data. According to the scheme, algorithm compensation is carried out on each frame of spectral data, so that system distortion introduced by the spectrometer can be accurately compensated in a full spectrum range, the extraction precision of spectral features is greatly improved, and the magnitude order improvement of measurement precision is realized; in addition, the optimized spectral data more truly reflects the physical characteristics of the tested sample, the dependence of an optical model on empirical parameters is reduced, and the generalization ability and prediction reliability of the model are improved.

The invention relates to the field of environmental monitoring, and particularly discloses a non-negative matrix factorization and adaptive peak recognition fluorescence feature extraction and traceability system, which comprises a spectral data preprocessing module, a spectral data non-negative matrix factorization module, a component number automatic selection module, an adaptive peak recognition module, a feature library construction module and a similarity comparison module. An improved non-negative matrix factorization model is adopted to decompose the three-dimensional fluorescence spectrum matrix of a single sample, and an optimal component number K is automatically determined through multiplicative update rule iterative optimization; the self-adaptive peak identification module carries out selective filtering, accurately extracts the position and intensity of a fluorescence peak through multiple mechanisms, and carries out peak position calibration in a neighborhood; the Hungary algorithm is adopted to carry out characteristic peak matching to calculate the comprehensive similarity between the samples, and rapid and accurate identification of the pollution source is realized. The method has the advantages of high resolution, strong anti-interference capability, low requirement on the number of samples, automation and the like, and is suitable for water quality fingerprint feature extraction of a water sample in a complex environment and real-time source tracing of sewage.

The invention provides a semiconductor epitaxial growth control method and system, a control device and a related product, and relates to the technical field of semiconductor detection.The semiconductor epitaxial growth control method comprises the steps that reflection spectrum data is obtained in real time and analyzed to obtain an optical feature set, then the optical feature set is input into an optical transmission model for iterative optimization, and a predicted growth rate is obtained; and comparing the predicted growth rate with the target growth rate, generating deviation information, and outputting a process adjustment control instruction, so that the main control system performs closed-loop correction on the process parameters in the next process period. According to the invention, the optical feature set can dynamically represent the influence of the interface transition layer on the reflection spectrum, so that the inversion precision and robustness of the real epitaxial structure state are improved; closed-loop correction is performed on the process parameters based on the predicted growth rate, so that the growth rate drift can be inhibited in time, the process disturbance can be adaptively compensated, the manual calibration and intervention requirements are reduced, and the production stability, repeatability and batch-to-batch consistency are improved.

The invention relates to the technical field of medical hygiene, and particularly discloses a hygiene visual intelligent monitoring and guiding system suitable for surgical hand disinfection, hand motion data is collected by fusing a multi-modal motion capture module with a depth camera and an inertial sensor, and a high-risk area identification module analyzes the risk of microbial residues in finger seams and fingernail edges. The three-dimensional motion modeling module constructs a biomechanical model to calculate a trajectory, the dynamic compliance guiding module marks deviation in real time through optical projection, the multi-source data fusion module integrates a risk value and a spectral data evaluation effect, and the intelligent decision engine module outputs an operation matching degree and a correction instruction based on a neural network. Omnibearing monitoring, accurate risk assessment, real-time action guidance and quantitative effect evaluation of the hand disinfection process of an operator are realized, the standardization and effectiveness of surgical hand disinfection operation are finally improved, and the infection risk of a surgical site is reduced.

The invention provides a rapid evaluation method and system for the soil fertility level, and relates to the technical field of soil fertility detection.The rapid evaluation method comprises the steps that an original absorbance spectral data curve is obtained by scanning a soil sample, the moisture peak absorbance, the actual peak position wavelength and the moisture synthetic frequency characteristic peak width are extracted, and a spectral data basis is provided; a water-salt optical path coupling factor is calculated, water-salt correction operation is carried out in combination with a volume correction coefficient, a virtual dry soil spectrum data curve is obtained, interference of salt on a water optical path is quantified, and a water background and salt masking are eliminated; constructing a characteristic virtual baseline of the organic matter and the total nitrogen, calculating an integral area to obtain characteristic intensity of the organic matter and the total nitrogen, and providing a physical input variable for quantitative inversion; introducing a water-salt optical path coupling factor to invert the organic matter content and the total nitrogen content, and correcting signal depression caused by salt stress; the effective fertility index is calculated and graded by combining the organic matter content and the total nitrogen content of the soil, and the actual fertilizer supply capacity of the soil is comprehensively evaluated.

The invention relates to a geological disaster early warning algorithm based on hyperspectral and Internet of Things data fusion analysis, and relates to the technical field of geological disaster monitoring and early warning, the geological disaster early warning algorithm comprises the following steps: S1, preprocessing hyperspectral data and Internet of Things data, and respectively extracting spectral-spatial features and dynamic topology time sequence features; s2, fusing the spectrum-space features and the time sequence features to generate cross-modal joint features; s3, performing parameter optimization on the fusion features through a quantum-classical hybrid optimization algorithm, and calculating a disaster risk probability; and S4, based on the optimization result and the risk probability, executing an edge-cloud collaborative early warning decision. According to the method, through non-negative tensor ring decomposition of the hyperspectral data and dynamic topology modeling of the Internet of Things, the limitation of a traditional single data source in temporal-spatial resolution and physical relevance is solved, multi-dimensional joint extraction of spectrum-space-mechanical characteristics can be realized, and the characterization precision of a rock-soil body deformation evolution law is remarkably enhanced.

The invention discloses a correction method for saturation spectral line intensity in a laser spectrum, and relates to the technical field of spectral data analysis, the method comprises the following steps: identifying saturation characteristics in the spectrum, determining a saturation point based on an intensity threshold, and recording the original intensity of the saturation point; determining a target spectrum peak wavelength range containing a saturation point; eliminating saturation points from the wavelength range of the target spectrum peak, and retaining unsaturation points; fitting the unsaturated point through a nonlinear function to obtain the maximum fitting strength; if the maximum fitting strength is lower than the original strength, performing strength correction by adopting a linear extension mode by taking an unsaturated point as a reference, and otherwise, performing strength correction directly based on a fitting model; and outputting the corrected spectral intensity data, and calculating the maximum intensity of the spectral peak and the integral area of the spectral peak. According to the method, the accuracy of saturation judgment and the reliability of a correction result are improved, and a direct intensity data basis and core derivative parameters are provided for quantitative analysis.

The invention discloses a mixed gas absorption spectrum analysis method and system based on variational mode decomposition, specific laser is injected into an optical resonant cavity unit, and a detector unit continuously monitors the light intensity change and records a light intensity attenuation signal; pre-processing the recorded light intensity attenuation signal; carrying out VMD decomposition on the preprocessed ring-down signal to obtain a plurality of IMF components, respectively introducing CO2 and CO gases with known concentrations into the optical resonance unit, and recording spectral data of each single gas component; calculating the similarity and contribution degree of each IMF component, and setting a weight combination to form a joint score; the gas with the highest joint score is selected, the score is compared with an adaptive threshold value, and when the score is higher than the adaptive threshold value, the IMF component is marked as the characteristic component of the corresponding gas; and finally, gathering and outputting the marked IMF components according to gas types to obtain characteristic signals of the gases, thereby realizing multi-component gas separation in the mixed gas absorption spectrum.

The invention relates to the technical field of shoe body detection, in particular to a shoe body quality detection method based on image visual analysis, which comprises the following steps: synchronously acquiring three-dimensional contour and surface spectral information of a to-be-detected shoe body through three-dimensional scanning and spectral imaging equipment to obtain original point cloud spectral data; performing registration and fusion processing on the original point cloud spectral data to construct a registered hyperspectral three-dimensional grid model containing space coordinates and hyperspectral reflectivity values; and calculating a normal vector field and a curvature distribution diagram of the surface based on the hyperspectral three-dimensional grid model so as to quantify geometric features of the surface of the shoe body. According to the invention, through registration and fusion processing, surface spectral information collected by a spectral imaging device is reversely projected and interpolated to each space coordinate vertex of an initial three-dimensional point cloud constructed by a three-dimensional scanning device, so that a registered hyperspectral three-dimensional grid model is constructed, and a surface chemical characteristic spectrum is endowed with three-dimensional space geometric coordinates. And the detection capability is improved.

The invention provides a multi-platform satellite thermal infrared hyperspectral atmospheric ammonia monitoring method and system, and belongs to the technical field of satellite remote sensing. The method comprises the following steps: collecting thermal infrared hyperspectral data, atmospheric state parameters, earth surface parameters and instrument characteristic parameters of a stationary satellite or a polar orbit satellite; based on a thermal infrared radiation transmission physical mechanism, inputting the preprocessed standardized parameters into a fast radiation transmission forward model to obtain a simulated spectrum consistent with the actually measured data format of a stationary satellite or a polar orbit satellite; constructing a cost function containing observation error constraint and NH3 prior information constraint on the basis of the simulated spectrum and the actually measured spectrum in combination with an optimization estimation theory, solving a minimum value of the cost function by adopting a Levenberg-Marquardt iterative algorithm, and performing inversion to obtain an atmospheric ammonia concentration profile; and performing quality control and column concentration conversion on an inversion result, and verifying the precision by combining multi-source observation data to obtain an atmospheric ammonia concentration data set.

The invention discloses a temperature and gas detection system and method based on an ultraviolet to near infrared spectrum, relates to the technical field of spectral measurement, and solves the technical problems that the space in a smelting furnace is occupied and the complexity of the system is increased due to the fact that two sets of independent equipment are adopted for temperature measurement and environment monitoring in the smelting process respectively in the prior art. The method comprises the following steps: acquiring spontaneous radiation light of a to-be-detected target; carrying out dispersive light splitting on the spontaneous radiation light to obtain spectral data of a full wave band from ultraviolet light to near-infrared light; performing denoising processing on the spectral data to obtain a spectrum of the to-be-detected target; obtaining temperature data according to the spectral analysis of the to-be-detected target; calculating based on the spectrum and a pre-established spectrum database to obtain gas concentration; the temperature of the to-be-measured target can be accurately measured, and the measurement accuracy is improved.

The invention discloses a molten steel element content on-line detection system based on a laser-induced breakdown spectroscopy technology, and relates to the technical field of molten steel element content on-line detection, the molten steel element content on-line detection system comprises an excitation system, an acquisition system, a data preprocessing module and a component detection module; the excitation system is used for emitting laser to the molten steel surface to generate plasma; the acquisition system is used for collecting signal light radiated by the plasma; the data preprocessing module is used for carrying out baseline correction and noise reduction processing on the collected original spectral signals; and the component detection module comprises an element content prediction model based on a gating specific expert attention network LSEA-Net and is used for carrying out quantitative analysis on the preprocessed spectral data to obtain the element content of the molten steel. The technology has the advantages that sample preparation is not needed, online rapid analysis and multi-element synchronous detection are achieved, and the technology is suitable for the high-temperature environment and the like, and the requirement of a steelmaking site for real-time component detection is met very well.

The invention provides a polarization modulation coupling time control phase locking high-temperature noise suppression in-situ Raman method and device, and relates to the technical field of spectrum detection. The method comprises the following steps: exciting pulse linear polarization laser to irradiate a sample to obtain Raman scattering light; a parallel vibration component (P) and a vertical vibration component (S) are obtained, and original spectral data containing polarization information and residual high-temperature noise are collected by adopting an ISCCD detector synchronously triggered by gating. Background noise is removed and a parallel vibration component (P) is extracted through polarization differential processing in combination with adaptive filtering (dynamically adjusting parameters according to signal distribution) and principal component analysis (PCA); and finally, performing signal-to-noise ratio optimization on the pure component by applying a phase locking algorithm to finally obtain a Raman spectrum result with a high signal-to-noise ratio. According to the invention, the in-situ Raman spectrum of the material within the temperature range of room temperature to 3000 DEG C can be measured, and the detection of weak signals is more sensitive.

The invention belongs to the technical field of water pollutant identification, and particularly relates to a water pollutant distinguishing method and system based on multi-source spectrum fusion. According to the method, ultraviolet visible spectrum data and three-dimensional fluorescence spectrum data are effectively integrated in a multi-source spectrum data fusion mode, so that the characteristic information of water pollutants is captured, stably appearing characteristic absorption wavebands can be positioned by utilizing the ultraviolet visible spectrum data, and the water pollutants can be accurately detected. The method comprises the following steps: identifying a characteristic sub-region capable of representing the fluorescence response characteristic of a pollutant through three-dimensional fluorescence spectrum data, in a characteristic extraction stage, performing first-order derivative operation on an ultraviolet visible spectrum to position an absorption peak position, calculating curvature distribution of a three-dimensional fluorescence spectrum to identify the characteristic sub-region, and when a fusion characteristic is generated, performing first-order derivative operation on the three-dimensional fluorescence spectrum to identify the characteristic sub-region; according to the method, the key operation parameters are subjected to grid search optimization, and the optimal key operation parameter combination is screened out by performing multiple rounds of iterative training and evaluation on a preset training set and a verification set, so that the recognition performance of the real-time collected water sample is improved.

The application discloses a kind of full automatic waste fabric identification sorting control system, it is related to resource reusing technical field, including identification analysis module, the type, color, component and fibre fineness data of waste fabric on the surface of automatic conveying module are acquired and analyzed by identification analysis module, obtain the component composition and its weight distribution of each area of waste fabric, generate data form and send to data module, and sorting action module is controlled by sorting strategy control module to carry out sorting screening;Using machine vision technology and fabric visual identification model determines the type of waste fabric and divides its area, so that different parts of fabric can be detected and identified, improve the component identification precision of fabric;By extracting the color value of different areas and correcting the color value of spectral data in the identification analysis module, the influence of the color on the surface of the fabric on light, especially visible light, is eliminated, making the spectral data closer to the truth, and the analysis result is more accurate.

The invention relates to the technical field of spectrum detection, in particular to a Raman spectrum-based rice oil detection method, which comprises the following steps: extracting main and auxiliary peaks to construct a peak group structure, performing sliding comparison to output a conformity region, identifying jump to remove an abnormal signal, calibrating a peak position to generate a correction structure, and performing homing numbering to complete a detection scheme. According to the method, the feature matching rule is constructed by combining the peak position spacing, the intensity ratio and the wavenumber index, the recognition precision of the multi-peak structure in the spectrum is enhanced, non-continuous response signals are recognized and eliminated through derivative change trend and proportion deviation constraint, and the accuracy of stable response extraction is improved. Dynamic calibration and position adjustment of peak positions are realized by utilizing an offset trend sequence, the consistency of a peak group structure in an overall direction and the continuity of a local structure are ensured, a numbering system and structure attribution division is completed after homing sorting, and the integrity of feature expression in a complex sample and the orderliness and reliability of spectral data processing are improved.

The invention discloses a troposphere BrO vertical profile inversion method, medium and equipment based on machine learning, and the method comprises the steps: obtaining BrO vertical profile data, inputting the profile data, aerosol characteristic parameters, geometric angle parameters and earth surface characteristic parameters into an atmospheric radiation transmission model, and obtaining BrO DSCD data; the obtained DSCD and the profile data set are preprocessed; training a troposphere BrO vertical distribution inversion model combining a convolutional neural network CNN and a long short-term memory network LSTM by using the preprocessed DSCD and profile data set; and performing QDOAS processing on MAX-DOAS spectral data to be inverted to obtain DSCD data, inputting the preprocessed DSCD data, aerosol parameters and geometric angle parameters into the trained CNN-LSTM troposphere BrO vertical distribution inversion model to perform data inversion, and finally obtaining an inversion result of troposphere BrO vertical distribution. According to the invention, the accuracy and stability of an inversion result can be significantly improved.

The invention discloses a method for detecting the nitrogen content of organ tissues of different varieties of oilseed rapes based on visible near-infrared spectroscopy, which comprises the following steps: firstly, collecting oilseed rape leaf, shell, stalk and root system samples under the conditions of multiple varieties and multiple nitrogen fertilizer levels, and acquiring spectral data within the range of 430-2500nm by using a visible near-infrared spectroscopy; and a Kjeldahl method is synchronously adopted to measure the real nitrogen content as a reference value. Preprocessing the spectral data, including de-noising, standard normal variable transformation, multivariate scatter correction and derivative processing, so as to weaken the influence of scattering and baseline drift; characteristic wavelengths related to the nitrogen content are screened through stepwise regression, variable projection importance, competitive self-adaptive reweighted sampling, a continuous projection algorithm and other methods, and a random forest model, a support vector machine model, a partial least squares discriminant analysis model, a partial least squares regression model, a support vector regression model, an XGBoost model and other models are combined. And respectively constructing a classification identification model and a regression prediction model.