34 results about "Spectral database" patented technology

The invention relates to the technical field of gas detection, and provides a doped gas concentration detection method and system based on a gas analyzer. The method comprises the following steps: constructing a standard spectrum database under the same pressure and temperature as a to-be-measured site; the method comprises the following steps: introducing sample gas into an optical absorption cell, carrying out wavelength scanning through a tunable diode laser in a plurality of adjacent absorption spectral line intervals containing target gas and interference gas, superposing a high-frequency sinusoidal modulation signal on a driving current, and synchronously detecting a second harmonic component of transmission laser to generate an actual measurement spectrum; fitting the actually measured spectrum with a synthetic spectrum generated on the basis of a standard spectrum database, and outputting a target gas concentration ratio as a detection result when a residual error meets a threshold value, so as to solve the technical problem that the detection precision is reduced due to spectrum overlapping caused by interference gas, and achieve the purpose that the detection precision is reduced by constructing the standard spectrum database and second harmonic spectrum fitting analysis. And the accuracy and the anti-interference capability of gas detection are improved.

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 provides a hyperspectral water quality detection device and a hyperspectral water quality detection method, and aims to realize water quality detection with small volume, high detection speed, high real-time performance and high precision. The device comprises a water tank, a black sealing box embedded into a transparent sample box, a hyperspectral camera, a full-spectrum light source and a processing module. A water sample is irradiated by a light source through a transparent sample box, a hyperspectral camera collects spectral data of the water sample, an embedded processing module preprocesses the data, and feature extraction and multi-parameter prediction are performed by using a water quality analysis algorithm based on a one-dimensional convolutional neural network and a multi-head attention structure. By constructing a water quality spectrum database to train the model, real-time parallel analysis and accurate prediction of spectrum data in a specific area can be realized, and predicted water quality parameters include but are not limited to indexes such as COD, turbidity, total nitrogen, total phosphorus, ammonia nitrogen and chlorophyll. The device can realize high-precision and multi-parameter water quality real-time detection, and is small in size, fast in response and wide in application range.

The application provides a kind of management and control chemical fast LIBS classification identification model construction method, medium and system, belongs to management and control chemical technical field, the application is by collecting mercury argon reference lamp spectrum to carry out wavelength calibration, according to atomic spectrum database extraction management and control element characteristic transition wavelength window and compressed to low-dimensional feature space by sparse self-encoder, with hydrogen line starker broadening and Boltzmann graph slope method real-time inversion electron density and electron temperature and to compressed characteristic matrix is physically normalized, calculates plasma drift index to adaptively adjust plasma parameter decoupling double-flow condition graph network update strategy, the normalized characteristic matrix and physical diagnostic value are input into the artificial intelligence network to output element concentration vector and class prediction probability, solve the technical problems that plasma parameter fluctuation leads to laser-induced breakdown spectroscopy classification feature instability, classification accuracy significantly decreases under the condition of cross-instrument and cross-matrix.

The invention relates to the cross technical field of biological medicine and data processing, and discloses a quality detection data analysis method for a close-period liquid medicine preparation. The method comprises the following steps: acquiring a transmitted light intensity signal through a multi-wavelength optical sensing array; performing compression on the edge side to generate a 6-dimensional low-dimension spectral feature vector; and after the data is uploaded to a cloud end, performing dynamic track matching based on a hierarchical hidden Markov model constructed by a historical full-life-cycle spectrum database, calculating a degradation risk index and triggering early warning. The system comprises a sensing array, an edge computing unit and a cloud collaborative analysis platform. According to the invention, through an edge-cloud collaborative architecture, the communication and calculation overhead is significantly reduced while the early warning precision is ensured, degradation early warning in advance of not less than 48 hours is realized, and the missing of an intervention window is effectively avoided.

A method and system for investigating and analyzing data from a mass spectrometer, the system including a spectral database, a count service for receiving scan data generated by the mass spectrometer and identifying a number of spectra in the data, a scaling service for receiving the scan data generated by the mass spectrometer, receiving the number of spectra from the count service, and initiating a plurality of query services, each query service of the plurality of query services corresponding to at least one spectra of the number of spectra and for querying the spectral database with the corresponding at least one spectra and returning a match between the corresponding at least one spectra and at least one known spectra from the spectral database, and a results service for retrieving each match, and formatting each match into an output data structure.

The invention relates to a method for predicting the content of elements in coal based on LIBS spectral feature optimization and machine learning, which comprises the following steps of: converting a preprocessed spectrum into a two-dimensional spectral matrix, matching a peak wavelength point of the two-dimensional spectral matrix with an NIST standard atomic spectrum database, identifying spectral lines belonging to target elements, and determining the content of the elements in the target elements according to the spectral lines. A characteristic wavelength set stably appearing in all coal sample spectrums is screened out to serve as candidate characteristics; further screening out a core characteristic wavelength subset which is most relevant to the content of the target element and has the minimum redundancy by applying a plurality of characteristic selection methods; and then selecting an optimal model combination by adopting a plurality of machine learning regression algorithms. According to the method, through a two-stage screening strategy combining physical spectral line identification (NIST matching) and statistical learning feature selection, irrelevant and redundant spectral information is eliminated to the greatest extent, anti-interference core features directly related to the target element content are reserved, and the quality of model input data is improved from the source.

The application discloses a temperature and gas detection system and method based on ultraviolet to near infrared spectrum, relates to the technical field of spectrum measurement, and solves the technical problem that the prior art adopts two separate devices to respectively measure temperature and monitor the environment in a smelting process, occupies space in a smelting furnace, and increases the complexity of the system; the application obtains spontaneous emission light of a target to be detected; the spontaneous emission light is dispersed and spectrally analyzed to obtain spectrum data of the full wave band of ultraviolet to near infrared light; the spectrum data are denoised to obtain the spectrum of the target to be detected; temperature data are obtained according to spectrum analysis of the target to be detected; and the concentration of gas is calculated based on the spectrum and a pre-established spectrum database; the temperature of the target to be detected can be accurately measured, and the measurement accuracy is improved.

The present application relates to camouflage recognition imaging system and camouflage recognition method, specifically relates to a kind of multispectral camouflage recognition imaging system and camouflage recognition method, solve the technical problems that existing camouflage recognition imaging system range of matter is small, application scene is limited, or camouflage recognition method cannot real-time imaging, real-time identification, or spectral database construction efficiency is low.The multispectral camouflage recognition system of the present application includes image acquisition unit, data processing unit and display storage unit, image acquisition unit uses snapshot type spectral imaging chip, can real-time exposure, real-time data processing;Data processing unit can obtain continuous spectral information by means of spectral fitting degree identification technology, according to the characteristic spectrum of different kinds of identification target selects feature database, changes identification target, greatly increases the identifiable spectral range.In the multispectral camouflage recognition method of the present application, spectral correction is included, can remove the interference of environmental light factor, work under different daylight color temperature.

The application relates to the technical field of biological medicine and data processing, and discloses a quality detection data analysis method for near-efficiency liquid medicine preparation. The method comprises the following steps: collecting a transmission light intensity signal through a multi-wavelength optical sensing array; generating a 6-dimensional low-dimensional spectral feature vector through edge side compression; after being uploaded to the cloud, performing dynamic trajectory matching based on a hierarchical hidden Markov model constructed based on a historical full-life-cycle spectral database, calculating a deterioration risk index, and triggering an early warning. The system comprises a sensing array, an edge computing unit and a cloud collaborative analysis platform. Through the edge-cloud collaborative architecture, the application can guarantee early warning accuracy while significantly reducing communication and calculation overheads, realize deterioration early warning for not less than 48 hours, and effectively avoid missing the intervention window.

The invention relates to a method for synchronously and rapidly detecting multiple parameters of respirable dust and silicon dioxide on a membrane, and belongs to the field of dust monitoring. The method comprises the following steps: S1, carrying out on-membrane sampling and Fourier infrared spectrum scanning on the quartz respirable dust, taking the quartz respirable dust as a spectrum library of a standard sample, and establishing a standard silicon dioxide and quartz respirable dust quality prediction model based on a spectrum absorption peak of silicon dioxide at 800cm <-1 >; s2, selecting different types of dust, carrying out on-film sampling and spectrum comparative analysis, and establishing spectrum databases and dynamic correction models of different types of dust; and S3, aiming at concentration calculation deviation caused by sampling flow fluctuation, establishing a flow compensation and concentration calculation model, and finally realizing synchronous detection of multiple parameters such as respirable dust mass concentration, silicon dioxide mass concentration and silicon dioxide mass percent.

The method comprises the following steps: establishing a near infrared spectrum database of key indexes of the salt-making mother liquor, and carrying out chemical component calibration by combining a laboratory standard method; introducing the mother liquor into an optical detection pool through a bypass circulation system, and collecting near infrared spectrum data; constructing a semi-supervised regression model between the spectral data and the target parameters, and calculating the sulfate radical content, the chloride ion content and the solid-to-liquid ratio in the mother liquor by utilizing the established semi-supervised regression model; a detection result is transmitted to a control system, and crystallization process parameters are adjusted. According to the scheme provided by the invention, the continuous and real-time monitoring of the salt-making mother liquor is realized through the combination of the bypass circulation system and the near infrared spectrum technology; the semi-supervised regression model is utilized, the flow model depth reversible network and the maximum mutual information coefficient are combined, overlapping signals and complex background interference in spectral data are effectively decoupled, and the accuracy and stability of multi-component quantitative analysis are improved.

The invention belongs to the technical field of coal quality detection, and particularly discloses a coal mineral component proportion estimation method and system, and the method comprises the steps: collecting an original coal sample hyperspectral image, and carrying out the preprocessing of the original coal sample hyperspectral image; the method comprises the following steps: determining types and proportions of minerals in a coal sample through a coal quality test method, collecting spectral data of each mineral in the coal sample by adopting a hyperspectral imager, and further establishing a spectral database of the minerals in the coal; constructing a 3D convolutional neural network model based on an encoder-decoder structure on the basis of the preprocessed coal sample hyperspectral image and the spectral database of the minerals in the coal; and inputting a to-be-detected original coal sample hyperspectral image into the 3D convolutional neural network model based on the encoder-decoder structure to carry out space-spectral feature fusion, introducing physical prior constraints, and estimating to obtain mineral proportion data in the coal sample. The problems that an existing coal quality detection method is low in precision and poor in applicability, and online real-time detection cannot be achieved are solved.

The utility model relates to the technical field of gas spectrum analysis, in particular to an automatic calibration and storage device of a gas spectrum database, which comprises a gas chamber, a sealing cover is arranged on the upper side of the gas chamber, and an adjustable light source module, a gas spectrum calibration module and an integral processor are sequentially arranged on the upper side of the sealing cover from left to right. A heat dissipation type storage module is arranged at the right end of the lower side of the gas chamber, two gas path assemblies are arranged in the gas chamber and can be divided into a calibration gas path assembly and a cleaning gas path assembly, the calibration gas path assembly is used for normal gas calibration and conveying, and the cleaning gas path assembly is used for cleaning the calibration gas path assembly. According to the utility model, a highly integrated modular structural design is adopted, and the adjustable light source module, the gas spectrum calibration module, the overall processor and the heat dissipation type storage module are organically integrated around the gas chamber, so that the integrated functions of automatic calibration, analysis and storage of gas spectrum data are realized; and independent maintenance and upgrading of each functional module are facilitated.

The invention belongs to the technical field of biological detection, and particularly discloses a rapid microbiological detection system and method. The method comprises the following steps: acquiring a morphology image of a microbial sample by adopting the laser scanning subsystem; identifying a microbial target from the morphology image of the microbial sample by adopting an image segmentation method, and extracting position coordinates of the microbial target; based on the position coordinates of the identified microbial target, calculating the shortest path starting from the initial position to access all microbial target points as a scanning track; controlling a laser scanning subsystem and a spectrum detection subsystem to perform spectrum detection on each target position according to the scanning track; and carrying out feature extraction on spectral data including but not limited to Raman spectrum, comparing the spectral data with a standard spectral database, identifying the category of a microbial target, and judging the activity and other characteristics of microorganisms. According to the invention, the species, activity and other characteristics of microorganisms can be rapidly identified.

The application discloses a Raman spectrum discrimination method and system, and relates to the field of Raman spectrum analysis. The method comprises the following steps: pre-processing a to-be-detected sample signal to obtain a first Raman scattering signal of the to-be-detected sample signal; calculating a correlation coefficient according to the first Raman scattering signal and a preset spectrum database; and obtaining a matching spectrum type of the to-be-detected sample according to the correlation coefficient. According to the method, the correlation coefficient is calculated according to the first Raman scattering signal and the preset spectrum database, and the matching spectrum type of the to-be-detected sample is obtained according to the correlation coefficient. Compared with the traditional method, the calculation amount is greatly reduced, the algorithm running time is shortened, the algorithm time consumption is greatly reduced, and the operation is simple.

The invention relates to the technical field of spectrum optimization and intelligent illumination, in particular to a spectrum optimization system for fresh keeping of picked leaf vegetables, comprising a light source module which is composed of a plurality of independent and controllable LED units with different wavelengths and is used for generating a composite spectrum for irradiating the leaf vegetables; the spectrum sensing module is directly aligned with the light emitting surface of the light source module and is used for collecting the composite spectrum in real time and outputting multi-channel spectrum data; the control module is connected with the spectrum sensing module and the light source module; the optimal preservation spectrum database is integrated in a memory of the control module and stores target spectrum data corresponding to different leaf vegetable types; according to the invention, a closed-loop control system composed of a spectrum sensing module, a spectrum reconstruction unit, a spectrum matching unit and a spectrum optimization unit is constructed, so that the technical span from fixed spectrum illumination to dynamic precise spectrum regulation and control is realized, and the technical problem that the traditional fixed spectrum cannot adapt to the physiological characteristics of different leaf vegetables is effectively solved.

The application relates to the technical field of data processing, and provides a pixel spectral database optimization method and related devices.The method comprises the following steps: acquiring an initial pixel spectral database; performing screening on initial pixel spectral data in the initial pixel spectral database by adopting a linear separability quantitative analysis method to obtain a first screened pixel spectral database subset; performing screening on the first screened pixel spectral database subset by adopting a nonlinear separability method to obtain a second screened pixel spectral database subset, and obtaining an optimized pixel spectral database.The pixel spectral database can be screened twice from the linear angle of low latitude and the nonlinear angle of high latitude, redundant data and noise in the pixel spectral database are removed, the pixel spectral database is optimized, and the data quality of the pixel spectral database is improved.

The present description provides a data query method and apparatus based on an interstellar molecule spectrum database, an electronic device, and a storage medium. The method comprises: in response to retrieval information inputted by a user and associated with an interstellar compound molecule, determining a compound molecule to be queried; determining, from a pre‑created interstellar molecule spectrum database, basic query information corresponding to the compound molecule; performing reasoning on the basic query information and determining molecular spectrum theoretical data corresponding to a data type to be calculated which has been determined by the user; and returning the basic query information, the molecular spectrum theoretical data, and the reasoning process to the user.

The invention relates to an intelligent monitoring nursing bed cleaning control method and system, electronic equipment and a storage medium. The method comprises the following steps: acquiring operation information of a nursing bed and work and rest information of a target object in real time; wherein the operation information comprises humidity information and pressure information; under the condition that the humidity information is greater than a humidity threshold value, determining a high-humidity area according to the humidity information; performing infrared scanning on the high-humidity area to obtain a reflection spectrum under a specific wavelength; based on the reflection spectrum, combining a convolutional neural network and a spectrum database to obtain a substance classification result; and according to a substance classification result, triggering a cleaning and drying process and dynamically adjusting cleaning parameters to complete cleaning of the nursing bed. Thus, the pollutants can be accurately classified, the pollutants are distinguished from common water stains, sweat and other interferents, and the problems that single humidity detection is prone to misjudgment and false triggering of cleaning are solved; hydroelectric resource waste and ineffective loss of equipment caused by cleaning with fixed cleaning parameters are reduced, and meanwhile, the cleaning and drying effects are guaranteed.

The invention relates to the technical field of material detection, in particular to an online detection device for trace impurities in an electronic-grade quartz crystal material. Collecting material data of a plurality of target materials, then creating a detection model, inputting a plurality of pre-processed data into the detection model, creating a standard impurity spectrum database based on the pre-processed data of each target material through the detection model, and performing calibration by combining a dynamic correlation matching algorithm with a signal enhancement factor to obtain a standard impurity spectrum database; the method comprises the following steps of: training a target material to detect the content of impurities contained in the target material, and finally inputting real-time data into a trained detection model to detect the content of impurities contained in a real-time material. By fusing a bionic micro-nano enrichment technology and a light trace-cavity enhanced Raman technology, a signal enhancement factor reaches 10 orders of magnitude; the method can accurately detect trace impurities and micro elements in the electronic-grade quartz crystal material, and adopts multispectral combination and anti-interference design, so that the detection error is less than or equal to + / -2%, and the detection sensitivity is improved.

The invention discloses a near infrared spectrum rapid detection method for active ingredients of a compound fertilizer, and belongs to the technical field of crossing of agricultural detection technology, spectral analysis and artificial intelligence. Comprising the following steps: collecting original near infrared spectrum data of a compound fertilizer sample and a corresponding standard chemical measurement value, constructing a multi-level spectrum database, constructing a rapid detection model based on an improved one-dimensional convolutional neural network model optimized by an improved particle swarm optimization algorithm, training the rapid detection model based on a preset data set, and obtaining a rapid detection result. And inputting near infrared spectrum data to be detected into the trained rapid detection model to obtain the content of the effective components of the compound fertilizer. According to the method, a high-diversity spectrum database is constructed through data enhancement, and the robustness of the model for resisting physical interference is improved; the improved one-dimensional convolutional neural network is adopted to accurately mine deep features of the spectrum; and an improved particle swarm optimization algorithm is introduced to optimize initial network parameters, so that the training stability and convergence efficiency are improved, and the effective components of the compound fertilizer are detected.

The invention relates to the technical field of SCR (Selective Catalytic Reduction) denitration catalyst detection, in particular to an SCR denitration catalyst poisoning multispectral cooperative detection method and a related device. Then, on the basis of the spectral data of the SCR denitration catalyst and a preset poisoning characteristic spectral database, the poisoning type of the SCR denitration catalyst is obtained; wherein the spectral data of the SCR denitration catalyst cover the optical morphology, Raman spectrum, X-ray fluorescence spectrum, X-ray photoelectron spectrum and the like of the SCR denitration catalyst, all-directional information of catalyst poisoning can be obtained from different angles, and information such as the optical morphology, molecular structure, elemental composition and chemical valence state is comprehensively analyzed; therefore, comprehensive characterization of multi-poison composite poisoning is achieved, the poisoning type of the catalyst is determined, rapid and early detection of SCR denitration catalyst poisoning is achieved, and the problems that in the prior art, SCR denitration poisoning detection is single in dimension, long in detection period and the like are solved.

The invention discloses a rapid detection method and system for nutritional ingredient defects of non-fried instant noodles, and relates to the technical field of food detection.The method comprises the steps that multiple types of non-fried instant noodles are synchronously scanned through a dual-band light source, a nutritional ingredient defect three-dimensional spectrum database is constructed, and a nutritional ingredient defect prediction channel is established; meanwhile, a visual image is collected for feature recognition and defect evaluation, and a visual nutrition defect evaluation channel is constructed; and carrying out weighted fusion on the two channels to form a multi-modal defect detection channel, carrying out conjoint analysis on spectrum and image data of the target instant noodle sample, and outputting a nutritional ingredient defect detection result. According to the method, the technical problems of long detection period and low accuracy of casual inspection of products depending on laboratory physical and chemical analysis or a single spectrum technology in the prior art are solved, and the technical effect of high-precision and rapid detection of the nutritional defects of the non-fried instant noodles through a multi-modal detection mechanism combining spectral analysis and image analysis is achieved.

The invention provides an aviation kerosene multi-parameter quantitative detection method and system, and belongs to the technical field of oil detection. Obtaining an aviation kerosene standard sample and constructing a standard sample spectrum database; obtaining a to-be-detected aviation kerosene sample and carrying out spectral analysis on the to-be-detected aviation And matching a spectral analysis result with a standard sample spectral database, and realizing multi-parameter quantitative detection of aviation kerosene based on a matching result. According to the aviation kerosene multi-parameter quantitative detection method, a plurality of exponential decay functions are superposed and a randon function is introduced to simulate variation noise between samples, so that the authenticity of the model is improved; the defect of spectral image blurring caused by the light scattering effect is effectively made up through scattering compensation; and cosine similarity matching is carried out based on the feature quantity in the feature extraction layer variable, so that the problem of excessive dependence on the high-dimension vector amplitude when Euclidean distance matching is used in traditional spectrum comparison is solved, and the matching precision is improved.

The application discloses an infrared spectrum atmospheric absorption interference correction method and system, and the method comprises the following steps: establishing a historical spectrum database; based on a to-be-corrected sample, screening from the historical spectrum database to obtain a near-neighbor reference base spectrum; setting a correction branch: a reference spectrum generation branch generates a reference spectrum of the to-be-corrected sample based on the to-be-corrected sample, the near-neighbor reference base spectrum and environmental and instrument parameters; a physical correction branch constructs a differential interference component based on a time sequence spectrum window, and carries out physical constraint correction by using a multi-spectrum robust estimation and an adaptive segmented scaling factor to obtain a physically corrected spectrum; a deep learning branch carries out deep residual learning on the to-be-corrected sample, the reference spectrum and the physically corrected spectrum to obtain a deep corrected spectrum; and according to the uncertainty or consistency of the correction branch, the corrected spectrum is adaptively fused to obtain a final corrected spectrum. The application realizes more fine and more robust correction of the infrared spectrum atmospheric absorption interference.

Platforms, systems, and methods for automated omics for generalization using spectral databases in synthetic biology development. According to one aspect, there is provided a system for converting raw data from an analytical and mass spectrometry instrument to model-ready data, comprising: computing hardware configured to: receive data from the analytical and mass spectrometry instrument, wherein the data includes measurement data from a set of control samples and a set of test samples; extract a set of peak lists comprising a set of test peak lists and a set of control peak lists from the received data; compress the extracted peak lists using a compression algorithm; identify a set of metabolites that correspond to a set of peaks from the compressed peak lists by comparing a set of mass-to-charge ratios and a set of retention times associated with the set of peaks with the mass-to-charge ratios.

The invention provides a plastic sorting system based on short-wave infrared multispectral camera imaging. The plastic sorting system comprises a spectral database establishing module, a key wave band screening module, a multispectral classification model building module and an industrial grade sorting device building module. The spectrum database establishment module is used for providing data support for subsequent wave band screening and model training; the key wave band screening module is used for executing a data preprocessing and fusion type feature extraction method and screening out a key wave band with the highest discrimination degree from original wave bands; the multi-spectral classification model construction module constructs a multi-spectral classification model based on the key wave band with the highest distinction degree, and precise identification of plastic types is achieved; the industrial grade sorting device building module is used for building a short wave infrared multispectral imaging plastic sorting device simulating an industrial production line, and real-time dynamic sorting is achieved. According to the invention, dimensionality reduction of the spectral band is realized reasonably and effectively, so that an expensive hyperspectral camera is replaced, and the hardware cost and complexity of the system are reduced.

The present invention provides a spectral measurement system and a spectral thickness measurement system. The spectral measurement system includes a light source module, a dispersive element, a detector, a displacement stage controller, and a displacement stage. The light source module provides incident light, which irradiates a first region on the sample to generate a first interference light, which is incident on the detector via the dispersive element to generate a first coherent spectrum. The displacement stage controller controls the displacement stage to move the sample so that the incident light irradiates a second region on the sample to generate a second interference light, the second and first regions partially overlap, the second interference light is incident on the detector via the dispersive element to generate a second coherent spectrum, and the first coherent spectrum is compensated based on the second coherent spectrum. The spectral thickness measurement system includes the spectral measurement system and obtains the thickness of the first region based on a spectral database and the compensated first coherent spectrum. The present invention moves the position of the sample by the displacement stage and compensates for spectral information on the sample that is lost due to the limitations of the spatial resolution of the detector.

The application discloses a kind of composite fertilizer effective component near-infrared spectrum rapid detection method, belong to the cross technical field of agricultural detection technology, spectral analysis and artificial intelligence.It includes the original near-infrared spectrum data of composite fertilizer sample and its corresponding standard chemical determination value, constructs multilevel spectral database, constructs rapid detection model based on the improved one-dimensional convolution neural network model optimized by improved particle swarm optimization algorithm, based on the preset data set, the rapid detection model is trained, the near-infrared spectrum data to be measured is input into the rapid detection model after training, and the effective component content of composite fertilizer is obtained.The application constructs high diversity spectral database by data enhancement, improves the robustness of model against physical interference;Improved one-dimensional convolution neural network can accurately mine deep spectral features;Improved particle swarm optimization algorithm is introduced to optimize network initial parameters, improve training stability and convergence efficiency, realize the detection of composite fertilizer effective component.