894 results about "Ft ir spectra" patented technology

The invention discloses a near infrared spectrum rapid detection system and a real-time analysis method, and the system comprises an environmental parameter and spectrum information collection module, a data processing module, an intelligent evaluation model construction and optimization module, a real-time analysis and evaluation module, and a result feedback and optimization module. The intelligent evaluation model construction and optimization module is used for constructing an intelligent evaluation model by fusing a neural network and a support vector machine, taking the characteristic vector information which is obtained after processing and represents environmental parameters and is associated with infrared spectrum information as input, and taking an expected evaluation result which is associated with the infrared spectrum information of a sample as output; training the intelligent evaluation model; and the real-time analysis and evaluation module is used for inputting the environment parameter information and the infrared spectrum information which are acquired in real time into the trained intelligent evaluation model to obtain an evaluation result of the infrared spectrum information. The system and the method have the advantages of high detection speed, high accuracy and the like.

The invention relates to the technical field of remote monitoring, in particular to a remote monitoring system and method based on a cloud platform. The method comprises the following steps: acquiring multispectral video image data of an industrial site; analyzing the difference characteristics of visible light and infrared spectrum based on the original data of the multispectral video image, identifying the characteristic mode of electromagnetic interference, and generating spectrum difference interference fingerprint data; identifying an uninterfered area according to the spectrum difference interference fingerprint data to obtain spectrum reliability distribution data; extracting non-interfered spectrum section information based on the spectrum reliability distribution data; performing complementary fusion on the non-interfered spectrum section information and the interfered area to obtain texture enhanced video data; and analyzing the time-varying characteristic of electromagnetic interference by using the texture enhanced video data to obtain interference time sequence characteristic data. According to the invention, dynamic calibration of interference is realized by analyzing the deformation characteristic of the calibration frame, and the adaptive capacity of the system to a changing electromagnetic environment is remarkably improved.

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.

The invention provides a blood glucose monitoring and early warning method. The blood glucose monitoring and early warning method comprises the steps of a, collecting a skin surface spectral signal I (lambda), b, collecting skin surface temperature T, humidity H, environment light intensity L and environment electromagnetic interference E parameters, c, preprocessing the spectral signal I (lambda), removing noise N (I) through wavelet transform, and d, calculating the blood glucose monitoring and early warning parameters. And the signal intensity is corrected through an ambient light intensity compensation model. According to the blood glucose monitoring and early warning method, a non-invasive acquisition mode combining near infrared spectrum and visible light spectrum is adopted, and a multi-modal regression model and a dynamic interference compensation technology are innovatively introduced, so that the problem that the precision of a sensor is influenced by skin temperature, humidity and environmental electromagnetic interference in the prior art is solved; the timeliness and the stability of signal purification are improved by utilizing preprocessing and optimization algorithms of spectral signals, such as wavelet transform, empirical mode decomposition and fast Fourier transform.

The invention relates to the technical field of food component analysis, in particular to a food quality component analysis method and system based on a near infrared spectrum technology, and the method comprises the following steps: obtaining a food sample set, and obtaining original near infrared spectrum data based on the food sample set, combining the original near infrared spectrum data to form an original infrared spectrum matrix; preprocessing the original infrared spectrum matrix to generate a processed spectrum matrix; and extracting the characteristic wavelength of the near infrared spectrum information based on the processed spectrum matrix, constructing a food quality quantitative analysis model according to the characteristic wavelength in combination with a partial least squares regression algorithm, and outputting a component content predicted value according to the food quality quantitative analysis model. And generating a quality component analysis report of the to-be-detected food based on the component content predicted value. The method has the advantages that food quality and ingredients can be quickly, nondestructively and accurately detected, and detection precision and accuracy are improved.

The invention relates to the technical field of coal gangue recognition, in particular to a coal gangue AI recognition method and system based on infrared spectrum measurement, and the method comprises the steps: obtaining a near infrared spectrum of each contrast reference sample and a near infrared spectrum of each to-be-detected sample, and respectively recording the near infrared spectrums as each contrast infrared spectrum and each to-be-detected infrared spectrum; determining a contrast absorption vector and a to-be-detected absorption vector; determining the absorption peak saliency of each wave band; obtaining a target wave band; determining an absorption level vector of any target wave band; determining the absorption peak recognition degree and the potential information characteristic value of any target wave band in each infrared spectrum to be detected; determining the coal gangue recognition degree; and obtaining a training set and a training label of the classification model, and identifying the coal gangue in the to-be-detected sample. According to the method, the problem that the key feature information of the coal gangue cannot be accurately extracted due to the fact that the infrared spectrum data acquisition process is easily interfered by noise or redundant information is solved, so that the coal gangue recognition accuracy is improved.

The invention relates to the technical field of spectrum analysis, in particular to an intelligent protein quality detection method and system based on spectrum analysis, and the method comprises the following steps: collecting infrared spectrum data of a preset number of protein samples; determining a local fluctuation coefficient of the transmittance of each wavelength; determining a fluctuation coefficient of the infrared spectrum data of each protein sample; screening the infrared spectrum data of all the protein samples based on the fluctuation coefficient to obtain fluctuation data of each suspected impurity; equally dividing each piece of suspected impurity fluctuation data into each fragment; determining an abnormal index of the suspected impurity fluctuation data in each segment; correcting a spatial domain weight and a value domain weight of each filtering window in a bilateral filtering algorithm based on the abnormal index, and denoising infrared spectrum data of the protein sample; and determining a difference coefficient of the infrared spectrum data of each protein sample, and detecting the protein quality. Therefore, the accuracy of protein quality detection is improved.

The invention discloses a method and a system for rapidly detecting the content of aluminum in bauxite based on a near infrared spectrum, and aims to meet the cross-field requirements of geoscience, resource exploration, artificial intelligence and the like. A traditional bauxite Al2O3 detection method is high in cost, expensive in used instrument and low in efficiency, and application of the traditional bauxite Al2O3 detection method in the fields of scientific research, industry and management is limited. In order to solve the problems, the method comprises the following steps: establishing a sample database by utilizing collected near infrared spectrum data of bauxite and Al2O3 standard values corresponding to samples; according to the method, a deep learning model based on DenseNet and Roformer is constructed and trained, so that the deep learning model is suitable for capturing a nonlinear relationship between a bauxite spectrum and the Al2O3 content of the bauxite spectrum, and rapid detection and analysis of Al2O3 of the bauxite can be simply and efficiently realized.

The invention discloses a soil heavy metal inversion method and system integrating satellite remote sensing and near-end sensing, and the method comprises the steps: collecting a soil sample, and measuring the soil heavy metal content and a soil visible light-near infrared spectrum; obtaining a time sequence multispectral image of a research area, calculating a spectral index, and selecting and screening bare soil pixels through a threshold value to obtain a bare soil image; performing spectrum correction on the bare soil image; obtaining a joint dictionary and a sparse coefficient through sparse representation and dictionary learning, and reconstructing a hyperspectral image of the bare soil image; converting the one-dimensional spectral data into a two-dimensional spectrogram by using continuous wavelet transform, extracting spectral features in combination with a 2D-CNN algorithm, and constructing a soil heavy metal inversion model; and using the trained inversion model to predict the soil heavy metal content of the research area based on the reconstructed hyperspectral image. According to the method, satellite remote sensing and near-end sensing are integrated to obtain a large-scale accurate soil heavy metal content distribution map, deep features are extracted in combination with a 2D-CNN algorithm, and the inversion model precision and model efficiency are improved.

The invention relates to the technical field of remote sensing spectrums, in particular to an urban black and odorous water body recognition method and system based on remote sensing spectrums. The method comprises the following steps: acquiring infrared spectrum and visible spectrum data based on a remote sensing satellite or an unmanned aerial vehicle, processing the data, eliminating noise and abnormal values in the data, and extracting spectral features to obtain a spectral feature data set; according to the invention, through infrared and visible spectrum data collected by a remote sensing satellite or an unmanned aerial vehicle and through combination with urban water body data of a geographic information system, the dimension and quality of the data are improved, so that monitoring of the urban black and odorous water body is not limited to surface coverage, and sensitivity of data analysis can be accurately regulated and controlled by dynamically adjusting a spectrum band weight; the method has the advantages that the method adapts to variable environmental conditions, recognition precision of various polluted water bodies is improved, continuity and accuracy of long-term monitoring are guaranteed by continuously updating and optimizing the processing flow, and urban water quality management and decision-making processes are effectively supported.

A rock mass mechanical parameter prediction method based on an underground powerhouse of a pumped storage power station relates to the technical field of mechanical parameter prediction, and comprises the following steps: identifying an embedding position and distribution characteristics of a sulfur-containing shale interlayer by constructing a three-dimensional geologic structure model and lithologic information of a construction area, and establishing a structural domain partition parameter prediction area system; constructing a nonlinear regression function model taking infrared spectrum characteristic factors and environment control parameters as input and taking elasticity modulus and the like as output by combining infrared spectrum characteristic data and a historical evolution track; the model is embedded into a regional system for dynamic assignment, and a parameter space-time change atlas is constructed; the laser point cloud data and the micro-seismic monitoring data are fused to invert the actual response of the surrounding rock; when the monitoring value deviates from the prediction map, triggering parameter degradation function re-calibration to complete parameter dynamic updating and prediction closed loop; according to the method, the accuracy of rock mass mechanical parameter prediction of the pumped storage power station underground powerhouse can be improved.

The invention provides a harmful organic matter infrared spectrum detection method based on segmented multi-dimensional harmonic decomposition, relates to the technical field of infrared spectrum analysis, and aims to solve the problems of noise interference, peak overlapping and inaccurate signal extraction during harmful organic matter detection in the prior art. Obtaining a sample solution to be detected and obtaining an original infrared spectrum of the sample solution to be detected; the method comprises the following steps of: segmenting a dynamic wave band, and performing local signal decomposition on each segment by adopting a harmonic primary function with a proper order; the absorption peak of the target harmful organic compound is accurately extracted by utilizing wave band coupling and characteristic peak identification; stable and accurate quantitative analysis is realized by combining partial least squares, so that the detection sensitivity and accuracy are improved, and the safety of food contact materials is ensured; according to the method, the accuracy and reliability of signal denoising, peak identification and quantitative analysis under a complex matrix are effectively improved by adopting a wave band segmentation strategy, segmented multi-dimensional harmonic decomposition and wave band coupling and characteristic peak identification.

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.

The invention provides a near-infrared spectrum rapid detection method and system for effective components of a herbicide, which are applied to the technical field of agricultural detection.The near-infrared spectrum rapid detection method comprises the steps of obtaining near-infrared spectrum data of a to-be-detected soil area and moisture content and compactness numerical values reflecting soil texture differences, and calculating the effective components of the herbicide on the basis of a preset interval to which the moisture content numerical value belongs; according to the method, the corresponding preset scattering correction parameter is selected, the actual scattering correction parameter is obtained through compactness numerical calculation, fine adjustment of the scattering correction parameter is achieved, and the quality and reliability of spectral data are improved through spectral correction; and the corrected spectrum is analyzed, so that the real condition of the herbicide in the soil can be reflected more accurately. Therefore, the interference of complex soil texture in the field on near infrared spectrum herbicide detection is effectively overcome, and the beneficial effect of rapidly and accurately detecting the effective components of the herbicide under the complex soil condition is achieved.

The invention belongs to the technical field of spectrum detection and analysis, and discloses an infrared spectrum-based rubber auxiliary component analysis and quality control method and a system thereof, the method comprises the following steps: using a Fourier transform infrared spectrometer to collect infrared spectrum data of a rubber auxiliary mixture in real time, and preprocessing the infrared spectrum data; based on the preprocessed infrared spectrum data, a partial least square regression method is adopted to construct a multi-component quantitative model for component identification and quantitative analysis; according to a result of component identification and quantitative analysis, generating a content report and quality control information of the rubber auxiliary component; according to the method, the influence of environmental interference on infrared spectrum data can be effectively reduced, and real-time and accurate quantitative analysis and quality control can be carried out on each component in the rubber additive production process.

The invention relates to the technical field of infrared spectroscopic analysis, in particular to a spectroscopic analysis-based corn kernel nutrient abundance and deficiency state evaluation method, which comprises the following steps of: irradiating corn kernels by using near-infrared and mid-infrared light sources, collecting reflected spectral signals, and recording spectral signal data; and based on the spectral signal data, extracting key wavelength region data to obtain key spectral data. According to the invention, the near-infrared light source and the mid-infrared light source are combined, so that the capability of acquiring nutrient information of the corn kernels is enhanced, the reflection characteristic of spectral signals is utilized, the interference of environmental factors on data acquisition is reduced, and the stability of spectral data is improved. The screening of key wavelength regions reduces irrelevant information, so that the spectral data is more concentrated on the characteristic wave bands influencing the nutrient state. The spectral signals are processed through wavelet transform, the resolution ratio of the characteristic peak value is improved, the local change trend of the spectral signals is enhanced, and characteristic extraction is more accurate.

The invention discloses a pixel-level multispectral fusion small target detection method and system based on mask enhancement. The method comprises the following steps: acquiring remote sensing image data to be processed, wherein the data comprises a multispectral image pair of a visible light spectrum and an infrared spectrum; splitting the multispectral modal image pair and inputting the split multispectral modal image pair into a feature fusion module at the same time to obtain respective masks, then obtaining feature information after mask processing according to the respective masks, performing feature fusion on the feature information and the split original image to obtain a first fusion result of a single modal feature, and then performing channel splicing and merging to obtain a second fusion result of the single modal feature; obtaining a result after the second fusion; inputting a result after the second fusion into an SE module, screening out effective information features, inhibiting redundant features, and obtaining an image after pixel-level fusion; and inputting the image after pixel-level fusion into a lightweight detection network to obtain a detection result. According to the invention, the detection precision of small targets in multi-modal remote sensing image target detection is improved.

The invention discloses an infrared spectrum chemometrics analysis method and system, and relates to the technical field of infrared spectrum analysis. Comprising the following steps: acquiring a plurality of infrared spectrum images with known spectral characteristic parameters, and preprocessing the infrared spectrum images to obtain a training sample image set; based on the training sample image set, establishing a neural network prediction model, taking the infrared spectral image as input and the spectral characteristic parameters as labels, and training to obtain a spectral characteristic parameter extraction model; acquiring an infrared spectrum image of an object to be detected, preprocessing the infrared spectrum image, inputting the preprocessed infrared spectrum image into the trained model, outputting spectrum characteristic parameters, and acquiring environmental parameters; correcting the spectral characteristic parameters based on the environmental parameters, and further correcting the spectral characteristic parameters in combination with the physical characteristic parameters of the to-be-detected article to obtain an accurate value; and completing the stoichiometric analysis of the property and content of the to-be-detected article based on the accurate spectral characteristic parameters. The true values of the spectral characteristic parameters are accurately simulated, and the accuracy of an analysis result is remarkably improved.

The invention provides a near infrared spectrum qualitative discrimination method and system, and the method comprises the steps: collecting a near infrared spectrum data set, and converting a one-dimensional spectrum sequence into a polar coordinate image; constructing a lightweight convolutional network architecture, and processing the polar coordinate image; k samples of each type are dynamically extracted from the training set, a support set is constructed based on the K samples and a few-sample strategy, and a query support similarity matrix is constructed through a cosine similarity algorithm; and judging the test set based on the query support similarity matrix, the support set and a preset threshold to obtain the highest similarity between the test set and the support set, and if the highest similarity exceeds the preset threshold, judging that the test set is a corresponding category. According to the method, the cost and time of sample collection and preparation can be reduced, the feature information of the spectral data is effectively extracted, and the method has relatively high flexibility during discrimination.

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 discloses a passive Fourier infrared spectrum gas telemetering control system, which relates to the technical field of telemetering control and comprises an environment sensing module, a dynamic interference correction module, a multi-modal fusion module, a decision display module and an execution feedback module. The environment sensing module synchronously obtains three-dimensional topographic data, a surface reflection polarization state and atmospheric turbulence intensity parameters through a laser radar, a multispectral polarization camera and a quantum meteorological sensor, generates a dynamic topographic-meteorological coupling field model, and corrects spectral signals in real time based on the coupling field model. The terrain reflectivity, the slope angle, the polarization state and the turbulence intensity are comprehensively considered, the accuracy and robustness of spectral signal correction are improved, the AR holographic projection unit is used for converting four-dimensional data into a three-dimensional visual scene in real time, a three-level early warning mechanism is established, and the decision making efficiency and the response speed are improved.

The invention discloses a space-space-ground-well integrated carbon dioxide sequestration leakage monitoring method and system, and relates to the technical field of carbon dioxide sequestration leakage monitoring, and the method comprises the steps: carrying out the remote sensing measurement of a target region through a synthetic aperture radar and a satellite, and determining a carbon dioxide sequestration leakage monitoring boundary; performing photogrammetry by using an unmanned aerial vehicle, and constructing a three-dimensional real scene model Surveying the target area on site to obtain surveying and mapping data; based on the key monitoring area, an absorption spectrum and earth surface leakage flux are obtained through infrared spectrum monitoring and vorticity monitoring; gravity monitoring data, storage area underground water parameters, carbon dioxide plume distribution conditions and logging sound wave frequency distribution characteristic graphs are obtained through gravity monitoring, underground water monitoring, optical fiber monitoring and sound wave logging monitoring. And through comprehensive analysis, according to a comprehensive analysis result, whether carbon dioxide leaks is judged, and a carbon dioxide sequestration leakage monitoring result is obtained. The accuracy of carbon dioxide sequestration monitoring can be effectively improved.

The invention relates to the technical field of infrared light analysis, in particular to a method and a system for rapidly determining nicotine content in tobacco tar by utilizing near infrared spectrum. The method comprises the following steps: performing multi-angle near infrared spectrum acquisition and multi-depth spectrum acquisition on a tobacco tar sample to obtain optimal angle spectrum data and optimal depth spectrum data, and performing tensor fusion on the optimal angle spectrum data and the optimal depth spectrum data to obtain a complete tobacco tar spectrum data set; dividing the complete tobacco tar spectrum data set to obtain a tobacco tar window spectrum fragment data set, and performing tobacco tar window competitive screening on the tobacco tar window spectrum fragment data set to obtain a primary tobacco tar characteristic window set; and performing stable tobacco tar characteristic wave band screening on the primary tobacco tar characteristic window set to obtain a stable tobacco tar characteristic wave band set. According to the invention, rapid, real-time and high-precision detection of nicotine content is realized, and personal error and operation complexity are significantly reduced.

The invention relates to the technical field of American ginseng quality detection, in particular to an American ginseng quality detection method and system based on S-transformation and multi-task deep learning. Near infrared spectrum data of American ginseng to be detected are extracted, and time-frequency conversion is performed on the near infrared spectrum data through S-transformation; obtaining a time-frequency characteristic pattern used for representing time-frequency domain information of the near infrared spectrum; and inputting the to-be-detected American ginseng time-frequency feature map into a pre-trained multi-task deep learning model, and identifying the producing area of the to-be-detected American ginseng and predicting the quality index of the to-be-detected American ginseng by using the multi-task deep learning model. The multi-task deep learning model comprises a feature extraction network used for extracting time-frequency features of each detection task of the feature map, a feature interaction network used for enhancing feature complementation between the tasks and performing feature fusion, and a multi-task head network used for executing a production place identification task and a quality index prediction task according to the time-frequency features and outputting the tasks. The method can be suitable for quality analysis and origin classification of small American ginseng samples.

The invention provides a control method for automatically collecting and eliminating target infrared characteristics, and relates to the technical field of target infrared characteristic elimination, and the method comprises the steps: obtaining real-time target and environment infrared data of a target of which the infrared characteristics are to be eliminated, carrying out the noise reduction processing, and obtaining continuous infrared spectrum data through spectrum integration; according to each environment background temperature, temperature adjustment is performed on each target surface, wavelet transform is performed on infrared spectrum data to obtain the temperature rise characteristic of the target, parameter optimization is performed on temperature adjustment, and region classification is performed on the adjusted target region to obtain a target region classification result. And carrying out smooth temperature adjustment on the classification areas exceeding the judgment threshold according to the temperature of the centers of the classification areas, calculating the phase, frequency and amplitude of an interference radiation spectrum according to the difference value between the real-time infrared radiation distribution data of the target and the environment and the position of a radiation interference source, adjusting interference waves, and eliminating the infrared characteristics of the target according to the interference principle.

The invention discloses a greenhouse gas infrared spectrum fitting method based on instrument linear function iterative inversion, and relates to the technical field of atmospheric greenhouse gas infrared spectrum measurement. According to the greenhouse gas infrared spectrum fitting method based on instrument linear function iterative inversion, through iterative calibration based on infrared spectrum parameters and the instrument linear function, greenhouse gas concentration is obtained, a calibration spectrum is fitted, dynamic conditions such as current environment temperature and gas pressure are obtained, and an absorption cross section reference value in a database is combined, so that the greenhouse gas infrared spectrum fitting method based on instrument linear function iterative inversion is obtained. According to the method, the absorption cross section value under the real condition can be calculated for each wavenumber point, errors caused by fixed parameter hypothesis are eliminated, in addition, the matching degree of a calibration spectrum and an actual measurement spectrum is remarkably improved by iteratively optimizing the center wavenumber and the resolution width of an instrument, and particularly under the conditions of high noise and complex multi-component interference, the accuracy of the calibration spectrum is greatly improved. And the overall precision of spectrum fitting is improved, so that the accuracy and reliability of greenhouse gas concentration inversion data are ensured.

The invention relates to the technical field of gasoline detection, in particular to a near infrared spectrum interference correction method and a near infrared spectrum interference correction system for vehicle gasoline detection. During vehicle gasoline near infrared spectrum correction, near infrared spectrum data of interference components in gasoline are collected, and characteristic wave bands of the interference components are analyzed; then whether the characteristic wave bands of the target component and the interference component coincide or not is judged to determine candidate interference wave bands, different interference correction methods are determined according to the interference indexes of the corresponding candidate interference wave bands, that is, different interference correction methods are adopted for different candidate interference correction wave bands, and the pertinence and accuracy of interference correction are improved.

The invention provides a soil quick-acting potassium measuring method and system and a medium, and relates to the technical field of soil data measurement, and the method comprises the following steps: extracting near infrared spectrum data of soil to be measured; solving Euclidean distances between one row and other rows in the spectrum matrix, and taking the number of distances smaller than a preset threshold value as the number of neighbors of the row; traversing the spectrum matrix, constructing a neighbor number array, normalizing a density vector, and transposing and assigning the density vector as an initial weight of a competitive adaptive reweighted sampling (CARS) method; determining an optimal regularization parameter through a Bayesian optimization method, fitting a Lasso model, obtaining a coefficient matrix of the Lasso model, and determining a regression coefficient of each spectral feature; and screening spectral features with non-zero regression coefficients in the near infrared spectral data of the soil to be detected to form an effective feature matrix. According to the method, the calculation complexity of the CARS method is reduced, parameter setting is simplified, overfitting is prevented, the model fitting precision is improved, and the prediction performance of soil available potassium measurement is improved.

The invention discloses a pea seed quality detection method and device based on an infrared spectrum, and relates to the technical field of seed screening. Comprising the following steps: collecting infrared spectrums of sample pea seeds with known quality parameters, carrying out denoising and enhancement pretreatment, and establishing a neural network prediction model in combination with the quality parameters (water content, crude fiber, fat, starch and protein); after same preprocessing is carried out on the infrared spectrum of a seed to be detected, the trained model is input, and quality parameters of the seed to be detected are predicted; calculating a nutrition richness factor and a water swelling potential factor based on the predicted value, and combining with a surface image to extract a color channel value to calculate a color uniformity coefficient; the quality index is calculated by integrating information such as color uniformity, circularity and surface roughness, and is compared with the preset judgment threshold to generate the quality judgment result, so that the quality judgment result of the pea seeds can be quickly output, and the requirements of modern agriculture on high-throughput detection and intelligent analysis are met.

The invention belongs to the technical field of target detection, particularly relates to an infrared weak and small target detection method based on cross-domain feature migration, and provides a novel detection network which is SA-YOLO for short. Firstly, a position guidance enhancement module is designed under the inspiration of a multispectral feature migration idea, and infrared spectral feature response of a weak and small target is effectively enhanced through a triple structure of position perception attention, multi-scale significance fusion and position guidance filtering. Secondly, a spectrum-space sensing module is provided, and a progressive voidage and a double-branch attention mechanism are adopted, so that robustness modeling of multi-scale spatial features is realized. Then, a sensitivity stability loss function is innovatively proposed, and the problem of sudden drop of scale sensitivity of a weak and small target is relieved through dynamic penalty weight and composite normalization. And finally, an infrared small target data set named as ISUAV-D is also collected. Experimental results show that SA-YOLO shows significant advantages in a strong light interference environment and a tiny target detection scene.