47 results about "Hyperspectral reflectance" patented technology

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 belongs to the technical field of image processing, and particularly relates to an intelligent agricultural product pesticide residue detection method and system based on a spectrum technology, and the method comprises the steps: obtaining the hyperspectral reflectivity according to the original hyperspectrum, dark current and whiteboard image data; calculating an illumination confidence factor by using the average spectral intensity; performing linear scaling on the obtained standard biological matrix background spectrum according to the illumination confidence factor to obtain a local background estimated value and obtain a net residual spectrum; generating a pesticide residue distribution thermodynamic diagram by combining the cosine similarity of the net residual spectrum and the target pesticide standard fingerprint spectrum and the noise suppression weight based on the illumination confidence factor; and pesticide residue detection is carried out according to the statistical characteristics of the connected region of the suspected defect mask. Through adaptive background deduction and noise suppression, the problems of uneven curved surface illumination and strong background interference are solved, and the detection accuracy is improved.

The invention discloses a coastal salt marsh wetland non-photosynthetic vegetation coverage inversion method, which comprises the following steps: S1, obtaining hyperspectral reflection curves of soil, non-photosynthetic vegetation and photosynthetic vegetation of the coastal salt marsh wetland of the Yellow River Delta, simulating and measuring spectral characteristics of samples under different moisture contents (dry, low humidity, medium humidity and saturation) in a laboratory, and calculating the coverage of the non-photosynthetic vegetation in the coastal salt marsh wetland; key wave bands of non-photosynthetic vegetation and other ground features are screened and distinguished; s2, constructing a moisture-insensitive hyperspectral NPV index (MINI), and determining an optimal wave band combination through wave band traversal; and S3, constructing mixed data of soil, non-photosynthetic vegetation and photosynthetic vegetation under different moisture conditions in a laboratory, and evaluating inversion precision and stability of MINI under different moisture conditions. S4, acquiring a coastal salt marsh wetland hyperspectral image, and calculating an NDVI value and an MINI index value based on the processed image; and S5, performing spectral unmixing on the mixed pixels based on a triangular space method, and inverting the non-photosynthetic vegetation, photosynthetic vegetation and soil coverage of the research area. The method provided by the invention solves the technical problems of low precision and poor stability of the traditional NPV index inversion coverage under the condition that the coastal salt marsh wetland is influenced by tides and the moisture change is obvious.

The invention discloses a copper alloy surface wiredrawing quality detection method and system, and the method comprises the steps: carrying out the multi-source defect feature collaborative analysis of hyperspectral reflectivity data, linear array image data and surface three-dimensional point cloud data, and obtaining an oxidation defect marking graph, a scratch distribution graph and a texture anomaly graph; inputting the oxidation defect mark graph, the scratch distribution graph and the texture anomaly graph into a pre-trained convolutional neural network, wherein the convolutional neural network outputs and obtains defect space coordinates; and calculating the unit area defect density, the maximum defect area and the deepest scratch depth according to the defect space coordinates, and determining the surface wire drawing quality of the copper alloy according to the unit area defect density, the maximum defect area and the deepest scratch depth. The industrial problems of multi-modal data splitting, missed judgment of tiny defects and misalignment of grading standards can be well solved.

A method for performing point-to-point whiteboard parameter ratio correction on a hyperspectral image includes: capturing hyperspectral data of a standard reference whiteboard in advance as white(x,y,w), and storing the records, then captures hyperspectral data of a sample as sample(x,y,w); then selecting an unobstructed and unshaded whiteboard area within a certain range of the hyperspectral data sample(x,y,w) of the test sample, and labeling the area as Area A; calculating a spectral average of the ROI on the sample image as SA(w); calculating a spectral average of whiteboard data in the same position as the ROI as WA(w); dividing the two spectral averages, and obtaining a correction coefficient alpha(w)=WA(w). / SA(w); multiplying the whiteboard parameter ratio correction coefficient alpha(w) by a sample reflectance image matrix after whiteboard parameter ratio correction to obtain a final hyperspectral reflectance image matrix REFL(x,y,w)=alpha(w).*sample(x,y,w). / white(x,y,w).

This invention discloses a method for inverting equivalent water thickness in plant leaves across species. This method selects equivalent water thickness with more explicit physical meaning as a representative indicator of plant leaf water content. Based on traditional vegetation indices, it combines continuous removal of preprocessing parameters to construct absorption feature parameters, thereby reducing noise in hyperspectral data while deeply mining its inherent spectral characteristics. It innovatively introduces one-heat coding technology to explicitly process species information and employs a Bayesian optimized machine learning model, effectively improving the accuracy, stability, and computational efficiency of cross-species water inversion. This invention can accurately invert the water status of crops across species, providing a new method for rapidly acquiring crop water status based on hyperspectral reflectance characteristics. It can be widely applied in intercropping and relay cropping patterns and large-scale survey and analysis scenarios, greatly improving the accuracy and speed of water stress diagnosis.

The present application belongs to the technical field of agricultural remote sensing and crop yield estimation, and specifically discloses a winter wheat yield estimation method based on a combination of spectral characteristic indexes in key growth periods. The method is based on winter wheat canopy hyperspectral reflectance data, and first constructs spectral characteristic indexes FVI and a spectral characteristic index set. Then, the core indexes related to yield estimation are obtained through progressive screening, and the index combination optimization is carried out in each growth period respectively, and further growth period combination optimization is carried out, so as to determine the target key growth period combination for yield estimation. On this basis, a cross-period dynamic feature is further constructed, and a two-stage XGBoost winter wheat yield estimation model oriented to key growth period information utilization is established, so as to realize accurate estimation of winter wheat yield.

This invention relates to the field of optical monitoring technology for water environment, and discloses a method for analyzing and characterizing eutrophication components of water bodies based on hyperspectral feature inversion. The method includes: retrieving the intrinsic absorption spectrum sequence of pure water as a physical constraint benchmark; calculating the ratio of the hyperspectral reflectance to be measured to the benchmark to generate a modulation vector; determining the fractional-order differential sequence related to the sampling wavelength; extracting trough features through morphological baseline correction; completing the fractional-order differential transformation by combining the order sequence; extracting the characteristic trough depth and skewness parameters; and outputting the concentrations of chlorophyll a and colored soluble organic matter using an unmixing model. This invention utilizes the benchmark to drive the dynamic evolution of the differential operator, achieving topological separation of the scattering background and trace component absorption characteristics in different bands, suppressing local waveform distortion caused by background scattering heterogeneity, and improving the accuracy of component inversion under complex matrices.

The application belongs to the technical field of plant disease nondestructive detection, and discloses a nondestructive detection method and system for citrus Huanglongbing based on hyperspectral reflection and transmission fusion, which comprises the following steps: acquiring hyperspectral reflection spectrum data and hyperspectral transmission spectrum data of a to-be-detected citrus leaf, obtaining single reflection characteristic band data and single transmission characteristic band data after pretreatment and band selection; based on the single reflection characteristic band data, the single transmission characteristic band data and the spectrum data after feature-level fusion of the two, first, second and third prediction calculation results are obtained through respective optimal classification models; the corresponding weights are determined according to the classification accuracies of the optimal classification models, and the first, second and third prediction calculation results are subjected to decision-level fusion calculation to obtain the Huanglongbing grade of the citrus. The method introduces feature-level and decision-level fusion, has the discrimination ability of sample surface physical form and internal structure composition, and can quickly and nondestructively detect the Huanglongbing of the citrus.

The invention discloses a deep learning-based carbon nitrogen ratio estimation method for hyperspectral reflectivity of a leaf, and belongs to the technical field of agricultural phenotype monitoring. The core of the method lies in that a spectrum adaptive multi-scale convolutional neural network is constructed, and a network framework sequentially comprises an integral-differential collaborative feature extraction module used for enhancing local and global features of a spectrum; the multi-scale spectrum convolution module is used for extracting multi-scale absorption characteristics related to the carbon and nitrogen substances; the spectrum adaptive self-attention module is used for modeling inter-band long-range dependence and focusing a key spectrum area; and the KAN module is used for high-precision fitting of the complex nonlinear relation between the spectral characteristics and the carbon nitrogen ratio. According to the invention, full-band spectral information is fully utilized through the architecture, the problems of insufficient feature utilization and low model precision in the prior art are solved, lossless and high-precision estimation of the carbon-nitrogen ratio of the leaf is realized, and the method is suitable for precision agriculture and ecological environment monitoring.

The invention provides a corn biomass hyperspectral estimation method and system based on fractional order differential optimization. The method comprises the following steps: collecting corn planting data of a target area; processing the original hyperspectral gray value image to obtain a multiband surface reflectance image, and obtaining original hyperspectral reflectance through ROI (Region of Interest) extraction and SG filtering; performing FOD fractional order differential transformation on the original hyperspectral reflectivity before and after the NDVI mask processing to obtain FOD spectral reflectivity characteristics under different orders; constructing a plurality of dual-band spectrum indexes and a plurality of three-band spectrum indexes, and performing high-dimensional feature screening on the plurality of dual-band spectrum indexes and the plurality of three-band spectrum indexes to form a candidate feature set; and by taking the candidate feature set as an input variable and the aboveground biomass actually measured on the ground as an output variable, constructing an aboveground biomass estimation model, carrying out training optimization on the aboveground biomass estimation model, and outputting an optimal estimation strategy of the aboveground biomass of the corn.

This invention belongs to the field of environmental monitoring technology and discloses a remote sensing inversion method for dissolved inorganic nitrogen and silicate in estuaries based on salinity synergy. The method includes the following steps: S1. During a field survey in the land-sea interaction zone of the estuary, remote sensing reflectance, salinity, and nutrient data are collected simultaneously. The measured hyperspectral reflectance is simulated as the equivalent reflectance of the satellite band using the spectral response function of the target satellite, which is used to construct the training and validation datasets for the model. S2. A nonlinear inversion model of remote sensing reflectance and salinity is established, and a nutrient mixture model of salinity and nutrients is constructed. The training dataset is used for model training. S3. The satellite remote sensing reflectance data of the estuarine area to be predicted is input into the trained nonlinear inversion model and combined with the nutrient mixture model to generate the spatial distribution of DIN concentration and DSi concentration in the estuarine area. This method achieves high-precision remote sensing inversion prediction of DIN concentration and dissolved silicate DSi concentration in the estuarine area.

The application relates to the technical field of metal mine development, and discloses a soil heavy metal content estimation method, device, equipment and medium, the method comprising the following steps: collecting soil samples in a target area, and determining the contents of various heavy metal elements in the soil samples; acquiring visible light-near infrared hyperspectral reflectance data and X-ray fluorescence spectrum data of the soil samples; calculating competition coefficients among the heavy metal elements by using an L-V competition model, and determining a competition element set of a target heavy metal element; based on band importance of a random forest, a multi-objective optimization function is constructed, the function aims to maximize spectral response differences of the target heavy metal element and minimize spectral response differences of competition elements of the target heavy metal element, and the NSGA-II algorithm is adopted to screen an optimal band bit feature set in combination with wavelength interval constraints, and a machine learning regression model is trained to estimate the heavy metal content of soil in a to-be-measured area. Through the scheme, the inversion precision and stability of the model are significantly improved.

The application discloses a kind of soil-vegetation synergic interaction radiation transmission model construction method and system, belong to soil remote sensing monitoring technical field.The method includes: obtaining dynamic parameter, the parameter of target area crop is output soil state parameter and vegetation state parameter by running crop growth model;Parameter conversion is carried out by calling mechanism constraint transfer operator;Coupling radiation transmission model, and based on the sensor response function convolution method of radiation transmission mechanism, the hyperspectral reflectance data output by the canopy radiation transmission module is smoothed, and the equivalent reflectivity of each band is calculated by numerical integration;Output soil-crown complex mixed reflection characteristic data.The application breaks through the bottleneck problem of traditional radiation transmission model in soil organic carbon remote sensing monitoring under vegetation coverage scene, and finally achieves significant technical effect in spectral simulation accuracy, cross-platform applicability, scene generalization ability and application support value.

The invention discloses a reconstruction method and system for full-field-of-view hyperspectral reflectivity. The method comprises the following steps of: firstly, acquiring radiance data of a target scene through hyperspectral imaging, and imaging diffuse reflection standard plates arranged at a plurality of positions under the same illumination and observation geometric conditions to obtain corresponding radiance data of the standard plates; secondly, geometric registration and multi-view-field splicing are carried out on the target scene and the standard board image, and reference radiance distribution covering the range of the target scene is constructed; on the basis, a continuous reference radiance field covering a full view field is reconstructed by utilizing a band-by-band surface fitting technology, and finally, a conversion function is established by combining with the nominal reflectivity of a standard plate, so that pixel-by-pixel reflectivity calculation and output are realized. The system comprises a hyperspectral imaging unit, an illumination and geometric control unit, a standard board laying unit and a data processing unit. The method is suitable for hyperspectral reflectivity consistency reconstruction of a large-scale target scene.

The invention provides a fine simulation method and system for total primary productivity of tree species in the technical field of ecological remote sensing and carbon cycle simulation. The method comprises the following steps: S1, acquiring leaf hyperspectral reflectivity data, photosynthetic physiological parameter data and canopy hyperspectral remote sensing image data of different tree species in a target area; s2, extracting characteristic wave bands based on the maximum carboxylation rate in the hyperspectral reflectivity data of the leaves and the photosynthetic physiological parameter data, constructing a characteristic wave band combination, and constructing a Vcmax hyperspectral inversion model based on the characteristic wave band combination; step S3, tree species classification is carried out by using the canopy hyperspectral remote sensing image data, and a dynamic Vcmax distribution diagram is generated in combination with the Vcmax hyperspectral inversion model; and S4, replacing static Vcmax parameters of the ecological process model based on the dynamic Vcmax distribution diagram, and inputting driving data into the ecological process model to simulate a GPP space-time pattern. The method has the advantages that the accuracy and practicability of tree species total primary productivity simulation are greatly improved.

The invention discloses a spectrum identification method for early warning of summer corn nitrogen phosphorus and potassium nutrition imbalance, and relates to the technical field of agricultural information monitoring and spectral analysis, and the method comprises the following steps: S1, obtaining hyperspectral reflectivity data of leaves at different leaf positions of a summer corn target plant in a key growth period; and S2, based on the hyperspectral reflectivity data, calculating a nitrogen-phosphorus-potassium collaborative diagnosis spectral index. According to the spectrum identification method for realizing early warning of the summer corn nitrogen-phosphorus-potassium nutrition imbalance, comprehensive judgment and early risk quantification of the summer corn nitrogen-phosphorus-potassium nutrition imbalance condition are realized by constructing a collaborative diagnosis spectrum index fused with nitrogen, phosphorus and potassium multi-element information and introducing a dynamic decision matrix based on a leaf position response time sequence rule. According to the method, the limitation of traditional single element diagnosis is overcome, and spectrum weak abnormal signals of different leaf positions of crops before visible symptoms appear can be captured, so that the early warning opportunity is advanced.

The embodiment of the invention discloses a hyperspectral reflectivity determination method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring a target hyperspectral image and a target depth image; determining pixel points having a mapping relationship between the target hyperspectral image and the target depth image, and for each pixel point of at least part of the pixel points of the target hyperspectral image, determining the object distance of the pixel point based on the object distance of the pixel point having the mapping relationship with the pixel point in the target depth image; acquiring a plurality of standard hyperspectral images respectively acquired for the standard plates under different object distances; and for each pixel point of at least part of the pixel points of the target hyperspectral image, determining the reflectivity of the pixel point by using the spectral characteristics of the pixel point, the spectral characteristics of the reference pixel point and the standard reflectivity of the standard plate. According to the scheme, the reflectivity corresponding to each pixel point can be determined one by one by taking the pixel point as the minimum unit, so that the reflectivity of the target area can be accurately obtained.

The invention relates to the technical field of corn canopy nitrogen content monitoring, and discloses a corn canopy nitrogen content monitoring method and system, electronic equipment and a storage medium, and the method comprises the steps: obtaining multi-source remote sensing data of a target corn field area, the multi-source remote sensing data comprises a multi-spectral image and a hyperspectral reflectivity, the method comprises the following steps: constructing a multispectral vegetation index and a texture index based on a multispectral image, constructing a hyperspectral vegetation index, constructing a mixed feature data set by using the multispectral vegetation index, the hyperspectral vegetation index and the texture index, and carrying out feature screening on the mixed feature data set by adopting an LASSO regression algorithm to obtain an optimal feature combination; and training a random forest model through a training sample set formed by the optimal feature combination and a corresponding corn leaf nitrogen concentration measured value to form a corn canopy nitrogen content monitoring model, and obtaining a corn canopy leaf nitrogen content predicted value of the target corn field area through the corn canopy nitrogen content monitoring model. And the nitrogen condition of the corn plant in the target area can be effectively estimated.

The invention discloses a pineapple picking control method and system based on image recognition and positioning and a picking machine, and relates to the technical field of agricultural equipment.The method comprises the following steps that firstly, images of a pineapple planting area are collected, a pineapple target to be picked is recognized through a machine learning network, and the position of the pineapple target is marked as a target task point; then the pineapple picking machine is controlled to go to a task point, images of to-be-picked pineapple plants are collected, and a sliding distance is set to gradually slide downwards for height monitoring; the width and curvature changes of the detection image in the adjacent height monitoring intervals are analyzed to determine the gliding distance, and then the initial shearing height is set; acquiring hyperspectral reflection data below the height, sliding to a detection interval according to a preset step length, matching with the stem component data, and stopping sliding to determine an accurate shearing interval; finally, the accurate shearing position is determined for picking, and efficient and accurate pineapple picking is achieved.

The present application belongs to the technical field of crop disease monitoring, and particularly relates to a hyperspectral wheat scab severity monitoring method based on UFD and Triplet Attention-1D-CNN, comprising the following steps: S101. Adopting unwinding Fourier decomposition (UFD) to perform spectral enhancement processing on wheat canopy hyperspectral reflectance data; S102. Constructing an oversampling strategy based on class proportion IR regulation to perform balanced processing on the spectral samples; S103. Training a one-dimensional convolutional neural network model based on the balanced spectral samples to realize wheat scab severity grading identification. A complete technical scheme from data preprocessing, sample balancing to intelligent identification is formed, rapid, accurate and automatic grading monitoring of wheat scab severity is realized, and effective technical support is provided for precision agriculture decision-making.

The invention discloses a soil pollution traceability method and system based on multi-source spectrum identification, and belongs to the field of soil pollution traceability. The method comprises the following steps: S100, collecting original samples of at least two types of potential pollution sources and background soil samples of a research area; s200, constructing a structured spectrum fingerprint database; s300, taking the spectrum fingerprint database as a prior knowledge base and a training sample set, and training a classification and regression model by adopting a supervised learning algorithm; and S400, acquiring data of a hyperspectral reflection spectrum, a laser-induced breakdown spectrum and a three-dimensional fluorescence spectrum consistent with the data of the step S200 for a polluted soil sample to be traced under laboratory or in-situ conditions, inputting the data into the classification and regression model, and outputting the probability that the soil sample to be traced belongs to each potential pollution source category. According to the invention, rapid and efficient pollution traceability can be realized, so that the defects of the existing method in the aspects of timeliness, accuracy and quantification are overcome.

The invention discloses a hyperspectral detection method for the sugar degree and hardness of tomatoes. The method comprises the following steps: firstly, acquiring a hyperspectral reflection or transmission spectrum of tomatoes in a range of 400-1000nm, and carrying out pretreatment such as standard normal transformation, multivariate scatter correction, Savitzky-Golay smoothing and orthogonal signal correction on the original spectrum; then, characteristic wavelength screening is carried out on the preprocessed spectrum through a competitive adaptive reweighted sampling (CARS), an irreducible variable elimination (UVE) or a UVE-CARS joint algorithm, and a characteristic spectrum matrix is obtained; a Kennard-Stone (KS) algorithm is further adopted to divide a training sample into a correction set and a prediction set, a tomato sugar degree model and a tomato hardness model are constructed based on partial least squares regression (PLSR), and an optimal main factor number is determined through cross validation; and inputting the characteristic spectrum of the tomato to be detected into the model, and outputting the predicted values of sugar degree and hardness. The method has the advantages of no damage to fruits, high modeling stability, good prediction precision and the like, and is suitable for the fields of tomato quality evaluation, grading, postharvest treatment and the like.

The invention relates to a blueberry SSC content prediction method based on Caputo fractional derivative hyperspectral pretreatment, and belongs to the technical field of blueberry SSC content prediction. The method comprises the following steps: acquiring a blueberry sample, and carrying out hyperspectral imaging acquisition and SSC content determination; constructing a region of interest for a blueberry sample image, obtaining visible light-near infrared hyperspectral reflectivity data, performing Savitzky-Golay smoothing processing, constructing an overall sample, and performing descriptive statistical analysis on a training set and a test set; a Caputo fractional derivative algorithm is introduced to pre-process the spectral reflectivity of the whole sample, and a spectral feature screening algorithm based on SPA and CARS combination is adopted to carry out dimension reduction and feature optimization; constructing a regression prediction model of the SSC content of the blueberries by taking the obtained characteristic wave bands as model independent variables and the SSC values of the corresponding blueberries as dependent variables; and realizing blueberry SSC content prediction based on the regression prediction model. The objective of the invention is to solve the technical problem that spectral feature response wavebands related to SSC are difficult to accurately mine and screen in the prior art.

This application provides a rapid quality assessment method for recycled aggregates based on hyperspectral imaging. The method acquires hyperspectral reflectance data of the recycled aggregates, determines the target region of the aggregate to be tested based on the hyperspectral reflectance data, and extracts the geometric morphology parameters of the target region. It assesses the dispersion of reflectance energy distribution based on the reflectance sequence of each pixel in the characteristic band, determining the spectral disorder index. Based on the spectral disorder distribution of each pixel within a preset spatial neighborhood, it performs spatial correlation analysis to determine the mortar adhesion persistence index. Based on the mortar adhesion persistence index, spectral evolution rate, and geometric morphology parameters, it calculates a comprehensive quality index and assesses the quality. This method significantly improves the robustness of mortar identification in complex environments, corrects optical deviations in irregularly shaped aggregates, and enables real-time and accurate rapid quality assessment of recycled aggregates.

This invention belongs to the field of environmental monitoring technology and discloses a remote sensing inversion method for dissolved inorganic nitrogen and silicate in estuaries based on salinity synergy. The method includes the following steps: S1. During a field survey in the land-sea interaction zone of the estuary, remote sensing reflectance, salinity, and nutrient data are collected simultaneously. The measured hyperspectral reflectance is simulated as the equivalent reflectance of the satellite band using the spectral response function of the target satellite, which is used to construct the training and validation datasets for the model. S2. A nonlinear inversion model of remote sensing reflectance and salinity is established, and a nutrient mixture model of salinity and nutrients is constructed. The training dataset is used for model training. S3. The satellite remote sensing reflectance data of the estuarine area to be predicted is input into the trained nonlinear inversion model and combined with the nutrient mixture model to generate the spatial distribution of DIN concentration and DSi concentration in the estuarine area. This method achieves high-precision remote sensing inversion prediction of DIN concentration and dissolved silicate DSi concentration in the estuarine area.

The present application relates to the technical field of food quality detection, in particular to a fish freshness evaluation method and system, comprising: obtaining hyperspectral reflectance curve, surface texture microscopic image and volatile odor substance concentration spectrum of the fish to be measured, extracting protein degradation characteristic spectral absorption band, microstructure integrity multi-level entropy set and microbial metabolism indicative odor fingerprint. The three types of characteristics are input into an improved fusion evaluation algorithm, which adopts a weighted decision fusion framework, introduces a time decay factor dynamic weight distribution and a feature reliability adaptive correction mechanism, and outputs a dynamic freshness quantitative index and a shelf life stage label through feature-level fusion and time series evolution analysis. The present application can match the time sequence change of fish quality, weaken the interference of characteristic fluctuation, and improve the stability and accuracy of the evaluation result.

The invention provides a wetland vegetation LAI inversion method based on unmanned aerial vehicle multi-source remote sensing. The method comprises the following steps: collecting multi-source remote sensing data of a target area through an unmanned aerial vehicle; the multi-source remote sensing data comprises hyperspectral reflectivity data and point cloud data; performing feature extraction and screening on the multi-source remote sensing data to obtain an optimal feature subset; and performing inversion modeling and result comparison of the leaf area index through the optimal feature subset to obtain an optimal modeling scheme. According to the method, multi-source remote sensing characteristic variables are optimized and screened, and the LAI inversion model of the wetland vegetation is constructed by using a machine learning algorithm, so that the LAI estimation precision of the wetland vegetation is improved.

The invention presents a method for analyzing soil contamination by pollutants, particularly organic pollutants, by hyperspectral analysis of reflection and / or photoluminescence, characterized in that said analysis is carried out with a first piece of equipment by illuminating a sample with a light source and with at least one spectral sensor sensitive to a spectrum ranging from thermal infrared to ultraviolet. Abstract figure: 2

The invention relates to the technical field of soil content monitoring, in particular to a soil selenium content hyperspectral inversion method, which comprises the following steps: firstly, obtaining visible-near infrared hyperspectral reflectivity data and a selenium content measured value of a soil sample; original spectral data is subjected to standard normal variable transformation combined with second-order differential preprocessing, so that effective spectral signals are enhanced; then, screening out a key characteristic wave band subset from the preprocessed spectral data by adopting a competitive self-adaptive reweighted sampling algorithm; and by taking the feature subset as an input variable, constructing a support vector regression initial model, and introducing a myxomycete algorithm to globally optimize key parameters of the model so as to obtain an optimal inversion model. The method effectively solves the problems of weak spectral response and limited inversion precision of the soil selenium element, and realizes rapid and high-precision quantitative prediction and spatial mapping of the soil selenium content.