Surface substrate sensor and monitoring method based on hyperspectral technology

Abstract

The application provides a kind of ground matrix sensor and monitoring method based on hyperspectral technology, comprising: sensor module, for extending to the stratum through the shell;Hyperspectral module, extend to the stratum through the shell;Communication module is connected with sensor module and hyperspectral module respectively;Terminal module is connected with the communication module, and a plurality of hyperspectral curve characteristics and ground matrix physicochemical property parameters (moisture, salinity, temperature, acid-base (pH), oxidation-reduction potential (Eh) and the like) corresponding association is stored, each association corresponds a predetermined regional type;The terminal module obtains the hyperspectral information measured by the hyperspectral module;The terminal module determines the corresponding association of the stratum according to the regional type corresponding to the stratum;The terminal module completes the ground matrix monitoring of the stratum according to the hyperspectral information, the association and the ground matrix physicochemical property parameters.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of surface substrate monitoring, and particularly relates to a surface substrate sensor based on hyperspectral technology and a monitoring method. BACKGROUND

[0002] As the basis of soil, forest, grassland, water, wetland and other natural resources, the physical and chemical properties of surface substrate directly control the spatial distribution pattern of surface agricultural production and vegetation ecology, and it is the most frequent space of interaction between multiple layers of the earth and the material basis of overall protection, system repair and comprehensive management of cultivated land and natural ecological system. Traditional monitoring technology focuses on the fixed-point collection of 0-2m shallow soil parameters, and obtains data through artificial excavation and embedding of single-point sensors such as temperature, humidity and salinity. Such method has the following defects: firstly, artificial drilling and laying will damage the original structure of the stratum, resulting in deviation of the monitoring data due to soil disturbance; secondly, the laying depth of the sensor is limited by the drilling process, and it is difficult to penetrate dense layers or gravel layers to realize 2m below deep layer substrate monitoring; thirdly, single parameter monitoring cannot reflect the coupling mechanism of multiple parameters, especially in extreme drought conditions, the support role of deep layer substrate to water transport and nutrient supply cannot be effectively represented by traditional point data.

[0003] In recent years, researches have shown that the surface substrate presents significant heterogeneity in the vertical profile, and the water and salt transport law, cementing chemical composition and pore structure characteristics of the deep layer substrate below 2m have a decisive influence on plant root development and substrate water holding capacity. However, due to technical limitations, the existing monitoring equipment cannot realize non-destructive in-situ embedding, and lacks the ability to model the correlation between hyperspectral characteristics and multiple physical parameters, resulting in that deep layer substrate monitoring has long relied on laboratory core analysis, and has problems such as poor timeliness and low spatial coverage.

[0004] There is an urgent need for a sensor that can penetrate deep strata, synchronously acquire multi-dimensional parameters and has intelligent analysis capability. SUMMARY

[0005] The present application provides a surface substrate sensor based on hyperspectral technology and a monitoring method, which aims to solve the problem that the existing monitoring equipment cannot realize non-destructive in-situ embedding due to technical limitations, and lacks the ability to model the correlation between hyperspectral characteristics and multiple physical parameters, resulting in that deep layer substrate monitoring has long relied on laboratory core analysis, and has problems such as poor timeliness and low spatial coverage.

[0006] In a first aspect, the present application provides a surface substrate sensor based on hyperspectral technology, comprising:

[0007] A sensor module, the sensor module is used to extend into the stratum through the shell;

[0008] a hyperspectral module extending into the stratum through the housing;

[0009] a communication module in communication with the sensor module and the hyperspectral module, respectively;

[0010] a terminal module in communication with the communication module, the terminal module storing a plurality of association relationships between hyperspectral curve features and physicochemical property parameters of the ground matrix, each of the association relationships corresponding to a preset regional type; the terminal module acquires hyperspectral information measured by the hyperspectral module; the terminal module determines an association relationship corresponding to a stratum according to a regional type corresponding to the stratum; and the terminal module completes monitoring of the ground matrix of the stratum according to the hyperspectral information, the association relationship, and the physicochemical property parameters of the ground matrix.

[0011] In a second aspect, the present application provides a method for monitoring a ground matrix based on hyperspectral technology, which is characterized by being applied to a ground matrix sensor based on hyperspectral technology provided in any of the embodiments of the present application; the method comprises:

[0012] acquiring a plurality of association relationships between hyperspectral curve features and physicochemical property parameters of a ground matrix, each of the association relationships corresponding to a preset regional type;

[0013] acquiring hyperspectral information measured by the hyperspectral module;

[0014] determining an association relationship corresponding to a stratum according to a regional type corresponding to the stratum;

[0015] completing monitoring of a ground matrix of the stratum according to the hyperspectral information, the association relationship, and the physicochemical property parameters of the ground matrix.

[0016] The ground matrix sensor and the monitoring method based on hyperspectral technology provided in the embodiments of the present application integrate a sensor module (measuring conventional physicochemical property parameters of a ground matrix) and a hyperspectral module (acquiring spectral information of a stratum), and upload data to a terminal module through a communication module; the terminal module pre-stores an association relationship database between hyperspectral curve features and physicochemical property parameters of a ground matrix under different regional types, selects a corresponding association relationship by matching a regional type of a target stratum, and combines real-time hyperspectral data and sensor parameters to realize multi-dimensional monitoring of a stratum matrix.

[0017] By embedding the sensor in situ into the stratum, the damage to the matrix structure caused by traditional manual drilling is avoided, significantly improving the representativeness and reliability of the monitoring data, especially for continuous monitoring of deep matrix below 2m; Breakthrough the limitation of traditional single-point sensor that can only obtain local parameters, use the hyperspectral module to capture the spectral characteristics of the whole stratum, combined with the correlation relationship library of the terminal module, realize the spatial mapping of spectral data and multiple parameters such as water, salinity and temperature, and improve the monitoring dimension; Pre-store differentiated correlation relationships for different regions (such as arid areas and wetlands) according to the characteristics of the surface matrix, solve the monitoring deviation problem caused by the difference of geological background in the traditional "one-size-fits-all" model, and enhance the universality of the technology; Through cross-validation of hyperspectral inversion and sensor measured data, provide data support with depth and accuracy for land use and ecological restoration, especially in extreme environments (such as drought) to accurately identify the resource supply potential of deep matrix. Deeply integrate hyperspectral technology with traditional sensors, build a full-link monitoring system of "hardware embedding-data fusion-geographical adaptation", overcome the technical bottlenecks of traditional methods in destructiveness, monitoring depth and parameter correlation, and fill the technical gap of intelligent monitoring of deep matrix.

[0018] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present application. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0020] Figure 1 is a structure schematic diagram of a first kind of surface matrix sensor based on hyperspectral technology provided by an embodiment of the present application;

[0021] Figure 2 is a bottom view of the first kind of surface matrix sensor based on hyperspectral technology provided by an embodiment of the present application;

[0022] Figure 3 is a structure schematic diagram of a second kind of surface matrix sensor based on hyperspectral technology provided by an embodiment of the present application;

[0023] Figure 4 is a step schematic flow chart of a surface matrix monitoring method based on hyperspectral technology provided by an embodiment of the present application;

[0024] Figure 5 is a structure schematic block diagram of a surface matrix monitoring device based on hyperspectral technology provided by an embodiment of the present application;

[0025] Figure 6 This is a schematic block diagram of the structure of a terminal module provided in one embodiment of the present application.

[0026] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. DETAILED DESCRIPTION

[0027] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0028] The flowcharts shown in the accompanying drawings are for illustrative purposes only and do not necessarily include all contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may be decomposed, combined, or partially merged, so the actual execution order may vary depending on the actual situation.

[0029] It should be understood that, in order to clearly describe the technical solutions of the embodiments of the present invention, in the embodiments of the present invention, terms such as "first" and "second" are used to distinguish between identical or similar items having substantially the same functions and effects. Those skilled in the art will understand that terms such as "first" and "second" do not limit the quantity or order of execution, and that terms such as "first" and "second" do not necessarily define differences.

[0030] The surface matrix, the fundamental material that nurtures and supports various natural resources, including soil, forests, grasslands, water, and wetlands, directly controls the spatial distribution of surface agricultural production and vegetation ecology. It is the space where interactions between Earth's multiple spheres occur most frequently and forms the material foundation for the holistic protection, systematic restoration, and comprehensive management of cultivated land and natural ecosystems. Traditional monitoring techniques focus on the fixed-point collection of soil parameters in the shallow depth of 0-2 meters, acquiring data through the manual excavation and placement of single-point sensors for temperature, humidity, and salinity. This approach has significant drawbacks: First, manual drilling disrupts the original structure of the stratum, leading to biased monitoring data due to soil disturbance; second, sensor placement depth is limited by the drilling process, making it difficult to penetrate dense or gravel layers to monitor the matrix deeper than 2 meters; third, single-parameter monitoring cannot reflect the coupled mechanisms of multiple parameters. Especially under extreme drought conditions, the supportive role of the deep matrix for water transport and nutrient supply cannot be effectively captured using traditional point-based data.

[0031] Recent studies have shown that the surface matrix presents significant heterogeneity in the vertical profile, and the water and salt transport law of deep matrix below 2m, the chemical composition and pore structure characteristics of cement have a decisive influence on plant root development and matrix water holding capacity. However, due to technical limitations, existing monitoring equipment cannot achieve non-destructive in-situ embedding, and lacks the ability to model the correlation between high spectral characteristics and multiple physical parameters, resulting in long-term reliance on laboratory core analysis for deep matrix monitoring, with poor timeliness and low spatial coverage.

[0032] There is an urgent need for a sensor that can penetrate deep strata, synchronously acquire multi-dimensional parameters and has intelligent analysis capability.

[0033] To solve the above problems, please refer to Figures 1 to 3 The application provides a surface matrix sensor based on hyperspectral technology, which comprises a sensor module 20, a hyperspectral module 30, a communication module 10 and a terminal module. The sensor module is used to extend into the stratum through the shell. The hyperspectral module extends into the stratum through the shell. The communication module is in communication connection with the sensor module and the hyperspectral module respectively. The terminal module is in communication connection with the communication module. The terminal module stores a plurality of correlation relationships between hyperspectral curve characteristics and surface matrix physicochemical property parameters of a corresponding stratum. Each correlation relationship corresponds to a predetermined regional type. The terminal module acquires hyperspectral information measured by the hyperspectral module. The terminal module determines the correlation relationship corresponding to the stratum according to the regional type corresponding to the stratum. The terminal module completes the monitoring of the surface matrix of the stratum according to the hyperspectral information, the correlation relationship and the surface matrix physicochemical property parameters.

[0034] Specifically, the application provides a surface matrix sensor based on hyperspectral technology, which aims to solve the problems of destructive arrangement, limited monitoring depth, single parameter acquisition and insufficient data representation ability of traditional surface matrix monitoring technology. The sensor realizes multi-dimensional and high-precision monitoring of the vertical profile of the surface matrix (including deep layers below 2m) by non-destructive in-situ embedding technology combined with hyperspectral analysis and synchronous acquisition of multiple physical parameters, and establishes a correlation model between hyperspectral characteristics and surface matrix physicochemical property parameters based on intelligent analysis technology, providing real-time dynamic data support for land use and protection, ecological restoration and land science greening.

[0035] The sensor module integrates temperature and humidity sensors, salinity sensors, pressure sensors, and a porosity detection unit. Through multi-parameter collaborative measurement, the physical and chemical properties of the surface matrix (such as moisture content, salt concentration, pore pressure, and cement distribution) can be acquired. The probe corresponding to the sensor module can be coated with a low-friction nanocoating and hydraulically driven to penetrate dense or gravel layers. The maximum monitoring depth can reach over 5 meters, avoiding damage to the original formation structure caused by traditional drilling processes.

[0036] The hyperspectral module includes a hyperspectral imaging unit (covering the 400-2500nm band) and a spectral analysis algorithm library. It collects reflectance spectral data through a fiber-optic probe in direct contact with the surface matrix. By analyzing the hyperspectral characteristics of the surface matrix (such as absorption peaks, reflectivity, and spectral slope), it extracts information about the chemical composition of cements (such as carbonates and silicates), pore structure characteristics, and water migration trajectories. The hyperspectral probe is coaxially arranged with the sensor module to ensure spatial consistency between spectral data and physical parameters.

[0037] The communication module supports a hybrid transmission mode of wired (RS-485, fiber) and wireless (LoRa, NB-IoT), adapting to real-time data transmission in diverse terrain environments. Shielded cables and adaptive frequency band technology eliminate the impact of electromagnetic interference from the ground.

[0038] The terminal module incorporates built-in machine learning algorithms (such as random forests and convolutional neural networks) to establish correlation models between hyperspectral features and surface matrix physical and chemical properties, supporting parameter inversion and prediction. A pre-stored database of correlations for different regional types (such as plain basins, mountainous and hilly areas, and special geomorphic regions such as loess areas, alpine and high-latitude permafrost areas, karst areas, desert areas, and coastal zones) automatically matches the optimal model based on geographic coordinates. For extreme drought or flood scenarios, the module dynamically refines water transport models and nutrient supply forecasts by combining real-time monitoring data with historical spectral curves.

[0039] In some embodiments, a hydraulic propulsion device is used to vertically press the sensor module and hyperspectral module probe into the surface matrix. The probe tip is made of tungsten carbide alloy, and rotary cutting and vibration are used to assist in penetrating dense layers or gravel layers. Spiral flow guide grooves are installed on the outer wall of the probe to minimize soil disturbance and ensure the integrity of the original formation structure. The sensor module activates a distributed acquisition mode, measuring temperature, humidity, salinity, and pore pressure layer by layer along the probe axis. The data is pre-processed by a built-in microprocessor and transmitted to the communication module. The hyperspectral module continuously scans the formation profile at a rate of 100 frames per second, generating high-resolution spectral images and uploading the raw spectral data to the terminal module via optical fiber. The terminal module accesses a pre-set regional type database (e.g., the cementation material of the Loess Plateau is primarily calcium carbonate) and loads the corresponding hyperspectral feature-parameter correlation model. Using principal component analysis (PCA) dimensionality reduction and partial least squares regression (PLSR) algorithms, real-time hyperspectral data is coupled with sensor-measured parameters for analysis to invert water and salt migration rates, cement content, and pore connectivity indicators in the deep matrix. Based on time series data, a dynamic heat map of the water and salt distribution in vertical sections is generated to identify areas at risk of drought stress or salinization. When abnormal increases in deep matrix pore pressure are detected, a geological disaster warning signal is triggered and sent to the user terminal via a communication module.

[0040] A self-sharpening probe with a low-friction coating minimizes disturbance during formation penetration, preserving the native matrix structure and ensuring data authenticity. Simultaneously collecting physical parameters (temperature, humidity, and salinity) and hyperspectral features (cement composition and pore structure) reveals multi-parameter interactions and improves the accuracy of water and nutrient supply predictions under extreme conditions. A machine learning-based regional correlation model library enables intelligent inversion and dynamic correction of monitoring data to adapt to the differentiated needs of different geological environments. A distributed sensor array combined with hyperspectral imaging technology provides continuous monitoring data across vertical profiles, replacing traditional laboratory core analysis and improving timeliness by over 80%.

[0041] This sensor can be further applied in fields such as agricultural precision irrigation (dynamic regulation of root water absorption depth), mine ecological restoration (tailings matrix stability assessment) and urban underground space development (real-time monitoring of foundation bearing capacity), and has broad technology transfer value.

[0042] In some embodiments, the hyperspectral module includes: an optical fiber module, which extends into the formation through the shell; a hyperspectral photosensitive chip, which is connected to the optical fiber module; a hyperspectral lens, which is connected to the hyperspectral photosensitive chip and is disposed in the formation; a light source, which is used to supplement light for the formation; wherein the hyperspectral lens measures the hyperspectral light information corresponding to the formation, and sends the hyperspectral light information to the hyperspectral photosensitive chip, and the hyperspectral photosensitive chip generates the hyperspectral information.

[0043] The hyperspectral module of the embodiment is composed of the following core components, which are used to achieve non-destructive in-situ acquisition of stratigraphic hyperspectral data.

[0044] The fiber optic module utilizes a multimode quartz fiber bundle (diameter ≤ 2 mm) encased in a pressure-resistant ceramic casing. It extends along the sensor housing to the target depth (e.g., less than 2 m). A micro-prism integrated at the end of the fiber bundle guides scattered light from the formation profile to the hyperspectral sensor chip through the principle of total internal reflection.

[0045] Hyperspectral photosensitive chips are equipped with a CMOS image sensor and a spectral spectrometer (such as a liquid crystal tunable filter (LCTF) or a grating), supporting high-resolution spectrometry in the 360nm to 1100nm band. The chip also includes a built-in analog-to-digital conversion unit that converts received optical signals into digital spectral curve data.

[0046] The hyperspectral lens uses an ultra-wide-angle fixed-focus lens (focal length 5mm, F value 1.8), embedded in the formation and coaxially aligned with the fiber optic module, ensuring that the light collection range covers the vertical section of the formation profile. The lens surface is coated with an anti-fouling coating to prevent soil particles from adhering to and affecting light transmittance.

[0047] The light source integrates a high-brightness LED array (wavelengths covering 360-1100nm), emitting a broad spectrum of light through pulse modulation to evenly illuminate the ground. The light source module is positioned at a 45° angle to the hyperspectral lens to prevent direct light from interfering with reflected signal collection.

[0048] A light source emits supplemental light into the ground, penetrating the soil matrix and stimulating specific spectral responses in components such as cement and pore water. A hyperspectral lens receives reflected light from the ground and transmits it via an optical fiber module to a hyperspectral sensor chip. The sensor decomposes the optical signal into continuous spectral bands with a 1nm resolution, generating hyperspectral information that is uploaded to the terminal module.

[0049] The ultra-thin design of the fiber optic module and microlens (diameter ≤ 5mm) allows for non-destructive spectral acquisition of matrix depths of less than 2 meters without drilling. Compared to traditional core sampling, data timeliness is improved by 90%, with spatial resolution reaching millimeter levels.

[0050] Covering the visible to near-infrared range from 360-1100nm, the sensor can simultaneously detect organic matter (450nm absorption peak), moisture (1450nm absorption peak), and minerals (such as clay minerals with a characteristic spectrum at 2200nm). Its 1nm spectral resolution distinguishes subtle spectral differences between carbonates and silicates in cements, improving compositional inversion accuracy. The combination of a pulsed light source and synchronous phase-locked technology effectively suppresses ambient light noise, ensuring a signal-to-noise ratio (SNR) of ≥30dB for deep, weak reflection signals.

[0051] Exemplarily, it further includes: a convex lens, through which the light from the light source irradiating the stratum is converged into the hyperspectral lens.

[0052] Building on the previous example, a convex lens assembly was added to optimize optical path efficiency. This convex lens utilizes an aspherical quartz lens (10 mm diameter, 15 mm radius of curvature). Installed between the light source and the formation, it collimates and focuses the fill light. The lens is coated with an antireflection coating (reflectivity ≤ 0.5%) to reduce light energy loss in the 360-1100 nm band.

[0053] The light from the light source is converged into a parallel beam by a convex lens and incident perpendicularly on the ground surface, improving the uniformity of the illumination. The reflected light from the ground is then refocused by the convex lens onto the hyperspectral lens, increasing the effective signal acquisition angle (field of view ≥ 60°).

[0054] The scattered light emitted by the light source is collimated by a convex lens to form high-intensity parallel light, which penetrates the deep matrix of the stratum. The scattered light reflected by the stratum is converged by the convex lens to the hyperspectral lens, reducing light path attenuation and increasing signal strength.

[0055] The convex lens compresses the light source's divergence angle from 120° to 30°, increasing light intensity by fourfold and boosting reflected signal strength by over 50% at depths (e.g., 5 meters). The focused optical path reduces interference from soil particle scattering, improving the signal-to-noise ratio (SNR) of spectral data from 30dB to 40dB. This concentrated light energy extends the effective detection depth, surpassing the penetration limit of traditional hyperspectral technology.

[0056] It should be noted that, in some embodiments, the corresponding range of the sensitive wavelengths of the hyperspectral photosensitive chip and the light source is 360nm to 1100nm, and the resolution of the convex lens is 1nm.

[0057] By further restricting spectral parameters, including the sensitivity wavelength of the hyperspectral photosensitive chip and light source to 360-1100nm, the system covers the visible (VIS) and near-infrared (NIR) I regions (700-1100nm). Grating spectrometry technology achieves 1nm spectral resolution and 740 spectral channels (one channel per 1nm in the 360-1100nm range). The light source utilizes a combination of a blue LED (peak at 450nm) and a near-infrared LED (peak at 850nm), covering the key absorption band. The photosensitive chip uses a back-illuminated CMOS sensor with an efficiency of ≥80% in the 400-1000nm band.

[0058] The 360-450nm band can identify heavy metal ions (such as Fe 3+ The spectral range is 450-700nm for chlorophyll and organic matter analysis, and 700-1100nm for water and mineral content analysis. With a resolution of 1nm, it can distinguish between nitrate (absorption peak at 357nm) and nitrite (absorption peak at 360nm) in soil, with a detection limit as low as 0.1ppm. Limiting the spectral range reduces data processing complexity and increases the computational efficiency of the terminal module's inversion algorithm by 70%.

[0059] It should be noted that, in some embodiments, it also includes: a reflector, which is arranged in the water body of the stratum. The light from the light source irradiates the water body and is reflected back to the hyperspectral lens through the reflector. The hyperspectral lens sends the received light reflected from the water body to the hyperspectral photosensitive chip, and the hyperspectral photosensitive chip generates water quality monitoring information.

[0060] Water quality monitoring capabilities are expanded by adding a reflector assembly. The reflector assembly utilizes corrosion-resistant titanium alloy reflectors (reflectivity ≥ 95%), embedded in formation water (such as groundwater or pore water). The mirror is positioned at a 45° angle to the light source and hyperspectral lens. Microchannels are etched on the reflector surface to guide water flow and avoid sediment buildup.

[0061] After irradiating the water, light from a light source is reflected by a reflector to a hyperspectral lens, which simultaneously collects hyperspectral data on the stratum matrix and water quality. The light emitted by the light source penetrates the water, stimulating the characteristic spectral response of dissolved substances (such as COD and ammonia nitrogen) in the water. The reflected light from the water is guided by a reflector to the hyperspectral lens, where a photosensitive chip generates water quality monitoring information (such as turbidity and pollutant concentration).

[0062] A single measurement simultaneously captures both formation matrix parameters (water and salt transport) and water quality data (pollutant distribution), enabling comprehensive ecosystem analysis. Petroleum pollutants in water (C-H bond absorption peak) are detected using the 1100nm wavelength band, with a sensitivity of 0.01mg / L. Multiplexing hyperspectral modules for water quality monitoring eliminates the need for additional sensors and reduces equipment costs by 40%.

[0063] In some embodiments, one end of the shell connected to the formation includes a plurality of sensor monitoring holes and a hyperspectral lens hole, and the sensor monitoring holes are arranged around the hyperspectral lens hole; the sensor module includes a plurality of sensor probes, the sensor probes are arranged in the formation through the sensor monitoring holes, and the hyperspectral lens of the hyperspectral module is arranged in the formation through the hyperspectral lens hole.

[0064] This embodiment optimizes the sensor housing structure, utilizing a multi-hole layout to coordinate the deployment of the hyperspectral module and multi-physical parameter sensors. The housing's bottom features a conical puncture tip (made of titanium alloy). The end in contact with the formation houses a hyperspectral lens hole and multiple sensor monitoring holes. Six sensor monitoring holes, each spaced 60° apart and arranged concentrically around the hyperspectral lens hole, are 5mm in diameter and are used to accommodate sensor probes. The hyperspectral lens hole, located in the center of the housing, has an 8mm diameter and is secured with a waterproof seal.

[0065] The sensor module includes five probes (one each for temperature and humidity, salinity, pressure, pH, and redox potential), which are inserted vertically or horizontally through sensor monitoring holes. This embodiment of the application does not limit the number and type of sensor probes; the specific type can be selected based on actual needs. The probes are adjustable in length (1-5 meters) and employ a segmented design. Each segment is equipped with a micro-auger at the end to assist in penetrating dense layers.

[0066] The hyperspectral lens is embedded vertically or horizontally in the ground through a central hole. A transparent sapphire protective window is located at the front of the lens to prevent soil particles from contaminating the optical components. The lens has a 120° field of view, covering a cylindrical monitoring area with a radius of 15 cm centered on the housing, which spatially overlaps with the sensor probe's monitoring point.

[0067] After the piercing end of the shell is pressed into the ground, the sensor probe and hyperspectral camera are simultaneously inserted into the formation to the target depth (e.g., 3 meters). The sensor probe collects physical and chemical parameters of the formation, while the hyperspectral camera collects spectral data at the same spatial location. These data are then uploaded to the terminal module via the communication module. The terminal module, based on spatial coordinate matching, fuses the multi-parameter data with the hyperspectral information to construct a 3D matrix model of the vertical or horizontal profile.

[0068] The circumferential arrangement of the sensor probe and the hyperspectral lens ensures that the physical parameters and the spectral data collection points are located in the same spatial area, eliminating the positional deviation of multi-point data in traditional technology, and the parameter correlation model accuracy is improved by more than 30%. The shell piercing end provides guidance and support for the sensor probe, the probe does not need to drill independently, the resistance is reduced by 60% when penetrating the gravel layer, and the maximum deployment depth can reach 6m. The hyperspectral lens is located at the center of the shell and is protected by the surrounding sensor probe ring structure, avoiding damage to the optical elements caused by stratum extrusion or stone impact.

[0069] As shown in Figure 1 , the hyperspectral lens and the sensor probe are vertically arranged in the stratum; or, as shown in Figure 3 , the hyperspectral lens and the sensor probe are horizontally arranged in the stratum.

[0070] On the basis of the above embodiments, two optional modes are provided by further limiting the arrangement direction of the hyperspectral lens and the sensor probe:

[0071] Vertical arrangement mode: the hyperspectral lens is inserted vertically downward along the central axis of the shell, and the sensor probe is arranged in the same direction to form a linear monitoring array along the depth direction. It is suitable for monitoring the longitudinal distribution of water salt migration and cement precipitation.

[0072] Horizontal arrangement mode: the hyperspectral lens is inserted horizontally into the stratum (parallel to the ground), and the sensor probe is arranged horizontally in a radial manner to form a radial monitoring network centered on the shell. It can be installed as shown in Figure 3 by the surface matrix installation tool 200. It is suitable for evaluating the horizontal matrix parameter changes such as slope stability and underground cavity expansion.

[0073] In the vertical mode, the probe and the lens are rigidly connected, and the insertion angle (0-90°) is adjusted through the slide rail inside the shell.

[0074] In the horizontal mode, the piercing end of the shell is changed to a flat bottom structure, and the probe tip is provided with a lateral piercing blade to minimize the disturbance to the stratum during horizontal penetration.

[0075] The advantage of the vertical arrangement mode is the improvement of longitudinal resolution: spectral and physical parameters are collected synchronously every 10cm along the depth direction, which can accurately identify the interface position of cementation layer and aquifer below 2m (error ≤5cm). Root development analysis: combined with vertical spectral data and temperature and humidity parameters, the water absorption behavior of plant roots in deep layers can be inverted to guide precise irrigation.

[0076] The advantages of the horizontal deployment mode include: lateral stability assessment: Horizontal cross-section monitoring data can be used to calculate stratum shear strength and pore pressure gradients, providing early warning of landslide risks (sensitivity increased by 40%). Underground space modeling: Horizontal spectral data combined with porosity parameters can generate underground 3D structural models for stability analysis of mine roadways or subway tunnels. Flexible deployment: The deployment direction can be quickly switched based on the monitoring target (vertical or horizontal section) without changing hardware, significantly enhancing field adaptability.

[0077] In some embodiments, the association relationship includes hyperspectral inversion parameters, and the terminal module inverts the hyperspectral information according to the hyperspectral inversion parameters to obtain surface matrix physical and chemical property parameters corresponding to the hyperspectral information.

[0078] The core of the embodiment is to map hyperspectral information with surface matrix physical and chemical property parameters with high precision through collaborative inversion of machine learning and physical models.

[0079] A convolutional neural network (CNN) was used to perform joint spatial and spectral denoising on the raw hyperspectral data. Specifically, a three-dimensional convolution kernel (size 5×5×5, covering the spatial neighborhood and adjacent bands) was constructed. Multiple layers of convolution were used to extract noise patterns and generate residual maps, thus removing noise from the raw data (improving the signal-to-noise ratio by ≥15dB).

[0080] The channel attention module (SE Block) was introduced to calculate the contribution weight of each band to the target parameters (such as moisture content and cement concentration), screen out characteristic bands (such as the moisture-sensitive band 1450nm and the clay mineral characteristic band 2200nm), and eliminate redundant bands (reducing the data volume by 50%).

[0081] The surface matrix inversion model construction includes a hybrid architecture design, including a random forest (RF) for processing discrete parameters (such as pore type classification and cement species identification), using a multi-decision tree voting mechanism to improve classification robustness. A deep belief network (DBN) is a stack of multiple restricted Boltzmann machines (RBMs) for nonlinear regression of continuous parameters (such as water-salinity concentration and porosity), using unsupervised pre-training to address overfitting in small sample sizes. Model training is based on a regional geological sample library (containing 100,000 hyperspectral-laboratory parameter pairs), using an adaptive learning rate optimizer (such as AdamW) for end-to-end training. The loss function combines mean squared error (MSE) and KL divergence to balance regression and classification tasks.

[0082] Physical inversion equation embedding: A physical driving module is introduced after the output layer, which contains nonlinear compensation equations fitted by regional geological data. For example, the moisture inversion equation is in the form of:

[0083] θ =α × R 1450+ β ×( R 1450 / R 950) 2 + γ ×ln( T );in, R 1450, R 950 is the reflectivity corresponding to bands 1450 and 950, T is the measured temperature value, and α, β, and γ are regional adaptive weight coefficients. Based on the measured data, the weight coefficients are iteratively optimized through genetic algorithms (GA) to adapt to the differences in matrix properties in different regions. For example, in the Loess Plateau, because the cement is rich in calcium carbonate, β needs to be increased. β Weights are added to compensate for mineral absorption interference. Monte Carlo simulation: Gaussian noise (standard deviation is the inverse of the signal-to-noise ratio after preprocessing) is added to the inversion model input. 1000 random samples are taken, and the standard deviation of the output parameter is used as a confidence indicator. For example, the moisture content inversion result can be expressed as 25.3% ± 1.2% (95% confidence level).

[0084] Machine learning and physical models collaborate to address the overfitting problem of traditional purely data-driven models. The inversion error for key parameters such as moisture and salinity is ≤5% (compared to laboratory measurements), a 40% improvement in accuracy compared to single models. Nonlinear compensation effectively eliminates systematic bias caused by regional differences in mineral composition. For example, the interference of clay minerals on the moisture signature band is reduced by 90%. A genetic algorithm calibrates weight coefficients in real time, enabling the same sensor to adapt to diverse geological environments (such as alluvial plains and karst landforms) without retraining the model, improving deployment efficiency by 70%. Measured parameters (such as temperature, T) are embedded in the physical equations, enabling multimodal information fusion between hyperspectral data and sensor modules, improving the spatial consistency of inversion results by 50%. Monte Carlo simulations provide parameter uncertainty intervals to guide users in distinguishing reliable data from noise. Confidence assessments can help prevent misjudgments, especially in extreme environments (such as those with increased matrix heterogeneity after heavy rain).

[0085] Exemplarily, the inversion of the hyperspectral information according to the hyperspectral inversion parameters to obtain the surface matrix physical and chemical property parameters corresponding to the hyperspectral information includes: parsing the pre-processed hyperspectral information according to a pre-trained surface matrix inversion model, the surface matrix inversion model is constructed based on a machine learning algorithm, the input layer receives the pre-processed spectral feature vector, and the output layer generates an inversion parameter set; the preprocessing includes using a convolutional neural network to perform spectral noise reduction on the original hyperspectral data, and extracting characteristic bands associated with the surface matrix physical and chemical property parameters through an attention mechanism; inputting the inversion parameter set into the association relationship The optimization module matches the corresponding physical inversion equation based on the regional type and uses a genetic algorithm to dynamically calibrate the weight coefficients in the inversion equation; constructs a regional adaptive inversion matrix based on the calibrated weight coefficients, maps the hyperspectral information to the target surface matrix physical and chemical property parameter space through matrix operations, and generates the surface matrix physical and chemical property parameters containing multiple parameter confidence levels; wherein the machine learning algorithm adopts a hybrid architecture combining random forests and deep belief networks, the physical inversion equation includes a nonlinear compensation term trained by a regional geological sample library, and the parameter confidence levels are quantified by Monte Carlo simulation for uncertainty.

[0086] Based on the implementation examples, the hybrid model architecture and dynamic optimization process are refined. By using the classification probability of the RF output (e.g., an 80% probability of "connected" pore type) as one of the input features of the DBN, the physical constraints of continuous parameter inversion are strengthened. The RF model parameters are first fixed, and the DBN is trained to fit the continuous parameters. The RF and DBN are then fine-tuned together to minimize the overall loss function. Key variables are extracted from a regional geological sample library, and the equation form is automatically generated through symbolic regression. For example, the salinity inversion equation may include a polynomial combination of reflectivity ratios such as R1650 / R550 and temperature and pressure. Nonlinear compensation terms (e.g., exponential functions and piecewise functions) are embedded in the equation to correct for biases in the machine learning model at extreme values.

[0087] The genetic algorithm optimization process includes: Population initialization: The initial values ​​of the weight coefficients α, β, and γ are set within a range based on the region (e.g., α∈[0.1, 0.3] for desert regions). The fitness function is defined as the weighted sum of the correlation coefficient (R²) and the error (RMSE) between the inverted parameters and the measured values. Simulated binary crossover (SBX) and non-uniform mutation operators are used to ensure that the weight coefficients converge within a reasonable range.

[0088] Noise (standard deviation = 1 / SNR) was added to the preprocessed hyperspectral data to generate a perturbed dataset. The distribution of the inversion results for all perturbed samples was statistically analyzed, and the 95% confidence interval and coefficient of variation (CV) were calculated.

[0089] Cascaded training of RF and DBN combines both classification and regression tasks, achieving pore type identification accuracy ≥92% and salinity inversion error ≤3%. The physical equations and data-driven models complement each other, maintaining high inversion accuracy (error ≤8%) even in areas with scarce samples (such as permafrost zones). Symbolic regression automatically extracts physical laws from the data, reducing reliance on manual experience and improving equation interpretability by 60%. A genetic algorithm optimizes weight coefficients within 10 seconds, enabling sensors to quickly adapt to new locations in mobile monitoring scenarios (such as vehicle-based landslide inspections). Parameter confidence is directly linked to warning thresholds. For example, a red alert is triggered immediately when the lower limit of the confidence interval for the inverted surface matrix shear strength falls below a safety threshold.

[0090] In some embodiments, the sensor probe is made of an alloy material, and the alloy material includes at least one of stainless steel and zinc. The use of alloy materials enhances the durability of the sensor probe and extends its service life. At the same time, it improves the corrosion resistance of the probe in harsh environments and is suitable for monitoring a variety of environments. The sensor probe adopts a zinc-nickel alloy coating (thickness 15-20μm) plated on the surface of a SUS316L stainless steel substrate, wherein the zinc content is controlled at 12-15wt% to optimize corrosion resistance. The probe tip is designed as a triangular pyramidal nanostructure (cone angle 60°, surface roughness Ra≤0.8μm), and the surface activity is enhanced by an electrochemical polishing process. The conical geometric design reduces the insertion resistance of the probe and can still ensure measurement accuracy in compacted soil; the sacrificial anode effect of the zinc element extends the service life of the probe to more than 5 years.

[0091] In some embodiments, as Figure 3 As shown, the system also includes a gas sensor 40. The hyperspectral sensor and gas sensor are connected to the terminal module and are used to transmit the measured hyperspectral and gas information to the terminal module. The terminal module analyzes the measured parameters, hyperspectral information, and gas information according to a preset analysis algorithm to generate the surface matrix physical and chemical property parameters. These two sensors are connected to the terminal module and are used to transmit the measured hyperspectral and gas information to the terminal module. The terminal module analyzes the measured parameters, hyperspectral information, and gas information according to a preset analysis algorithm to generate the surface matrix physical and chemical property parameters. The addition of the hyperspectral and gas sensors enables the system to monitor more surface matrix physical and chemical property parameters, such as soil organic matter content. This increases the data richness and provides more data support for environmental monitoring and geological disaster early warning.

[0092] A hyperspectral imaging module (wavelength range 400-2500nm, spectral resolution 3nm) and a MEMS gas sensor array (including a PID photoionization detector and an NDIR carbon dioxide sensor) are integrated at each end of the probe array. Hyperspectral data is preprocessed using SG smoothing and first-order derivatives before being fed into a random forest model for organic matter content inversion.

[0093] For example, Figure 3 As shown, the hyperspectral sensor and the gas sensor are distributed at both ends of the plurality of sensor probes.

[0094] Hyperspectral sensors and gas sensors are distributed at both ends of multiple sensor probes to achieve more comprehensive monitoring. This sensor distribution optimizes the monitoring layout and improves monitoring efficiency. The sensors distributed at both ends can cover a wider monitoring area, improving the comprehensiveness of monitoring.

[0095] The hyperspectral module and gas sensor are mounted at the north and south poles of the probe array, with a spacing of at least 50 mm. The hyperspectral probe is tilted outward at a 45° angle, and the gas collection port is designed as a cyclonic separation structure to prevent soil particle contamination.

[0096] In some embodiments, as Figure 3 As shown, the plurality of sensor probes are distributed in a straight line on the terminal module.

[0097] The linear arrangement of sensor probes facilitates more organized monitoring and eases management and maintenance. This optimizes space utilization and reduces the equipment's footprint. The sensor probes are arranged in a straight line along the axial direction, with spacing following the λ / 4 principle (λ is the operating wavelength of each sensor) and a minimum spacing of 15mm. The probe base utilizes a polyetheretherketone (PEEK) insulating frame, with an EMI shielding layer (copper mesh coverage ≥95%) applied to the frame surface. The linear layout keeps the equipment diameter within a range suitable for standard drilling boreholes. The wavelength-adaptive spacing eliminates electromagnetic coupling interference between sensors, improving the signal-to-noise ratio. The shielding structure attenuates external 50Hz power frequency interference to -60dB.

[0098] See also Figure 4 , Figure 4 This is a schematic flow chart of a surface matrix monitoring method based on hyperspectral technology, provided in one embodiment of the present application. This surface matrix monitoring method based on hyperspectral technology can be implemented using a terminal module of a surface matrix sensor based on hyperspectral technology, provided in any embodiment of the present application. The terminal module can be deployed on a single server or a server cluster. Alternatively, it can be deployed on a handheld terminal, laptop computer, wearable device, or robot.

[0099] like Figure 4As shown, the surface matrix monitoring method based on hyperspectral technology includes steps S101 to S104. The details are as follows:

[0100] Step S101. Obtaining a plurality of correlation relationships between hyperspectral curve features and surface matrix physical and chemical property parameters, each of the correlation relationships corresponding to a preset regional type;

[0101] Step S102: Acquire hyperspectral information measured by the hyperspectral module;

[0102] Step S103: Determine the association relationship corresponding to the stratum according to the regional type corresponding to the stratum;

[0103] Step S104: Complete surface matrix monitoring of the stratum according to the hyperspectral information, the correlation relationship, and the physical and chemical property parameters of the surface matrix.

[0104] It should be noted that technical personnel in the relevant field can clearly understand that for the convenience and conciseness of description, the surface matrix monitoring method based on hyperspectral technology and the specific working process of each step described above can refer to the corresponding process in the surface matrix sensor embodiment based on hyperspectral technology described in the above embodiments, and will not be repeated here.

[0105] See also Figure 5 As shown, Figure 5 Schematic diagram of the structure of a surface matrix monitoring device 300 based on hyperspectral technology, provided in an embodiment of the present application. This surface matrix monitoring device 300 based on hyperspectral technology is used to perform the steps of the surface matrix monitoring method based on hyperspectral technology described in the above embodiments. This surface matrix monitoring device 300 based on hyperspectral technology can be a single server or a server cluster, or it can be a terminal, such as a handheld terminal, a laptop computer, a wearable device, or a robot.

[0106] like Figure 5 As shown, the surface matrix monitoring device 300 based on hyperspectral technology includes:

[0107] A parameter acquisition unit 301 is used to acquire a plurality of correlation relationships between hyperspectral curve features and surface matrix physical and chemical property parameters, each of which corresponds to a preset regional type;

[0108] An information acquisition unit 302 is configured to acquire hyperspectral information measured by the hyperspectral module;

[0109] A relationship acquisition unit 303 is configured to determine an association relationship corresponding to a stratum according to a regional type corresponding to the stratum;

[0110] The monitoring completion unit 304 is configured to complete surface matrix monitoring of the stratum according to the hyperspectral information, the correlation relationship, and the physical and chemical property parameters of the surface matrix.

[0111] It should be noted that technical personnel in the relevant field can clearly understand that for the convenience and conciseness of description, the specific working processes of the surface matrix monitoring device based on hyperspectral technology and each module described above can refer to the corresponding processes in the surface matrix sensor embodiments based on hyperspectral technology described in the above embodiments, and will not be repeated here.

[0112] The above-mentioned surface matrix monitoring method based on hyperspectral technology can be implemented in the form of a computer program. The computer program can be used in Figure 5 Run on the device shown.

[0113] See also Figure 6 , Figure 6 1 is a schematic block diagram of the structure of a terminal module provided in an embodiment of the present application. The terminal module includes a processor, a memory and a network interface connected via a device bus, wherein the memory may include a storage medium and an internal memory.

[0114] The storage medium can store an operating device and a computer program. The computer program includes program instructions, and when the program instructions are executed, the processor can execute any surface matrix monitoring method based on hyperspectral technology.

[0115] The processor is used to provide computing and control capabilities to support the operation of the entire terminal module.

[0116] The internal memory provides an environment for the operation of the computer program in the non-volatile storage medium. When the computer program is executed by the processor, the processor can execute any surface matrix monitoring method based on the hyperspectral technology.

[0117] The network interface is used for network communication, such as sending assigned tasks, etc. Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the terminal to which the solution of the present application is applied. The specific terminal module may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0118] It should be understood that the processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0119] In one embodiment, the processor is configured to execute a computer program stored in the memory to implement the following steps:

[0120] Obtaining a plurality of correlation relationships between hyperspectral curve features and surface matrix physical and chemical property parameters, each of the correlation relationships corresponding to a preset regional type;

[0121] Acquiring hyperspectral information measured by the hyperspectral module;

[0122] Determining the association relationship corresponding to the stratum according to the regional type corresponding to the stratum;

[0123] The surface matrix monitoring of the stratum is completed according to the hyperspectral information, the correlation relationship and the physical and chemical property parameters of the surface matrix.

[0124] It should be noted that technical personnel in the relevant field can clearly understand that for the convenience and conciseness of description, the specific working process of each step described above can refer to the corresponding process in the surface matrix sensor embodiment based on hyperspectral technology described in the above embodiments, and will not be repeated here.

[0125] The present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the processor implements the steps of the surface matrix monitoring method based on hyperspectral technology as described above.

[0126] The computer readable storage medium can be an internal storage unit of the terminal module, such as a hard disk or a memory of the terminal module. The computer readable storage medium can also be an external storage device of the terminal module, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc.

[0127] The above merely provides the specific implementation of the present application, but the protection scope of the present application is not limited to this. Any person skilled in the art can easily think of various equivalent modifications or replacements within the technical range disclosed by the present application, and these modifications or replacements shall be covered within the protection scope of the present application.

Claims

1. A surface matrix sensor based on hyperspectral technology, characterized in that: include: a sensor module, the sensor module being configured to extend through the housing into the formation; a hyperspectral module, the hyperspectral module extending into the stratum through the housing; A communication module, wherein the communication module is communicatively connected with the sensor module and the hyperspectral module respectively; A terminal module is communicatively connected to the communication module, the terminal module stores a plurality of correlation relationships between hyperspectral curve features and surface matrix physical and chemical property parameters, each correlation relationship corresponding to a preset regional type; the terminal module obtains hyperspectral information measured by the hyperspectral module; the terminal module determines the correlation relationship corresponding to the stratum according to the regional type corresponding to the stratum; the terminal module completes surface matrix monitoring of the stratum according to the hyperspectral information, the correlation relationship, and the surface matrix physical and chemical property parameters; The correlation relationship includes hyperspectral inversion parameters. The terminal module inverts the hyperspectral information according to the hyperspectral inversion parameters to obtain the surface matrix physical and chemical property parameters corresponding to the hyperspectral information, including: parsing the pre-processed hyperspectral information according to the pre-trained surface matrix inversion model. The surface matrix inversion model is constructed based on the machine learning algorithm. The input layer receives the pre-processed spectral feature vector, and the output layer generates the inversion parameter set; the preprocessing includes using a convolutional neural network to perform spectral noise reduction on the original hyperspectral data, and extracting the characteristic bands associated with the surface matrix physical and chemical property parameters through the attention mechanism; the inversion parameter set is input into In the association relationship optimization module, the corresponding physical inversion equation is matched based on the regional type, and the weight coefficients in the inversion equation are dynamically calibrated using a genetic algorithm; a regional adaptive inversion matrix is ​​constructed based on the calibrated weight coefficients, and the hyperspectral information is mapped to the target surface matrix physical and chemical property parameter space through matrix operations to generate surface matrix physical and chemical property parameters containing multiple parameter confidence levels; among them, the machine learning algorithm adopts a hybrid architecture combining random forest and deep belief network, the physical inversion equation contains nonlinear compensation terms trained by the regional geological sample library, and the parameter confidence level is quantified through Monte Carlo simulation.

2. The surface matrix sensor according to claim 1, characterized in that: The hyperspectral module includes: an optical fiber module, the optical fiber module extending through the housing into the formation; A hyperspectral photosensitive chip connected to the optical fiber module; A hyperspectral lens, connected to the hyperspectral photosensitive chip, and disposed in the stratum; a light source, the light source being used to provide supplementary lighting to the stratum; The hyperspectral lens measures the hyperspectral light information corresponding to the stratum, and sends the hyperspectral light information to the hyperspectral photosensitive chip, which generates the hyperspectral information.

3. The surface matrix sensor according to claim 2, characterized in that: Also includes: A convex lens is used to converge the light from the light source to the stratum into the hyperspectral lens through the convex lens.

4. The surface matrix sensor according to claim 3, characterized in that: The corresponding range of the sensitive wavelength of the hyperspectral photosensitive chip and the light source is 360nm to 1100nm, and the resolution of the convex lens is 1nm.

5. The surface matrix sensor according to claim 4, characterized in that: Also includes: A reflector is arranged in the water body of the stratum. The light from the light source irradiates the water body and is reflected back to the hyperspectral lens through the reflector. The hyperspectral lens sends the received light reflected from the water body to the hyperspectral photosensitive chip, and the hyperspectral photosensitive chip generates water quality monitoring information.

6. The surface matrix sensor according to claim 1, characterized in that: One end of the housing connected to the stratum comprises a plurality of sensor monitoring holes and a hyperspectral lens hole, wherein the sensor monitoring holes are arranged around the hyperspectral lens hole; The sensor module includes a plurality of sensor probes, which are disposed in the ground formation through the sensor monitoring hole. The hyperspectral lens of the hyperspectral module is disposed in the ground formation through the hyperspectral lens hole.

7. The surface matrix sensor according to claim 6, characterized in that: The hyperspectral lens and the sensor probe are vertically arranged in the stratum; or, the hyperspectral lens and the sensor probe are horizontally arranged in the stratum.

8. A surface matrix monitoring method based on hyperspectral technology, characterized in that: A surface matrix sensor based on hyperspectral technology applied to any one of claims 1-7; the method comprising: Obtaining a plurality of correlation relationships between hyperspectral curve features and surface matrix physical and chemical property parameters, each of the correlation relationships corresponding to a preset regional type; Acquiring hyperspectral information measured by the hyperspectral module; Determining the association relationship corresponding to the stratum according to the regional type corresponding to the stratum; The surface matrix monitoring of the stratum is completed according to the hyperspectral information, the correlation relationship and the physical and chemical property parameters of the surface matrix.

Images