High and cold wetland surface layer coverage type monitoring system and method

By integrating multispectral radiation monitoring, adaptive environmental calibration, intelligent data processing, and multimodal transmission, the high-altitude wetland monitoring system solves the spatial and temporal limitations of traditional monitoring technologies, achieving high-precision, long-term stable coverage type monitoring, adapting to complex environments and terrains, and meeting the needs for rapid response.

CN120820495APending Publication Date: 2025-10-21CHENGDU UNIV OF INFORMATION TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202511015749.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-07-23
Publication Date
2025-10-21

AI Technical Summary

Technical Problem

Existing monitoring technologies cannot achieve real-time, multi-element collaborative perception of surface cover types in high-altitude wetlands. Traditional ground surveys are limited by the high-altitude environment, satellite remote sensing cannot capture instantaneous changes, and sensor networks lack multi-element collaborative perception capabilities, thus failing to meet the monitoring needs for rapid response.

Method used

It integrates a surface multispectral radiation monitoring unit, an adaptive environment calibration unit, an intelligent data processing unit, a low-power energy management unit, and a multimodal data transmission unit, including a red/near-infrared dual-band sensor array, an adaptive environment calibration module, an edge computing chip, a solar tracking module, and BeiDou short message and LoRa transmission modes, to achieve dynamic adjustment and multi-dimensional collaborative monitoring.

Benefits of technology

It has achieved high-precision, long-term stable monitoring of surface cover types in alpine wetlands, adapts to complex terrain and environment, ensures the real-time, continuous and reliable nature of monitoring, and improves monitoring efficiency and data integrity.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120820495A_ABST
    Figure CN120820495A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of monitoring systems, in particular to a high and cold wetland surface layer coverage type monitoring system and method. Comprising an earth surface multispectral radiation monitoring unit, a self-adaptive environment calibration unit, an intelligent data processing unit, a low-power-consumption energy management unit and a multi-modal data transmission unit, the earth surface multispectral radiation monitoring unit comprises a red light / near-infrared dual-band sensor array and an angle self-adaptive adjustment module, and can dynamically adjust a monitoring angle to adapt to topographic relief; the self-adaptive environment calibration unit is internally provided with a temperature, humidity and atmospheric scattering correction module and can correct the influence of environment interference on spectral data in real time; the intelligent data processing unit adopts an edge calculation chip and can locally complete reflectivity calculation and coverage type judgment; the method can adapt to complex terrains and dynamic environments, and efficient data processing and stable operation are achieved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of monitoring systems, and in particular to a system and method for monitoring surface cover types of alpine wetlands. Background Art

[0002] As a core component of the Qinghai-Tibet Plateau ecosystem, alpine wetlands have a direct impact on regional carbon cycling, hydrological regulation, and biodiversity conservation through dynamic changes in surface cover types, vegetation, water bodies, and permafrost. However, existing monitoring technologies face severe challenges: traditional ground surveys are limited by transportation barriers and labor costs in high-altitude environments, making it difficult to achieve large-scale, high-frequency monitoring; while satellite remote sensing can cover a wide area, it is limited by revisit cycles and interference from cloudy and snowy weather, making it impossible to capture instantaneous changes in wetland surface cover. For example, the average annual cloud cover rate in the alpine wetlands of the Qinghai-Tibet Plateau exceeds 70%, and the effective observation window for optical imaging is less than 30%, making it difficult to continuously track surface processes during key phenological periods, such as snowmelt and vegetation greening.

[0003] The core flaw of existing technologies is the lack of real-time perception and dynamic response capabilities. Although some studies have improved the accuracy of remote sensing classification by improving spectral indices or time series analysis algorithms, they still rely on periodic image acquisition and offline processing, and cannot meet the monitoring needs of wetland ecosystems for rapid response in the context of climate change. For example, the transformation process of swamp meadows to alpine meadows is often accompanied by instantaneous fluctuations in soil moisture, and traditional methods rely on multi-period image comparisons, resulting in monitoring lags of several weeks to months. In addition, existing sensor networks, such as soil moisture monitoring nodes, can achieve real-time collection of local parameters, but lack the ability to collaboratively perceive multiple factors such as vegetation type and water distribution, making it difficult to build a complete surface cover dynamic model.

[0004] Therefore, there is an urgent need for an automated monitoring system that can break through the limitations of time and space and realize collaborative perception of multiple factors to solve the fundamental deficiencies of existing technologies in real-time, continuity and environmental adaptability. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a high-altitude wetland surface cover type monitoring system, comprising:

[0006] Surface multispectral radiation monitoring unit, adaptive environmental calibration unit, intelligent data processing unit, low-power energy management unit and multimodal data transmission unit;

[0007] The surface multispectral radiation monitoring unit includes a red light / near infrared dual-band sensor array and an angle adaptive adjustment module, which can dynamically adjust the monitoring angle to adapt to the terrain;

[0008] The adaptive environmental calibration unit has built-in temperature, humidity and atmospheric scattering correction modules, which can correct the impact of environmental interference on spectral data in real time;

[0009] The intelligent data processing unit uses an edge computing chip to locally complete reflectivity calculation and coverage type discrimination;

[0010] The low-power energy management unit includes a solar tracking module and a battery health monitoring circuit, which can dynamically adjust energy distribution according to light intensity;

[0011] The multimodal data transmission unit integrates two transmission modes: Beidou short message and LoRa, and can automatically switch according to signal strength.

[0012] Preferably, the red light / near infrared dual-band sensor array includes 4 red light sensors and 4 near infrared sensors, which are distributed in a ring shape. The field of view angle of each sensor is 30°, and the monitoring wavelength ranges are 620-670nm and 830-870nm respectively. The sampling frequency of the sensor array can be adaptively adjusted within the range of 1-60 minutes through the intelligent data processing unit, and the sampling frequency will be automatically increased when a rapid change in the coverage type is detected.

[0013] Further preferably, the temperature correction module of the adaptive environmental calibration unit adopts a platinum resistance temperature sensor with a measurement range of -40°C to 60°C and an accuracy of ±0.1°C; the humidity correction module adopts a capacitive humidity sensor with a measurement range of 0-100%RH and an accuracy of ±2%RH; the atmospheric scattering correction module collects atmospheric optical thickness data in real time, establishes a scattering coefficient and spectral attenuation model, and performs band-specific correction on the original spectral data.

[0014] Further preferably, the intelligent data processing unit includes an FPGA chip and an ARM processor. The FPGA chip is responsible for high-speed spectral data acquisition and preprocessing, and the ARM processor is responsible for reflectance calculation, vegetation index analysis and coverage type discrimination. The built-in coverage type discrimination model can be remotely updated via OTA, supporting dynamic identification of new coverage types. The data storage module uses an industrial-grade SD card with a capacity of not less than 64GB, which can store monitoring data at intervals of 15 minutes for at least 3 years.

[0015] Further preferably, the intelligent data processing unit uses the following formula when calculating the normalized vegetation index:

[0016]

[0017] Among them, α IR is the reflectivity in the near-infrared band, α R is the reflectivity of red light band, k IR is the near-infrared band environmental correction coefficient, kR is the red light band environmental correction coefficient, δ is a minimum constant with a value of 0.001, which is used to avoid the denominator being zero. The environmental correction coefficient is dynamically generated by the adaptive environmental calibration unit according to the real-time monitored temperature, humidity and atmospheric optical thickness.

[0018] Further preferably, the low power consumption energy management unit adopts the following energy allocation formula:

[0019]

[0020] Among them, P 分配 The real-time power distribution of each unit, P 总 is the total output power of the system, I is the current light intensity, I max is the maximum light intensity threshold, SOC is the battery state of charge, SOC max is the battery full charge value, T is the current ambient temperature, T opt is the optimal operating temperature, ω1, ω2, and ω3 are the weight coefficients of light, battery status, and temperature, respectively, and their sum is 1. This formula is used to achieve dynamic energy allocation and ensure that the core functions of the system run first in extreme environments.

[0021] Further preferably, the transmission strategy of the multimodal data transmission unit satisfies the following formula:

[0022]

[0023] Among them, D 传输 is the actual amount of data transmitted, D 有效 is the effective data volume to be transmitted, SNR is the current signal-to-noise ratio, SNR 阈值 is the minimum communication signal-to-noise ratio, E is the residual energy, and E 临界 is the minimum operating energy of the system, β1 and β2 are the weight coefficients of signal quality and energy state respectively, the sum of which is 1. When SNR <SNR 阈值 And E>E 临界 When SNR≥SNR 阈值 and E≤E 临界 When transmitting, keep the LoRa mode and compress the data volume.

[0024] Further preferably, the coverage type discrimination logic of the intelligent data processing unit adopts a three-level decision mechanism:

[0025] The first-level decision is based on the preliminary classification of the relationship between the NDVI value and reflectivity calculated in claim 5. The second-level decision introduces a time series continuity check, and the classification is confirmed when the sampling results are consistent for three consecutive times. The third-level decision is combined with historical data comparison, and a manual review request is initiated when the deviation from the data of the same area in the past 30 days exceeds 30%.

[0026] The discrimination conditions are expanded to include the type of frozen soil cover:

[0027] When the red light reflectance is 0.30-0.50, the near-infrared reflectance is 0.25-0.45, the NDVI value is -0.1 to 0.0, and the soil temperature is below 0℃ for 5 consecutive days, it is judged to be frozen soil.

[0028] A method for monitoring the surface cover type of an alpine wetland, applied to any one of the above-mentioned systems for monitoring the surface cover type of an alpine wetland, comprising:

[0029] S1: When the system is initialized, the adaptive environmental calibration unit collects reference environmental parameters and establishes an initial calibration model;

[0030] S2: The surface multispectral radiation monitoring unit performs scanning data acquisition at preset angle intervals to obtain red light / near-infrared reflectance data at different angles;

[0031] S3: Perform angle normalization on the collected raw spectral data to convert the reflectance at different angles into vertical observation equivalent values;

[0032] S4: Call the formula of claim 5 to calculate the ratio of the corrected NDVI value to the reflectance;

[0033] S5: Combining the three-level decision-making mechanism with the extended coverage type discrimination criteria to identify the surface coverage type;

[0034] S6: The intelligent data processing unit stores the recognition results by timestamp and generates a data quality assessment report, wherein the data quality assessment report includes data integrity, correction accuracy, and classification confidence;

[0035] S7: The multimodal data transmission unit dynamically adjusts the transmission strategy according to the transmission strategy formula of the multimodal data transmission unit, and sends the monitoring results and the quality assessment report to the remote terminal;

[0036] S8: The low-power energy management unit adjusts the working mode of each unit, including sampling frequency, calculation accuracy, and transmission interval, according to the real-time energy status and energy allocation formula to achieve long-term battery life.

[0037] Further preferably, the angle normalization processing in S3 is specifically as follows: by establishing a relationship model between reflectivity and observation angle, the reflectivity data of the inclined observation is corrected to the equivalent value under the vertical observation angle, and the terrain slope factor is introduced in the correction process. When the slope exceeds 15°, the terrain shielding correction coefficient is additionally added to ensure the comparability of spectral data under different terrain conditions. The angle normalization processing can control the monitoring data deviation of different terrain areas within 5%.

[0038] Technical effect:

[0039] The innovative technology of this invention lies in its integrated surface multispectral radiation monitoring unit, which includes adaptive angle adjustment, adaptive environmental calibration, intelligent data processing, low-power energy management, and multimodal data transmission. This unit achieves high-precision monitoring through multi-dimensional collaboration. This system overcomes the problems of inaccurate and interrupted monitoring caused by a single fixed viewing angle, lack of dynamic environmental correction, poor real-time data processing, irrational energy allocation, and a single transmission mode. It can adapt to complex terrain and environments, ensure long-term stable operation, and improve monitoring efficiency and reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 This is the block diagram of the alpine wetland surface cover type monitoring system for this application;

[0041] Figure 2 This is a flow chart of the method for monitoring surface cover types in alpine wetlands for this application;

[0042] Figure 3 This is a schematic diagram of the overall structure of the monitoring device for this application;

[0043] Figure 4 A detailed flow chart for monitoring the surface cover types of alpine wetlands for this application;

[0044] Figure 5 This is a comparison chart of the coverage type identification results and actual measurement values ​​of the monitoring system of this application.

[0045] In the figure: 1. Tripod; 2. Data acquisition and control box; 3. Solar panel; 4. Power / signal cable; 5. Data sending / receiving antenna; 6. Red light / near infrared sensor. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0047] Traditional monitoring systems have the following technical problems: existing alpine wetland monitoring equipment mostly uses a single fixed-angle monitoring method, which is difficult to adapt to the spectral data deviation caused by terrain undulations; there is a lack of real-time environmental calibration mechanism, and environmental factors such as temperature and humidity can easily interfere with the accuracy of spectral measurements; data processing relies on remote terminals and cannot achieve local rapid analysis; the energy management strategy is simple and the endurance is insufficient in extreme climates; the data transmission mode is single and easily affected by complex terrain, resulting in transmission interruption, making it difficult to meet long-term high-precision monitoring needs.

[0048] Based on this, see Figure 1 This embodiment provides a high-altitude wetland surface cover type monitoring system, including a surface multi-spectral radiation monitoring unit, an adaptive environmental calibration unit, an intelligent data processing unit, a low-power energy management unit and a multimodal data transmission unit. The surface multi-spectral radiation monitoring unit includes a red light / near-infrared dual-band sensor array and an angle adaptive adjustment module, which can dynamically adjust the monitoring angle to adapt to the terrain undulations; the adaptive environmental calibration unit has a built-in temperature, humidity, and atmospheric scattering correction module, which can correct the impact of environmental interference on spectral data in real time; the intelligent data processing unit adopts an edge computing chip, which can locally complete reflectivity calculation and cover type discrimination; the low-power energy management unit includes a solar tracking module and a battery health monitoring circuit, which can dynamically adjust energy distribution according to light intensity; the multimodal data transmission unit integrates Beidou short message and LoRa transmission modes, and can automatically switch according to signal strength. The system realizes high-precision, long-term and stable monitoring of the surface cover type of high-altitude wetlands through multi-dimensional data fusion and environmental adaptive correction.

[0049] Through multi-unit collaborative design, this solution addresses the shortcomings of traditional systems in terrain adaptability, environmental interference resistance, data processing timeliness, energy endurance, and transmission stability, thereby improving the overall performance of alpine wetland monitoring. It is worth mentioning that the angle adaptive adjustment module of the surface multispectral radiation monitoring unit can dynamically adjust the monitoring angle according to terrain changes to ensure the consistency of spectral data acquisition; the adaptive environmental calibration unit collects temperature, humidity, and atmospheric parameters in real time and establishes a correction model to eliminate environmental interference; the edge computing chip of the intelligent data processing unit realizes local rapid analysis and reduces data transmission delays; the low-power energy management unit optimizes energy distribution through solar tracking and battery monitoring; and the multimodal data transmission unit switches transmission modes according to signal strength to ensure smooth data transmission.

[0050] The technical effects achieved by the above embodiments include: angle adaptive adjustment enables the system to obtain reliable spectral data in areas with complex terrain, avoiding classification errors caused by observation angle deviation; real-time environmental calibration improves the accuracy of spectral data and provides high-quality raw data for coverage type identification; local data processing enables rapid output of monitoring results to meet the needs of real-time response to environmental changes; dynamic energy management ensures that the system can still operate stably in insufficient light or extreme temperatures, extending the field work cycle; multi-mode transmission ensures the effective transmission of data in complex terrain, realizing the complete collection of monitoring data. The functions of each unit complement each other, comprehensively improving the accuracy, efficiency and reliability of surface cover type monitoring in alpine wetlands.

[0051] Traditional monitoring systems have the following technical problems: most existing sensors are single-band or configured with a small number of sensors, with a fixed sampling frequency, making it difficult to balance the monitoring range and data timeliness. Data omissions are prone to occur when the coverage type changes rapidly, and unreasonable spectral band range settings will affect the accuracy of reflectance data and cannot accurately support coverage type judgment.

[0052] Based on this, the red light / near-infrared dual-band sensor array includes 4 red light sensors and 4 near-infrared sensors, which are distributed in a ring. The field of view of each sensor is 30°, and the monitoring wavelength ranges are 620-670nm (red light) and 830-870nm (near-infrared). The sampling frequency of the sensor array can be adaptively adjusted within the range of 1-60 minutes through an intelligent data processing unit, and the sampling frequency will be automatically increased when a rapid change in coverage type is detected.

[0053] This solution solves the problem of incomplete data caused by the limited monitoring range and fixed sampling frequency of traditional sensors by optimizing the sensor array configuration and sampling strategy, and improves the quality and timeliness of spectral data collection. It is worth mentioning that: 4 red light and 4 near-infrared sensors are distributed in a ring, with a single field of view of 30°. When combined, they can achieve a larger coverage monitoring range and reduce monitoring blind spots; the wavelength range setting of 620-670nm for red light and 830-870nm for near-infrared light matches the spectral reflectance characteristics of different cover types on the wetland surface, ensuring that the reflection differences of vegetation, water bodies, bare soil, etc. can be effectively captured; the adaptive adjustment mechanism of the sampling frequency uses a lower frequency to save energy when the cover type is stable, and increases the frequency to capture key changes when changes are drastic, achieving a balance between efficiency and energy consumption.

[0054] The technical effects achieved by the above-mentioned embodiments include: the circularly distributed sensor array expands the monitoring range, enabling the system to fully cover the target area and avoid the limitations of single-point monitoring; the precise wavelength range setting ensures the effective differentiation of reflectance data of different coverage types, providing a reliable basis for subsequent NDVI calculation and type identification; the adaptive sampling frequency enables the system to capture data in a timely manner when the coverage type changes rapidly, such as when snowmelt or waterlogged areas recede, ensuring the continuity and integrity of monitoring; and the frequency is reduced during stable periods to reduce energy consumption and extend system endurance. These designs collectively enhance the scientific and practical nature of sensor data acquisition, laying the foundation for the efficient operation of the entire monitoring system.

[0055] Traditional monitoring systems have the following technical problems: existing monitoring equipment lacks an effective environmental calibration mechanism. Environmental factors such as temperature, humidity and atmospheric scattering can significantly affect the authenticity of spectral data, leading to deviations in reflectance calculations, which in turn affects the accuracy of coverage type identification. In addition, the calibration parameters are fixed and cannot adapt to dynamic changes in the field environment.

[0056] Based on this, the temperature correction module of the adaptive environmental calibration unit uses a platinum resistance temperature sensor with a measurement range of -40°C to 60°C and an accuracy of ±0.1°C. The humidity correction module uses a capacitive humidity sensor with a measurement range of 0-100% RH and an accuracy of ±2% RH. The atmospheric scattering correction module collects atmospheric optical thickness data in real time, establishes a scattering coefficient and spectral attenuation model, and performs band-specific correction on the original spectral data.

[0057] This solution solves the problem of spectral data distortion caused by environmental interference by integrating a high-precision environmental correction module and establishing a dynamic correction model, thereby improving the reliability of monitoring data.

[0058] It is worth mentioning that: the platinum resistance temperature sensor maintains an accuracy of ±0.1°C over a wide range of -40°C to 60°C, and can accurately capture the drastic temperature changes in alpine wetlands; the capacitive humidity sensor has a measurement range of 0-100% RH and an accuracy of ±2% RH, which can adapt to the monitoring needs of high-humidity environments in wetlands; the atmospheric scattering correction module collects the atmospheric optical thickness in real time, establishes a targeted relationship between the scattering coefficient and spectral attenuation in different bands, and realizes specific correction of the red light and near-infrared bands, avoiding the limitations of general correction methods.

[0059] The technical effects achieved by the above-mentioned embodiments include: high-precision temperature correction ensures the stability of spectral data under extreme temperature conditions, reducing sensor response deviations caused by temperature changes; precise humidity monitoring and correction eliminates the impact of high humidity environments on light propagation, ensuring the authenticity of reflectance data; band-specific atmospheric scattering correction effectively compensates for the differences in atmospheric scattering of light of different wavelengths, allowing red and near-infrared reflectance data to more accurately reflect surface cover characteristics. These correction measures work together to significantly improve the quality of the original spectral data, providing accurate basic data for subsequent NDVI calculations and cover type identification, reducing classification errors caused by environmental interference, and improving the overall accuracy of the monitoring system.

[0060] Traditional monitoring systems have the following technical problems: existing data processing units mostly rely on remote computing, which leads to data transmission delays and is easily restricted by network conditions; the data storage capacity is limited and cannot meet long-term monitoring needs; the processing algorithm is fixed and difficult to adapt to the coverage type judgment of different wetland environments; and there is a lack of effective data management mechanisms, which affects the timeliness and reliability of monitoring results.

[0061] Based on this, the intelligent data processing unit includes an FPGA chip and an ARM processor. The FPGA chip is responsible for high-speed spectral data acquisition and preprocessing, and the ARM processor is responsible for reflectance calculation, vegetation index analysis and coverage type discrimination. The built-in coverage type discrimination model can be remotely updated via OTA, supporting dynamic identification of new coverage types. The data storage module uses an industrial-grade SD card with a capacity of not less than 64GB, which can store monitoring data at intervals of 15 minutes for at least 3 years.

[0062] This solution solves problems such as data processing delays, insufficient storage, and model rigidity in traditional systems by optimizing data processing hardware configuration and software functions, thereby improving the efficiency and flexibility of data processing.

[0063] It is worth mentioning that the high-speed processing capability of the FPGA chip ensures real-time acquisition and preprocessing of spectral data, avoiding data accumulation; the ARM processor focuses on complex reflectance calculations, NDVI analysis, and type discrimination, with clear division of labor to improve processing efficiency; the OTA remote update mechanism enables the discrimination model to be optimized according to monitoring needs and environmental changes, supporting the identification of new coverage types, such as special vegetation and newly emerged waterlogged areas; the large-capacity storage of the 64GB industrial-grade SD card meets long-term data preservation needs, and the 15-minute interval storage strategy balances data density and storage capacity.

[0064] The technical benefits achieved by the above-mentioned embodiments include: The collaborative processing of FPGA and ARM enables rapid processing of spectral data, reducing the time delay between data acquisition and type discrimination, meeting the needs of real-time monitoring; the updateable discrimination model enables the system to adapt to the unique environments and cover types of different alpine wetlands, enhancing the system's versatility and scalability; and the large-capacity storage ensures the complete preservation of long-term monitoring data, providing a data foundation for long-term analysis of wetland ecological changes and preventing data loss from impacting research continuity. These features make the intelligent data processing unit the core of the system, effectively enhancing the practicality and scientific value of the monitoring system.

[0065] Traditional monitoring systems have the following technical problems: existing NDVI calculations mostly use a simple reflectance ratio formula, which does not consider the impact of environmental factors on different bands, resulting in deviations in the calculation results. When the reflectance values ​​are close, calculation errors such as the denominator being zero are prone to occur, affecting the accuracy of the vegetation index and thus reducing the reliability of cover type discrimination.

[0066] Based on this, the intelligent data processing unit uses the following formula to calculate the normalized vegetation index:

[0067]

[0068] Among them, α IR is the reflectivity in the near-infrared band, α R is the reflectivity of red light band, k IR is the near-infrared band environmental correction coefficient, k R is the red light band environmental correction coefficient, δ is a minimum constant with a value of 0.001, which is used to avoid the denominator being zero. The environmental correction coefficient is dynamically generated by the adaptive environmental calibration unit according to the real-time monitored temperature, humidity and atmospheric optical thickness.

[0069] Among them, α IR Represents the near-infrared band reflectivity, which is the reflection ratio of the ground surface to the near-infrared light received by the red light / near-infrared sensor, and directly reflects the optical characteristics of the ground object in the near-infrared band; α R Represents the red light band reflectivity, that is, the ratio of the ground surface’s reflection of red light received by the sensor, reflecting the optical characteristics of the ground object in the red light band. IR k is the near-infrared band environmental correction coefficient, which is dynamically generated by the adaptive environmental calibration unit based on the real-time monitored temperature, humidity, and atmospheric optical thickness. It is used to compensate for environmental factors such as changes in sensor sensitivity caused by low temperature and interference of enhanced light scattering caused by high humidity on near-infrared light reflectance data. Ris the red light band environmental correction coefficient, also generated by the adaptive environmental calibration unit. It corrects for the environmental characteristics of the red light band, such as the difference in the absorption of red light by water vapor in the atmosphere, to ensure the consistency of red light reflectance data under different environmental conditions. δ is a minimum constant with a value of 0.001. Its function is to avoid calculation errors caused by the zero value in the denominator of the formula, and to ensure that the α IR ·k IR With α R ·k R When the values ​​are close or even equal, the formula can still operate stably. This formula solves the problem that traditional NDVI calculation does not take into account the dynamic impact of the environment by introducing an environmental correction coefficient.

[0070] This solution improves the NDVI calculation formula, solves the problem of poor environmental adaptability and easy calculation errors of the traditional formula, and improves the accuracy of the vegetation index. It is worth mentioning that the k introduced in the formula IR and k R The correction coefficient, generated by the adaptive environmental calibration unit based on real-time environmental parameters, can specifically compensate for the different effects of temperature, humidity, and atmospheric scattering on the near-infrared and red light bands. The setting of δ effectively avoids the situation where the denominator is zero, ensuring the stability of the calculation. The formula structure retains the physical meaning of NDVI while enhancing adaptability to environmental changes.

[0071] The technical effects achieved by the above embodiments include: the introduction of the dynamic environmental correction coefficient enables the NDVI value to more realistically reflect the vegetation growth conditions and surface cover characteristics, and reduce the vegetation index deviation caused by environmental interference. For example, in a high humidity environment, the correction coefficient can eliminate the influence of water vapor on near-infrared light reflection, making the calculation results more accurate; the minimum value setting of δ ensures the effective calculation of the formula under various reflectivity combinations, avoiding system crashes caused by data anomalies; accurate NDVI values ​​provide a reliable basis for coverage type discrimination, making the distinction between vegetation, dead grass, bare soil and other types more accurate, and improving the sensitivity and recognition accuracy of the entire monitoring system to surface cover changes.

[0072] The recognition accuracy of this system is Figure 5 Comparative experiments have verified that in typical cover types such as permafrost, vegetation, and snow cover, the consistency between the monitoring results and the actual measurement values ​​is more than 95%, and the data deviation is controlled within 5%, which is significantly better than the traditional single sensor monitoring method.

[0073] Traditional monitoring systems have the following technical problems: existing energy management mostly adopts a fixed allocation mode, which does not take into account dynamic factors such as light intensity, battery status and ambient temperature. In insufficient light or extreme temperatures, unreasonable energy distribution is prone to occur, resulting in insufficient power supply to core functions or energy waste, affecting the long-term stable operation of the system in alpine wetlands.

[0074] Based on this, the low-power energy management unit adopts the following energy allocation formula:

[0075]

[0076] Among them, P 分配 The real-time power distribution of each unit, P 总 is the total output power of the system, I is the current light intensity, I max is the maximum light intensity threshold, SOC is the battery state of charge, SOC max is the battery full charge value, T is the current ambient temperature, T opt is the optimal operating temperature, ω1, ω2, and ω3 are the weight coefficients of light, battery status, and temperature, respectively, and their sum is 1. This formula is used to achieve dynamic energy allocation and ensure that the core functions of the system run first in extreme environments.

[0077] Among them, P 分配 It represents the real-time distribution power of each unit in the system, such as sensors, data processing units, and transmission units, which determines the operating status of each unit; 总 is the total output power of the system, provided by the solar panels and batteries, reflecting the total energy currently available to the system. I is the current light intensity, which directly affects the power generation efficiency of the solar panels; I max is the maximum light intensity threshold, which is a reference value set based on the historical light data of the alpine wetland and is used to normalize the current light intensity. SOC is the battery state of charge, which reflects the current storage level of the battery; SOC max is the battery's fully charged state value, serving as the normalized benchmark for the state of charge. T is the current ambient temperature, which affects device power consumption and battery performance; T opt The optimal operating temperature for the system, at which device power consumption is lowest and performance is most stable, is achieved. ω1, ω2, and ω3 represent weighting coefficients for light intensity, battery status, and temperature, respectively, summing to 1. These coefficients can be adjusted based on seasonal and environmental characteristics, such as increasing ω2 in winter when light intensity is low. This formula dynamically allocates energy by integrating multiple factors, resolving the issue of irrational energy utilization in extreme environments with traditional fixed allocation models. This ensures that core functions are prioritized during periods of energy shortage, extending the system's field operating time.

[0078] This solution addresses the poor adaptability of traditional fixed allocation models by introducing a multi-factor dynamic energy allocation formula, optimizing energy efficiency. Notably, the formula comprehensively considers light intensity, which determines solar input, battery state of charge, energy storage level, and ambient temperature, affecting device power consumption and battery performance. Weighting factors balance the influence of these factors. When sunlight is sufficient, the power of each unit can be appropriately increased to increase data collection frequency. When the battery charge is low, power is prioritized for core processing and transmission units. In extremely low temperatures, the power of the temperature control unit is increased to maintain device performance. The technical effects achieved by the above embodiments include: dynamic energy distribution enables the system to provide stable power supply in alpine wetlands with large light fluctuations, avoiding monitoring interruptions due to energy shortages; by prioritizing core functions such as data processing and key sensor operations, it ensures that key monitoring data can still be obtained when energy is tight; adjusts power according to ambient temperature to reduce energy waste at extreme temperatures and extend battery life; flexible setting of weight coefficients can adapt to the energy characteristics of different seasons, such as increasing the battery status weight when light is weak in winter, so that the system can operate efficiently under different climatic conditions throughout the year, providing continuous energy support for long-term field monitoring.

[0079] Traditional monitoring systems have the following technical problems: existing data transmission mostly adopts a single mode, which is easily affected by signal blocking in the complex terrain of alpine wetlands, resulting in data transmission interruption. It also does not take into account the remaining energy status, and may exhaust energy due to blind transmission, affecting the basic operation of the system and unable to guarantee the complete transmission of monitoring data.

[0080] Based on this, the transmission strategy of the multimodal data transmission unit satisfies the following formula:

[0081]

[0082] Among them, D 传输 is the actual amount of data transmitted, D 有效 is the effective data volume to be transmitted, SNR is the current signal-to-noise ratio, SNR 阈值 is the minimum communication signal-to-noise ratio, E is the residual energy, and E 临界 is the minimum operating energy of the system, β1 and β2 are the weight coefficients of signal quality and energy state respectively, the sum of which is 1. When SNR <SNR 阈值 And E>E 临界 When SNR≥SNR 阈值 and E≤E 临界 When transmitting, keep the LoRa mode and compress the data volume.

[0083] D 传输 The actual amount of data transmitted determines the size of the information transmitted each time; D 有效The amount of effective data to be transmitted, including key information such as monitoring results and status parameters. SNR is the current signal-to-noise ratio, reflecting the quality of data transmission. The higher the value, the more stable the signal. 阈值 is the minimum communication signal-to-noise ratio. When it is lower than this value, transmission is prone to errors. E is the remaining energy of the system, which affects subsequent operation; E 临界 The minimum operating energy of the system. If it is lower than this value, the system will be paralyzed. β1 and β2 are the weight coefficients of signal quality and energy status respectively, and the sum is 1. They can be adjusted according to the monitoring priority. For example, β1 can be increased for emergency monitoring. <SNR 阈值 And E>E 临界 The formula will automatically reduce D 传输 And switch to Beidou short message mode; when SNR≥SNR 阈值 and E≤E 临界 When D 传输 This formula solves the problems of low reliability and uncontrolled energy consumption in traditional single transmission modes in complex terrains, balances data integrity and system endurance, and ensures the effective transmission of monitoring data.

[0084] This solution solves the problems of low reliability and uncontrolled energy consumption in traditional single-mode transmission by establishing a dual-factor transmission strategy of signal and energy, and improves the stability and energy efficiency of data transmission.

[0085] It is worth mentioning that the formula combines the signal-to-noise ratio with the remaining energy to dynamically adjust the amount of transmitted data; when the signal is poor but the energy is sufficient, it switches to Beidou short messages to ensure the transmission of key data; when the energy is tight but the signal is good, the data volume is compressed to reduce energy consumption; the weight coefficient can be adjusted according to the monitoring priority to balance data integrity and energy consumption.

[0086] The technical effects achieved by the above embodiments include: multi-modal switching enables the system to still achieve data transmission in areas with weak signals such as valleys and dense forests, avoiding the loss of monitoring data; the energy-aware transmission strategy prevents system paralysis due to excessive transmission, ensuring the continuous operation of basic monitoring functions; dynamic adjustment of data volume enables key information to be transmitted first, and non-critical data to be supplemented when conditions are appropriate, thereby ensuring the timeliness of monitoring and saving energy; it is suitable for the complex terrain and communication conditions of high-altitude wetlands, achieving efficient and stable transmission of monitoring data, and providing complete data support for remote analysis.

[0087] The traditional monitoring system has the following technical problems: the existing cover type identification mostly relies on a single criterion, does not consider data continuity and historical comparison, is easily affected by instantaneous interference and leads to misjudgment, and does not include permafrost and other high-altitude wetland-specific cover types, and cannot fully reflect the surface characteristics of the wetland, affecting the integrity and accuracy of monitoring.

[0088] Based on this, the coverage type discrimination logic of the intelligent data processing unit adopts a three-level decision-making mechanism: the first-level decision is based on the preliminary classification of the relationship between the NDVI value and reflectivity calculated in claim 5, the second-level decision introduces time series continuity verification, and confirms the classification when the results of three consecutive samplings are consistent, and the third-level decision is combined with historical data comparison. When the deviation from the data of the same area in the past 30 days exceeds 30%, a manual review request is initiated. The discrimination conditions add frozen soil coverage type discrimination on the original basis: when the red light reflectance is 0.30-0.50, the near-infrared reflectance is 0.25-0.45, the NDVI value is -0.1 to 0.0, and the soil temperature is below 0°C for 5 consecutive days, it is judged to be frozen soil.

[0089] This solution solves the problems of high misjudgment rate and incomplete type coverage of traditional single criterion by building a multi-level decision-making mechanism and expanding the discrimination types, thereby improving the reliability and comprehensiveness of discrimination.

[0090] It is worth mentioning that: the first-level decision-making uses the corrected NDVI and reflectivity to lay the foundation for classification; the second-level decision-making filters out instantaneous interference, such as short-term snow accumulation and occasional occlusion, through continuous sampling verification; the third-level decision-making combines historical data to identify abnormal changes and ensure the rationality of classification; the newly added permafrost discrimination conditions integrate spectral characteristics and soil temperature to accurately identify the permafrost cover unique to alpine wetlands, filling the gaps in traditional classification.

[0091] The technical effects achieved by the above embodiments include: the three-level decision-making mechanism greatly reduces the misjudgment caused by single sampling errors, making the coverage type identification more stable, such as avoiding the misjudgment of short-term water accumulation as long-term water bodies; the inclusion of permafrost types makes the monitoring system more in line with the ecological characteristics of alpine wetlands, and fully captures all types of changes in the wetland surface from vegetation to permafrost; time series verification and historical comparison enhance the credibility of classification results, providing more accurate surface cover data for ecological research; the manual review mechanism provides timely warnings when abnormal changes occur, facilitating scientific researchers to intervene in analysis, taking into account both automation and professionalism, and improving the practical value of the monitoring system.

[0092] Traditional monitoring systems have the following technical problems: the existing monitoring methods have fixed processes, do not consider the impact of terrain differences on spectral data, lack standardized data processing steps, and do not include data quality assessment links. This leads to poor comparability of monitoring data in different regions, making it difficult to ensure the reliability of the results and affecting the scientific nature of subsequent ecological analysis.

[0093] Based on this, see Figure 2 The embodiment of the present application provides a method for monitoring the surface cover type of alpine wetlands, comprising the following steps:

[0094] Step 1: When the system is initialized, the adaptive environmental calibration unit collects reference environmental parameters, including temperature of 25°C, humidity of 50%, and atmospheric optical thickness of 0.1, and establishes an initial calibration model;

[0095] Step 2: The surface multispectral radiation monitoring unit performs scanning data acquisition at preset angle intervals with a step size of 15° to obtain red light / near-infrared reflectance data at different angles;

[0096] Step 3: Perform angle normalization processing on the collected original spectral data to convert the reflectance at different angles into vertical observation equivalent values; Step 4: Use the formula of claim 5 to calculate the ratio of the corrected NDVI value to the reflectance;

[0097] Step 5: Combining the three-level decision-making mechanism with the expanded cover type discrimination criteria, identify the surface cover type including the frozen soil type;

[0098] Step 6: The intelligent data processing unit stores the recognition results by timestamp and generates a data quality assessment report, including data integrity, correction accuracy, and classification confidence;

[0099] Step 7: The multimodal data transmission unit dynamically adjusts the transmission strategy according to the transmission strategy formula of the multimodal data transmission unit, and sends the monitoring results and the quality assessment report to the remote terminal;

[0100] Step 8: The low-power energy management unit adjusts the operating mode, sampling frequency, calculation accuracy, and transmission interval of each unit based on the real-time energy status and the energy allocation formula to achieve long-term battery life. This solution, through a standardized multi-step process and full-chain quality control, solves the problems of poor data comparability and lack of reliability of traditional methods, and improves the systematic and scientific nature of the monitoring method.

[0101] The scanning data acquisition process of the surface multispectral radiation monitoring unit is as follows: Figure 4 As shown in the figure, after obtaining the red light / near-infrared reflectance data, it is first normalized by angle, and then the cover type is identified through NDVI calculation and a three-level decision-making mechanism. The identification of the permafrost type requires the combination of soil temperature data for 5 consecutive days.

[0102] It is worth mentioning that: angle normalization eliminates the observation bias caused by terrain undulations, allowing data from different regions to be directly compared; the establishment of baseline environmental parameters ensures the consistency of the correction model; the data quality assessment report intuitively reflects the reliability of the results; and the dynamic adjustment of energy and transmission strategies adapts to the field environment. The technical effects achieved by the above embodiments include: standardized processes make monitoring data at different times and locations comparable, which facilitates the analysis of the spatiotemporal variation of surface cover in alpine wetlands; angle normalization improves the accuracy of data in complex terrain areas, avoiding misjudgment of types due to different observation angles; data quality assessment provides a reliability reference for subsequent analysis, facilitating the distinction between valid data and interference data; the entire process combines dynamic energy and transmission management to ensure that the method can be implemented in long-term field practice, providing a systematic and reliable monitoring method support for alpine wetland ecological research.

[0103] Traditional monitoring systems have the following technical problems: existing spectral data processing does not take into account the impact of terrain slope on observation angle, and there is a deviation between the reflectance data of inclined observations and vertical observations, resulting in incomparable data in different terrain areas. Especially in alpine wetlands with large slopes, this deviation will significantly affect the accuracy of coverage type discrimination.

[0104] Based on this, the angle normalization processing in step 3 is specifically as follows: by establishing a relationship model between reflectivity and observation angle, the reflectivity data of the oblique observation is corrected to the equivalent value under the vertical observation angle. The terrain slope factor is introduced in the correction process and obtained through the built-in tilt sensor of the sensor. When the slope exceeds 15°, the terrain shielding correction coefficient is additionally added to ensure the comparability of spectral data under different terrain conditions. The angle normalization processing can control the monitoring data deviation in different terrain areas within a smaller range.

[0105] This solution solves the problem of traditional angle correction ignoring the influence of terrain by introducing terrain slope factor and shading correction, and improves the consistency of different terrain data.

[0106] It is worth mentioning that: the reflectivity and observation angle relationship model is constructed based on optical principles, which can accurately convert reflectance data of different tilt angles; the tilt sensor obtains the terrain slope in real time, providing a quantitative basis for correction; the shading correction coefficient when the slope exceeds 15° is specifically aimed at the light blocking problem that may occur in steep areas, compensating for the impact of shadows on reflectivity; the entire processing process is completed automatically without human intervention. The technical effects achieved by the above embodiments include: angle normalization processing makes the oblique observation data consistent with the vertical observation data, eliminating the systematic deviation caused by terrain undulations; the introduction of terrain slope factors makes the correction more in line with the actual terrain, and reliable spectral data can still be obtained in complex areas such as slopes and depressions; shading correction ensures the validity of steep area data, avoiding misjudgment of shadows as water bodies or bare soil; the comparability of different terrain data is improved, making the coverage type distribution map of the entire alpine wetland monitoring area more accurate, providing more realistic basic data for ecosystem assessment.

[0107] The present invention also provides a device for monitoring the surface cover type of alpine wetlands, comprising a tripod 1, a data transmitting / receiving antenna 5 mounted on the top of the tripod 1, a data acquisition and control box 2 mounted on the tripod 1, a fixing frame mounted on the tripod 1, a red light / near-infrared sensor 6 mounted on the fixing frame, and the red light / near-infrared sensor 6 is located below the data transmitting / receiving antenna 5, and a solar panel 3 is also fixedly mounted on the tripod 1, wherein the data acquisition and control box 2, the solar panel 3, the transmitting / receiving antenna 5, and the red light / near-infrared sensor 6 are connected via a power / signal cable 4;

[0108] Specific examples Figure 3As shown, the tripod 1 provides stable support. The tripod 1 employed in the present invention is made of high-strength aluminum alloy, offering excellent low-temperature resistance and corrosion resistance. This ensures stability and durability in harsh environments, and it is lightweight, corrosion-resistant, and impact-resistant. The data acquisition and control box 2 houses an intelligent data processing unit and an adaptive environmental calibration module, offering excellent waterproof and dustproof properties, making it adaptable to various field environments. The solar panels 3 are connected to the battery via a power cable 4, and work in conjunction with a low-power energy management unit to achieve dynamic energy distribution. The solar panels 3 utilize high-efficiency monocrystalline silicon solar cells with a conversion efficiency exceeding 20%, forming a solar power generation array composed of parallel solar panels. An array of red / near-infrared sensors 6 is secured to the top of the tripod via an adaptive angle adjustment module. The data transmit / receive antenna 5 integrates Beidou and LoRa dual-mode transmission capabilities, ensuring data communication in complex terrain. The power / data cable 4 utilizes a special, low-temperature, and wear-resistant cable suitable for connecting mobile appliances with AC rated voltages of 0.6 / 1 kV and below, and requiring special requirements such as cold resistance and oil resistance. The data transmission / reception antenna 5 is a high-gain directional antenna that enables stable data transmission over long distances, unaffected by complex terrain and climatic conditions. The red light / near-infrared sensor 6 is a highly sensitive photoelectric conversion device that converts radiation energy into electrical signals. It operates stably in alpine environments, detecting wavelengths covering the red and near-infrared bands, ensuring accurate monitoring of surface cover types in alpine wetlands.

[0109] Example:

[0110] During the monitoring process, the data collector within the data acquisition and control box continuously collects reflected radiation data and environmental parameters every 15 minutes and transmits this data to the data processing module. The data processing module determines and classifies the wetland surface cover type in real time based on pre-set algorithms and criteria, such as snow cover, water, bare soil, dead grass, and vegetation. Once the data processing module receives the complete dataset, it analyzes it based on the criteria and ultimately outputs the monitoring results for the wetland surface cover type.

[0111] The discrimination conditions of the alpine wetland surface cover type monitoring system are as follows:

[0112] (1) If the red reflectance is greater than the infrared reflectance (αR>αIR), the red reflectance is greater than 0.65, the infrared reflectance is greater than 0.40, and the NDVI value is less than 0.0, it is judged to be snow cover.

[0113] (2) If the red light reflectance is greater than the infrared light reflectance (αR>αIR), the red light reflectance is less than 0.10, the infrared light reflectance is greater than 0.05, and the NDVI value is less than 0.0, then it is judged to be a water body.

[0114] (3) If the infrared reflectance is greater than the red reflectance (αIR>αR), the red reflectance is less than 0.10, the infrared reflectance is greater than 0.15, and the NDVI value is between 0.1 and 0.2, it is judged to be bare soil.

[0115] (4) If the infrared reflectance is greater than the red reflectance (αIR>αR), and the red reflectance is less than 0.15, the infrared reflectance is greater than 0.25, and the NDVI value is between 0.2 and 0.4, it is judged to be dead grass.

[0116] (5) If the infrared reflectance is greater than the red reflectance (αIR>αR), and the red reflectance is less than 0.05, the infrared reflectance is greater than 0.55, and the NDVI value is between 0.4 and 0.5, it is judged to be vegetation.

[0117] Among them, αR is the emissivity in the red band, αIR is the emissivity in the near-infrared band, and NDVI is the normalized difference vegetation index.

[0118] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. A system for monitoring the surface cover type of alpine wetlands, characterized in that: include: Surface multispectral radiation monitoring unit, adaptive environmental calibration unit, intelligent data processing unit, low-power energy management unit and multimodal data transmission unit; The surface multispectral radiation monitoring unit includes a red light / near infrared dual-band sensor array and an angle adaptive adjustment module, which can dynamically adjust the monitoring angle to adapt to the terrain; The adaptive environmental calibration unit has built-in temperature, humidity and atmospheric scattering correction modules, which can correct the impact of environmental interference on spectral data in real time; The intelligent data processing unit uses an edge computing chip to locally complete reflectivity calculation and coverage type discrimination; The low-power energy management unit includes a solar tracking module and a battery health monitoring circuit, which can dynamically adjust energy distribution according to light intensity; The multimodal data transmission unit integrates two transmission modes: Beidou short message and LoRa, and can automatically switch according to signal strength.

2. The alpine wetland surface cover type monitoring system according to claim 1 is characterized in that: The red light / near-infrared dual-band sensor array includes four red light sensors and four near-infrared sensors, which are distributed in a ring. The field of view of each sensor is 30°, and the monitoring wavelength ranges are 620-670nm and 830-870nm respectively. The sampling frequency of the sensor array can be adaptively adjusted within the range of 1-60 minutes through an intelligent data processing unit, and the sampling frequency is automatically increased when a rapid change in coverage type is detected.

3. The alpine wetland surface cover type monitoring system according to claim 1, characterized in that: The temperature correction module of the adaptive environmental calibration unit uses a platinum resistance temperature sensor with a measurement range of -40°C to 60°C and an accuracy of ±0.1°C; the humidity correction module uses a capacitive humidity sensor with a measurement range of 0-100%RH and an accuracy of ±2%RH. The atmospheric scattering correction module collects atmospheric optical thickness data in real time, establishes a scattering coefficient and spectral attenuation model, and performs band-specific correction on the original spectral data.

4. The alpine wetland surface cover type monitoring system according to claim 1, characterized in that: The intelligent data processing unit includes an FPGA chip and an ARM processor. The FPGA chip is responsible for high-speed spectral data acquisition and preprocessing, and the ARM processor is responsible for reflectivity calculation, vegetation index analysis and coverage type discrimination. The built-in coverage type discrimination model can be remotely updated via OTA, supporting dynamic identification of newly added coverage types. The data storage module uses an industrial-grade SD card with a capacity of not less than 64GB, which can store monitoring data at intervals of 15 minutes for at least 3 years.

5. The alpine wetland surface cover type monitoring system according to claim 1 is characterized in that: The formula for calculating the normalized vegetation index by the intelligent data processing unit is: Among them, α IR is the reflectivity in the near-infrared band, α R is the reflectivity of red light band, k IR is the near-infrared band environmental correction coefficient, k R is the red light band environmental correction coefficient, δ is a minimum constant with a value of 0.001, which is used to avoid the denominator being zero. The environmental correction coefficient is dynamically generated by the adaptive environmental calibration unit according to the real-time monitored temperature, humidity and atmospheric optical thickness.

6. The alpine wetland surface cover type monitoring system according to claim 1, characterized in that: The low power consumption energy management unit adopts the following energy allocation formula: Among them, P 分配 The real-time power distribution of each unit, P 总 is the total output power of the system, I is the current light intensity, I max is the maximum light intensity threshold, SOC is the battery state of charge, SOC max is the battery full charge value, T is the current ambient temperature, T opt is the optimal operating temperature, ω1, ω2, and ω3 are the weight coefficients of light, battery status, and temperature, respectively, and their sum is 1. This formula is used to achieve dynamic energy allocation and ensure that the core functions of the system run first in extreme environments.

7. The alpine wetland surface cover type monitoring system according to claim 1, characterized in that: The transmission strategy of the multimodal data transmission unit satisfies the following formula: Among them, D 传输 is the actual amount of data transmitted, D 有效 is the effective data volume to be transmitted, SNR is the current signal-to-noise ratio, SNR 阈值 is the minimum communication signal-to-noise ratio, E is the residual energy, and E 临界 is the minimum operating energy of the system, β1 and β2 are the weight coefficients of signal quality and energy state respectively, the sum of which is 1. When SNR <SNR 阈值 And E>E 临界 When SNR≥SNR 阈值 and E≤E 临界 When transmitting, keep the LoRa mode and compress the data volume.

8. The alpine wetland surface cover type monitoring system according to claim 1, characterized in that: The coverage type discrimination logic of the intelligent data processing unit adopts a three-level decision-making mechanism: The first-level decision is based on the preliminary classification of the relationship between the NDVI value and reflectivity calculated in claim 5. The second-level decision introduces a time series continuity check, and the classification is confirmed when the sampling results are consistent for three consecutive times. The third-level decision is combined with historical data comparison, and a manual review request is initiated when the deviation from the data of the same area in the past 30 days exceeds 30%. The discrimination conditions are expanded to include the type of frozen soil cover: When the red light reflectance is 0.30-0.50, the near-infrared reflectance is 0.25-0.45, the NDVI value is -0.1 to 0.0, and the soil temperature is below 0℃ for 5 consecutive days, it is judged to be frozen soil.

9. A method for monitoring the surface cover type of an alpine wetland, applied to a system for monitoring the surface cover type of an alpine wetland as claimed in any one of claims 1 to 8, characterized in that: include: S1: When the system is initialized, the adaptive environmental calibration unit collects reference environmental parameters and establishes an initial calibration model; S2: The surface multispectral radiation monitoring unit performs scanning data acquisition at preset angle intervals to obtain red light / near-infrared reflectance data at different angles; S3: Perform angle normalization on the collected raw spectral data to convert the reflectance at different angles into vertical observation equivalent values; S4: calling the intelligent data processing unit to calculate the normalized vegetation index formula to calculate the ratio of the corrected NDVI value to the reflectance; S5: Combining the three-level decision-making mechanism with the extended coverage type discrimination criteria to identify the surface coverage type; S6: The intelligent data processing unit stores the recognition results by timestamp and generates a data quality assessment report, wherein the data quality assessment report includes data integrity, correction accuracy, and classification confidence; S7: The multimodal data transmission unit dynamically adjusts the transmission strategy according to the transmission strategy formula of the multimodal data transmission unit, and sends the monitoring results and the quality assessment report to the remote terminal; S8: The low-power energy management unit adjusts the working mode of each unit, including sampling frequency, calculation accuracy, and transmission interval, according to the real-time energy status and energy allocation formula to achieve long-term battery life.

10. The method for monitoring surface cover types of alpine wetlands according to claim 9, characterized in that: The angle normalization processing in S3 is specifically as follows: by establishing a relationship model between reflectivity and observation angle, the reflectivity data of oblique observation is corrected to the equivalent value under the vertical observation angle. The terrain slope factor is introduced in the correction process. When the slope exceeds 15°, the terrain shielding correction coefficient is additionally added to ensure the comparability of spectral data under different terrain conditions. The angle normalization processing can control the monitoring data deviation in different terrain areas within 5%.