Intelligent maintenance method and system for vegetation recovery in high-altitude unmanned area

Abstract

The invention belongs to the technical field of intelligent maintenance management for vegetation recovery, and particularly relates to an intelligent maintenance method and system for vegetation recovery in a high-altitude unmanned area. Comprising the following steps: S1, collecting vegetation image data, soil humidity data, soil nutrient data and meteorological data in real time, and performing standardized preprocessing; s2, calculating to obtain current vegetation proportion data; judging that the target recovery area is an area needing to be supplemented and scattered with grass seeds; the grass seed supplementing and scattering amount is obtained through calculation; s3, determining a nonlinear mapping relation between the soil humidity and the vegetation transpiration, and a nonlinear mapping relation between the soil nutrient content and the vegetation absorptivity; training historical vegetation recovery data to obtain optimal water and fertilizer requirements of different vegetations under the current soil condition; (S4); the actual water spraying amount and the actual fertilizing amount are adjusted. And intelligent maintenance of vegetation recovery of the high-altitude unmanned area can be realized.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of vegetation restoration intelligent maintenance management, and particularly relates to an intelligent maintenance method and system for vegetation restoration in a high-altitude uninhabited area. BACKGROUND

[0002] High-altitude (altitude height ≥ 3000m) power transmission lines are usually built in high-altitude uninhabited areas, which are the core infrastructure for ensuring the delivery of clean energy (such as wind power and photovoltaic power) in the west and the power supply in remote plateau areas. During construction, the surface vegetation will inevitably be damaged. The high-altitude area has fragile characteristics such as low temperature, low oxygen, large diurnal temperature difference, strong ultraviolet rays, poor soil, and weak water and fertilizer retention capacity, and the natural recovery period of vegetation is as long as 5-10 years. If there is a lack of scientific maintenance, it is easy to cause soil erosion, land desertification and other problems, which not only threatens the stability of the tower foundation of the power transmission line, but also exacerbates regional ecological deterioration.

[0003] At present, the vegetation restoration and maintenance technology in high-altitude uninhabited areas still has many key bottlenecks, which is difficult to meet the actual needs of "high efficiency, precision and unmanned". Specifically, it includes the following aspects: First, the artificial maintenance has low feasibility and high cost. The environment in the high-altitude uninhabited area is harsh, the ecology is relatively fragile, and the transportation is closed and the material supply cycle is long.

[0004] Second, there are blind spots in remote communication and data transmission, and the precise control of vegetation restoration depends on the transmission of real-time data (such as soil moisture and vegetation coverage) and the issuance of remote instructions, but the high-altitude uninhabited area generally lacks 4G / 5G mobile communication network coverage.

[0005] Third, the utilization rate of rainwater and water fertilizer resources is low. In the high-altitude area, the precipitation is mainly in the form of rain and snow. The existing system lacks an efficient rainwater collection and storage scheme, and the rainwater loss rate is more than 60%. The dependence on external water sources further increases the maintenance cost. SUMMARY

[0006] The technical problem to be solved by the present application is to provide an intelligent maintenance method and system for vegetation restoration in a high-altitude uninhabited area, which can realize intelligent maintenance of vegetation restoration in a high-altitude uninhabited area.

[0007] The technical solution adopted by the present application to solve the technical problem is: an intelligent maintenance method for vegetation restoration in a high-altitude uninhabited area, comprising the steps of: S1, real-time collection of vegetation image data, soil moisture data, soil nutrient data and meteorological data, standardization preprocessing of the vegetation image data, soil moisture data and soil nutrient data; S2, based on the target restoration area data and the standardized vegetation image data in step S1, calculate the current vegetation percentage data; compare the current vegetation percentage data with the preset vegetation restoration target data, and when the current vegetation percentage data < the preset vegetation restoration target data, determine that the target restoration area is the area that needs to be reseeded with grass seeds; calculate the amount of grass seeds to be reseeded. S3 uses the random forest algorithm to screen key influencing factors, determine the nonlinear mapping relationship between soil moisture and vegetation transpiration, and the nonlinear mapping relationship between soil nutrient content and vegetation absorption rate; train historical vegetation restoration data to obtain the optimal water and fertilizer requirements of different vegetation under the current soil conditions. The optimal water and fertilizer requirements include the optimal watering amount, the optimal fertilizer application amount, and the optimal water-fertilizer ratio. S4; Based on the soil moisture data, soil nutrient data and meteorological data collected in real time in step S1, the nonlinear mapping relationship between soil moisture and vegetation transpiration, the nonlinear mapping relationship between soil nutrient content and vegetation absorption rate obtained in step S3, and the optimal watering amount, optimal fertilizer amount and optimal water-fertilizer ratio for different vegetation under the current soil conditions, adjust the actual watering amount and actual fertilizer amount.

[0008] Furthermore, the formula for calculating the optimal water spraying amount in step S3 is as follows: ; Where W represents the optimal water spraying rate; k1 is the soil moisture replenishment coefficient, which is determined according to soil type: 0.8 to 1.0 for sandy soil, 1.1 to 1.3 for loam, and 1.4 to 1.6 for clay. S max This represents the upper limit of suitable soil moisture for vegetation. S represents the current soil moisture sensor reading; A1 represents the area of ​​the irrigated region; f(H,T) is the high-altitude environment correction function, and its expression is: ; Where H represents altitude; T represents the diurnal temperature range.

[0009] Furthermore, the formula for calculating the optimal fertilizer application rate in step S3 is as follows: ; Where F represents the optimal fertilization rate; k2 is the nutrient absorption efficiency coefficient, which is 0.6 to 0.7 for herbaceous plants and 0.5 to 0.6 for shrubs; N i,target For the first The target content of nutrients; nutrients include nitrogen, phosphorus, and potassium; N i,current The current content of the i-th nutrient collected by the soil nutrient sensor; n represents the number of nutrient types; A2 represents the area of ​​the fertilization zone; G(V) is the vegetation growth stage correction function, and the functional expression of G(V) is: ; Where V represents the proportion of the vegetation growth cycle.

[0010] Furthermore, the formula for calculating the optimal water-fertilizer ratio in step S3 is as follows: ; Where R is the optimal water-fertilizer ratio; W represents the optimal water spraying rate; F represents the optimal fertilizer application rate; k3 is the vegetation type coefficient, with 0.8 for herbaceous plants and 1.2 for shrubs.

[0011] Furthermore, in step S1, the vegetation image data is preprocessed, including the Normalized Difference Vegetation Index (NDVI). The formula for calculating the NDVI is as follows: ; Where, ρ nir The reflectance in the near-infrared band of the image; ρ red is the reflectance of the red band in the image, with a value range of [-1, 1].

[0012] Furthermore, the formula for calculating the amount of grass seed supplementation in step S2 is as follows: ; Where Q represents the amount of grass seed supplementation; k4 is the correction coefficient for grass seed germination rate; C target Target vegetation coverage; C represents the current vegetation coverage; A represents the area of ​​the re-spreading zone; h(S) is the soil moisture influence function, where S is the current soil moisture. The function expression for h(S) is: .

[0013] Furthermore, in step S2, based on the target restoration area data and the vegetation image data from step S1, the current vegetation percentage data is calculated; specifically, this includes the following steps: S201, for the target restoration area data and the standardized vegetation image data in step S1, the improved YOLOv8 target detection algorithm is used to identify the distribution of individual vegetation and communities. The boundary between the vegetation-covered area and the bare soil area is extracted by the U-Net semantic segmentation model to obtain vegetation cover data, vegetation height data and vegetation density data. S202 uses the normalized vegetation index calculation formula to determine the vegetation growth status and identify abnormal vegetation areas by the difference in reflectance between the near-infrared band and the red band in the vegetation image, and obtains data on abnormal vegetation areas; abnormal vegetation areas include withered areas, diseased areas and insect-infested areas. S203, based on spatial analysis methods, uses vegetation cover data to subtract vegetation anomalous area data to obtain corrected current vegetation data; S204. Divide the corrected current vegetation data by the target restoration area data to obtain the current vegetation percentage data.

[0014] Furthermore, step S5 is included, which involves cross-validating the data obtained in steps S201 and S202 through data fusion.

[0015] Furthermore, the preset vegetation restoration target data in step S2 is 30%.

[0016] The intelligent maintenance system for vegetation restoration in high-altitude uninhabited areas adopts intelligent maintenance methods for vegetation restoration in high-altitude uninhabited areas, including a data acquisition and preprocessing system, a grass seeding system, an artificial intelligence integrated control system, and a water and fertilizer control system. The data acquisition and preprocessing system is used to acquire vegetation image data, soil moisture data, soil nutrient data and meteorological data in real time, and to perform standardized preprocessing on the vegetation image data, soil moisture data and soil nutrient data. The grass seed replanting system is used to calculate the current vegetation percentage data based on the target restoration area data and the standardized vegetation image data in step S1; compare the current vegetation percentage data with the preset vegetation restoration target data; when the current vegetation percentage data is less than the preset vegetation restoration target data, the target restoration area is determined to be the area that needs to be replanted with grass seeds; and calculate the amount of grass seeds to be replanted. The artificial intelligence integrated control system is used to screen key influencing factors through random forest algorithm, determine the nonlinear mapping relationship between soil moisture and vegetation transpiration, and the nonlinear mapping relationship between soil nutrient content and vegetation absorption rate; and is used to train historical vegetation restoration data to obtain the optimal water and fertilizer requirements of different vegetation under current soil conditions. The optimal water and fertilizer requirements include the optimal watering amount, the optimal fertilizer application amount, and the optimal water-fertilizer ratio. The water and fertilizer control system is used to adjust the actual watering amount and actual fertilizer application amount based on the soil moisture data, soil nutrient data and meteorological data collected in real time in step S1, the nonlinear mapping relationship between soil moisture and vegetation transpiration obtained in step S3, the nonlinear mapping relationship between soil nutrient content and vegetation absorption rate, and the optimal watering amount, optimal fertilizer application amount and optimal water-fertilizer ratio of different vegetation under the current soil conditions.

[0017] Compared with the prior art, the beneficial effects of the present invention are: the present invention provides an intelligent maintenance method and system for vegetation restoration in high-altitude uninhabited areas, which can realize intelligent maintenance of vegetation restoration in high-altitude uninhabited areas, and can also realize remote control of the intelligent maintenance system for vegetation restoration in high-altitude uninhabited areas. Attached Figure Description

[0018] Figure 1 This is the overall flowchart of the present invention; Figure 2 This is a system diagram of the present invention. Detailed Implementation

[0019] The following is in conjunction with the appendix Figure 1 and 2 The invention is further illustrated by the embodiments.

[0020] Intelligent maintenance methods for vegetation restoration in high-altitude uninhabited areas include the following steps: S1 collects vegetation image data, soil moisture data, soil nutrient data and meteorological data in real time, and performs standardized preprocessing on the vegetation image data, soil moisture data and soil nutrient data. S2, based on the target restoration area data and the standardized vegetation image data in step S1, calculate the current vegetation percentage data; compare the current vegetation percentage data with the preset vegetation restoration target data, and when the current vegetation percentage data < the preset vegetation restoration target data, determine that the target restoration area is the area that needs to be reseeded with grass seeds; calculate the amount of grass seeds to be reseeded. S3 uses the random forest algorithm to screen key influencing factors, determine the nonlinear mapping relationship between soil moisture and vegetation transpiration, and the nonlinear mapping relationship between soil nutrient content and vegetation absorption rate; train historical vegetation restoration data to obtain the optimal water and fertilizer requirements of different vegetation under the current soil conditions. The optimal water and fertilizer requirements include the optimal watering amount, the optimal fertilizer application amount, and the optimal water-fertilizer ratio. S4; Based on the soil moisture data, soil nutrient data and meteorological data collected in real time in step S1, the nonlinear mapping relationship between soil moisture and vegetation transpiration, the nonlinear mapping relationship between soil nutrient content and vegetation absorption rate obtained in step S3, and the optimal watering amount, optimal fertilizer amount and optimal water-fertilizer ratio for different vegetation under the current soil conditions, adjust the actual watering amount and actual fertilizer amount.

[0021] Specifically, the formula for calculating the optimal water spraying amount in step S3 is as follows: ; Where W represents the optimal water spraying volume, in liters (L). k1 is the soil moisture replenishment coefficient, which is determined according to soil type: 0.8 to 1.0 for sandy soil, 1.1 to 1.3 for loam, and 1.4 to 1.6 for clay. S max This represents the upper limit of suitable soil moisture for vegetation. S represents the current soil moisture sensor reading; A1 represents the irrigated area, in m². 2 ; f(H,T) is the high-altitude environment correction function, and its expression is: ; Where H represents altitude, in km; T represents the diurnal temperature range, measured in °C.

[0022] Specifically, the formula for calculating the optimal fertilizer application rate in step S3 is as follows: ; Where F represents the optimal fertilizer application rate, in L; k2 is the nutrient absorption efficiency coefficient, which is 0.6 to 0.7 for herbaceous plants and 0.5 to 0.6 for shrubs; N i,target For the first The target content of nutrients; nutrients include nitrogen, phosphorus, and potassium; N i,current The current content of the i-th nutrient collected by the soil nutrient sensor; n represents the number of nutrient types; A2 represents the area of ​​the fertilization zone, in m². 2 ; G(V) is the vegetation growth stage correction function, and the functional expression of G(V) is: ; Wherein, V represents the proportion of vegetation growth cycle, with a value range of 0 to 1.

[0023] Specifically, the formula for calculating the optimal water-fertilizer ratio in step S3 is as follows: ; Where R is the optimal water-fertilizer ratio; W represents the optimal water spraying rate, in L; F represents the optimal fertilizer application rate, in L; k3 is the vegetation type coefficient, with 0.8 for herbaceous plants and 1.2 for shrubs.

[0024] Specifically, step S1 involves preprocessing the vegetation image data, including the Normalized Difference Vegetation Index (NDVI). The formula for calculating the NDVI is as follows: ; Where, ρ nir The reflectance in the near-infrared band of the image; ρ red is the reflectance of the red band in the image, with a value range of [-1, 1].

[0025] Specifically, the formula for calculating the amount of grass seed supplementation in step S2 is as follows: ; Where Q represents the amount of grass seed supplementation, in liters (L). k4 is the correction coefficient for grass seed germination rate; C target Target vegetation coverage; C represents the current vegetation coverage; A represents the area of ​​the respreading zone, in meters (m²). 2 ; h(S) is the soil moisture influence function, where S is the current soil moisture. The function expression for h(S) is: .

[0026] Specifically, in step S2, the current vegetation percentage is calculated based on the target restoration area data and the vegetation image data from step S1; this includes the following steps: S201, for the target restoration area data and the standardized vegetation image data in step S1, the improved YOLOv8 target detection algorithm is used to identify the distribution of individual vegetation and communities. The boundary between the vegetation-covered area and the bare soil area is extracted by the U-Net semantic segmentation model to obtain vegetation cover data, vegetation height data and vegetation density data. S202 uses the Normalized Difference Vegetation Index (NDVI) calculation formula to determine the vegetation growth status (such as leaf chlorophyll content) by the difference in reflectance between the near-infrared band and the red band in the vegetation image, identify abnormal vegetation areas, and obtain data on abnormal vegetation areas; abnormal vegetation areas include withered areas, diseased areas, and insect-infested areas. S203, based on spatial analysis methods, uses vegetation cover data to subtract vegetation anomalous area data to obtain corrected current vegetation data; S204. Divide the corrected current vegetation data by the target restoration area data to obtain the current vegetation percentage data.

[0027] Specifically, it also includes step S5, which involves cross-validating the data obtained in steps S201 and S202 through data fusion; When there is a deviation between the vegetation growth status and the optimal water and fertilizer requirements, the backup decision tree model is automatically adjusted, and secondary corrections are made in combination with the special conditions of possible short-term drought, soil compaction, and snow and ice in high-altitude areas, so as to generate precise control commands.

[0028] Specifically, the preset vegetation restoration target data in step S2 is 30%.

[0029] The intelligent maintenance system for vegetation restoration in high-altitude uninhabited areas adopts intelligent maintenance methods for vegetation restoration in high-altitude uninhabited areas, including a data acquisition and preprocessing system, a grass seeding system, an artificial intelligence integrated control system, and a water and fertilizer control system. The data acquisition and preprocessing system is used to acquire vegetation image data, soil moisture data, soil nutrient data and meteorological data in real time, and to perform standardized preprocessing on the vegetation image data, soil moisture data and soil nutrient data.

[0030] Specifically, the data acquisition and preprocessing system includes a high-definition camera, a humidity sensor, a fertility sensor, a meteorological data logger, and a data collection and feedback terminal. The high-definition camera is used to acquire vegetation image data. The humidity sensor is used to acquire soil moisture data. The fertility sensor is used to acquire soil nutrient data. The meteorological data logger is connected to the meteorological department's database to acquire meteorological data. The data collection and feedback terminal performs standardized preprocessing on the vegetation image data, soil moisture data, and soil nutrient data.

[0031] The grass seed replanting system is used to calculate the current vegetation percentage data based on the target restoration area data and the standardized vegetation image data in step S1; compare the current vegetation percentage data with the preset vegetation restoration target data, and determine the target restoration area as the area that needs to be replanted with grass seeds when the current vegetation percentage data is less than the preset vegetation restoration target data; and calculate the amount of grass seed replanting.

[0032] Specifically, the grass seed application system includes a grass seed application control unit, a grass seed storage bin, multiple grass seed applicators, and multiple discharge control valves. Each grass seed applicator is connected to the grass seed storage bin via a grass seed delivery pipe, and each grass seed delivery pipe is equipped with a discharge control valve. Multiple grass seed applicators are evenly distributed above the soil in the target restoration area. Opening any one or more discharge control valves moves grass seeds from the grass seed storage bin to the grass seed applicator for application. The discharge control valves are solenoid valves. The grass seed application control unit determines whether the target restoration area requires grass seed application, calculates the amount of grass seed to be applied, and controls the opening and closing of the discharge control valves.

[0033] The artificial intelligence integrated control system is used to screen key influencing factors through random forest algorithm, determine the nonlinear mapping relationship between soil moisture and vegetation transpiration, and the nonlinear mapping relationship between soil nutrient content and vegetation absorption rate; and is used to train historical vegetation restoration data to obtain the optimal water and fertilizer requirements of different vegetation under current soil conditions. The optimal water and fertilizer requirements include the optimal watering amount, the optimal fertilizer application amount, and the optimal water-fertilizer ratio.

[0034] The AI-integrated control system is controlled via a remote control terminal. This remote control terminal can be a smartphone or tablet, etc.

[0035] The water and fertilizer control system is used to adjust the actual watering amount and actual fertilizer application amount based on the soil moisture data, soil nutrient data and meteorological data collected in real time in step S1, the nonlinear mapping relationship between soil moisture and vegetation transpiration obtained in step S3, the nonlinear mapping relationship between soil nutrient content and vegetation absorption rate, and the optimal watering amount, optimal fertilizer application amount and optimal water-fertilizer ratio of different vegetation under the current soil conditions.

[0036] Specifically, the water and fertilizer control system includes a water and fertilizer controller, a rainwater harvesting device, and a fertilizer storage device. Both the rainwater harvesting device and the fertilizer storage device are connected to the main irrigation pipe via drainage pipes. The main irrigation pipe is connected to irrigation branch pipes, which are staggered and shallowly buried in the soil within the target restoration area. Multiple drip irrigation nozzles with built-in solenoid valves are installed on the irrigation branch pipes. The water and fertilizer controller calculates the nonlinear mapping relationship between soil moisture and vegetation transpiration, the nonlinear mapping relationship between soil nutrient content and vegetation absorption rate, as well as the actual watering volume and actual fertilizer application volume. The rainwater harvesting device can be the device described in Chinese patent documents with application numbers CN201320761539.X, CN201320761539.X, etc. The fertilizer storage device can be a fertilizer storage tank, fertilizer storage box, etc. The fertilizer storage device is used to store liquid fertilizer.

[0037] This system is powered by a power supply system.

[0038] The specific embodiments described are preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. All equivalent changes made to the structure, shape, and principle of the present invention should be covered within the scope of protection of the present invention.

Claims

1. An intelligent maintenance method for vegetation restoration in high-altitude uninhabited areas, characterized in that, Including the following steps: S1 collects vegetation image data, soil moisture data, soil nutrient data and meteorological data in real time, and performs standardized preprocessing on the vegetation image data, soil moisture data and soil nutrient data. S2, based on the target restoration area data and the standardized vegetation image data in step S1, calculate the current vegetation percentage data; compare the current vegetation percentage data with the preset vegetation restoration target data, and when the current vegetation percentage data < the preset vegetation restoration target data, determine that the target restoration area is the area that needs to be reseeded with grass seeds; calculate the amount of grass seeds to be reseeded. S3 uses the random forest algorithm to screen key influencing factors, determine the nonlinear mapping relationship between soil moisture and vegetation transpiration, and the nonlinear mapping relationship between soil nutrient content and vegetation absorption rate; train historical vegetation restoration data to obtain the optimal water and fertilizer requirements of different vegetation under the current soil conditions. The optimal water and fertilizer requirements include the optimal watering amount, the optimal fertilizer application amount, and the optimal water-fertilizer ratio. S4; Based on the soil moisture data, soil nutrient data and meteorological data collected in real time in step S1, the nonlinear mapping relationship between soil moisture and vegetation transpiration, the nonlinear mapping relationship between soil nutrient content and vegetation absorption rate obtained in step S3, and the optimal watering amount, optimal fertilizer amount and optimal water-fertilizer ratio for different vegetation under the current soil conditions, adjust the actual watering amount and actual fertilizer amount.

2. The intelligent maintenance method for vegetation restoration in high-altitude uninhabited areas as described in claim 1, characterized in that, The formula for calculating the optimal water spraying amount in step S3 is: ； Where W represents the optimal water spraying rate; k1 is the soil moisture replenishment coefficient, which is determined according to soil type: 0.8 to 1.0 for sandy soil, 1.1 to 1.3 for loam, and 1.4 to 1.6 for clay. S max This represents the upper limit of suitable soil moisture for vegetation. S represents the current soil moisture sensor reading; A1 represents the area of ​​the irrigated region; f(H,T) is the high-altitude environment correction function, and its expression is: ； Where H represents altitude; T represents the diurnal temperature range.

3. The intelligent maintenance method for vegetation restoration in high-altitude uninhabited areas as described in claim 1, characterized in that, The formula for calculating the optimal fertilization amount in step S3 is: ； Where F represents the optimal fertilization rate; k2 is the nutrient absorption efficiency coefficient, which is 0.6 to 0.7 for herbaceous plants and 0.5 to 0.6 for shrubs; N i,target For the first The target content of nutrients; nutrients include nitrogen, phosphorus, and potassium; N i,current The current content of the i-th nutrient collected by the soil nutrient sensor; n represents the number of nutrient types; A2 represents the area of ​​the fertilization zone; G(V) is the vegetation growth stage correction function, and the functional expression of G(V) is: ； Where V represents the proportion of the vegetation growth cycle.

4. The intelligent maintenance method for vegetation restoration in high-altitude uninhabited areas as described in claim 1, characterized in that, The formula for calculating the optimal water-fertilizer ratio in step S3 is as follows: ； Where R is the optimal water-fertilizer ratio; W represents the optimal water spraying rate; F represents the optimal fertilizer application rate; k3 is the vegetation type coefficient, with 0.8 for herbaceous plants and 1.2 for shrubs.

5. The intelligent maintenance method for vegetation restoration in high-altitude uninhabited areas as described in claim 1, characterized in that, In step S1, the vegetation image data is preprocessed, including the Normalized Difference Vegetation Index (NDVI). The formula for calculating the NDVI is as follows: ； Where, ρ nir The reflectance in the near-infrared band of the image; ρ red is the reflectance of the red band in the image, with a value range of [-1, 1].

6. The intelligent maintenance method for vegetation restoration in high-altitude uninhabited areas as described in claim 1, characterized in that, The formula for calculating the amount of grass seed supplementation in step S2 is as follows: ； Where Q represents the amount of grass seed supplementation; k4 is the correction coefficient for grass seed germination rate; C target Target vegetation coverage; C represents the current vegetation coverage; A represents the area of ​​the re-spreading zone; h(S) is the soil moisture influence function, where S is the current soil moisture. The function expression for h(S) is: 。 7. The intelligent maintenance method for vegetation restoration in high-altitude uninhabited areas as described in claim 1, characterized in that, In step S2, the current vegetation percentage is calculated based on the target restoration area data and the vegetation image data from step S1; specifically, this includes the following steps: S201, for the target restoration area data and the standardized vegetation image data in step S1, the improved YOLOv8 target detection algorithm is used to identify the distribution of individual vegetation and communities. The boundary between the vegetation-covered area and the bare soil area is extracted by the U-Net semantic segmentation model to obtain vegetation cover data, vegetation height data and vegetation density data. S202 uses the normalized vegetation index calculation formula to determine the vegetation growth status and identify abnormal vegetation areas by the difference in reflectance between the near-infrared band and the red band in the vegetation image, and obtains data on abnormal vegetation areas; abnormal vegetation areas include withered areas, diseased areas and insect-infested areas. S203, based on spatial analysis methods, uses vegetation cover data to subtract vegetation anomalous area data to obtain corrected current vegetation data; S204. Divide the corrected current vegetation data by the target restoration area data to obtain the current vegetation percentage data.

8. The intelligent maintenance method for vegetation restoration in high-altitude uninhabited areas as described in claim 7, characterized in that, It also includes step S5, which involves cross-validating the data obtained in steps S201 and S202 through data fusion.

9. The intelligent maintenance method for vegetation restoration in high-altitude uninhabited areas as described in claim 1, characterized in that, The preset vegetation restoration target data in step S2 is 30%.

10. An intelligent maintenance system for vegetation restoration in high-altitude uninhabited areas, employing the intelligent maintenance method for vegetation restoration in high-altitude uninhabited areas as described in any one of claims 1-9, characterized in that, This includes a data acquisition and preprocessing system, a grass seed application system, an artificial intelligence integrated control system, and a water and fertilizer control system; The data acquisition and preprocessing system is used to acquire vegetation image data, soil moisture data, soil nutrient data and meteorological data in real time, and to perform standardized preprocessing on the vegetation image data, soil moisture data and soil nutrient data. The grass seed replanting system is used to calculate the current vegetation percentage data based on the target restoration area data and the standardized vegetation image data in step S1; compare the current vegetation percentage data with the preset vegetation restoration target data; when the current vegetation percentage data is less than the preset vegetation restoration target data, the target restoration area is determined to be the area that needs to be replanted with grass seeds; and calculate the amount of grass seeds to be replanted. The artificial intelligence integrated control system is used to screen key influencing factors through random forest algorithm, determine the nonlinear mapping relationship between soil moisture and vegetation transpiration, and the nonlinear mapping relationship between soil nutrient content and vegetation absorption rate; and is used to train historical vegetation restoration data to obtain the optimal water and fertilizer requirements of different vegetation under current soil conditions. The optimal water and fertilizer requirements include the optimal watering amount, the optimal fertilizer application amount, and the optimal water-fertilizer ratio. The water and fertilizer control system is used to adjust the actual watering amount and actual fertilizer application amount based on the soil moisture data, soil nutrient data and meteorological data collected in real time in step S1, the nonlinear mapping relationship between soil moisture and vegetation transpiration obtained in step S3, the nonlinear mapping relationship between soil nutrient content and vegetation absorption rate, and the optimal watering amount, optimal fertilizer application amount and optimal water-fertilizer ratio of different vegetation under the current soil conditions.

Images