Highly integrated inversion-based natural grassland productivity precision monitoring method and device

Abstract

The application discloses a kind of natural grassland productivity precision monitoring method and equipment based on height fusion inversion.The method comprises the following steps: in response to a monitoring instruction, distance observation data of a target herbaceous plant collected by a ranging device is obtained;Based on the geometric relationship between the ranging device and the target plant, the average height data of multiple plants in the target area is calculated using a processor in combination with the preset point information on the target plant and the empirical height equation;The height and the multispectral image features in the target area are fused, and a preset productivity fusion inversion model is used to generate natural grassland productivity data for the target area.The application realizes high-precision, near-real-time and low-cost monitoring of natural grassland productivity by fusing vegetation physical structure parameters and multispectral ecological information, providing reliable technical support for grassland ecological health assessment and fine management of animal husbandry.

Description

Technical Field

[0001] This application relates to the fields of intelligent sensing technology and grassland ecological monitoring technology, specifically to a method, monitoring device, electronic device, computer-readable storage medium, and computer program product for accurate monitoring of natural grassland vegetation productivity based on distance measurement, height estimation, and multispectral fusion. Background Technology

[0002] Accurately grasping the productivity of natural grasslands and its spatiotemporal dynamics is of great significance for grassland ecological protection, livestock carrying capacity assessment, rotational grazing regulation, and forage resource management. With the rapid growth in demand for smart ranches and precision grazing management, ranch monitoring technologies such as virtual fencing and remote water source monitoring have made progress in improving grazing efficiency, but they also place higher demands on high-frequency, low-cost, and scalable monitoring of grassland productivity.

[0003] Existing grassland vegetation productivity estimation techniques mainly include traditional ground quadrat sampling, satellite remote sensing inversion, and UAV remote sensing. While traditional ground quadrat sampling offers high accuracy, it is labor-intensive, has limited representativeness, and struggles to support large-scale and long-term continuous monitoring. Satellite multispectral and hyperspectral remote sensing images typically sample spectral indices such as Normalized Difference Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI), or combine spectral data with mechanistic models to estimate productivity. Although these methods overcome the time-consuming and labor-intensive nature of traditional ground sampling and achieve large-scale coverage, they still have limitations in practical applications.

[0004] (1) Temporal resolution and meteorological conditions: The inversion accuracy is easily affected by the revisit period and atmospheric conditions, making it difficult to achieve continuous high-frequency monitoring at the pasture scale;

[0005] (2) Lack of physical structural parameters: Existing spectral inversion models have difficulty directly obtaining key three-dimensional structural parameters such as vegetation height. Plant height, density and other structural parameters are the direct physical basis of biomass. Their lack leads to a theoretical bottleneck in the model when estimating productivity, which limits the further improvement of accuracy;

[0006] (3) Insufficient support for refined management: Limited ability to adapt to scale and respond in real time for refined management units such as fenced zones and rotational grazing areas.

[0007] In recent years, integrated "space-air-ground" monitoring technology combined with artificial intelligence has become a key focus in grassland ecological monitoring. The academic and industrial communities have conducted extensive research on productivity estimation based on the fusion of spectral and structural information. Related studies have used LiDAR, terrestrial laser scanning (TLS / handheld LiDAR), and UAV multispectral / hyperspectral methods to extract three-dimensional structural indicators such as vegetation canopy height and density, and combined these with machine learning models to improve the accuracy of grassland productivity estimation. However, these technologies still face significant limitations in the routine and operational monitoring of natural grasslands. On the one hand, while LiDAR can acquire detailed three-dimensional structures, the equipment is expensive and data processing is complex, making large-scale deployment difficult in pastures and similar settings. On the other hand, UAV systems are constrained by cost, airspace, weather conditions, and long data processing chains, making it difficult to meet the needs of routine monitoring of natural grasslands throughout their entire growth cycle, in near real-time, at low cost, and easily deployable.

[0008] Meanwhile, near-surface cameras (PhenoCam / near-surface camera networks) have developed rapidly in recent years. Because they can provide continuous images with high temporal resolution, they serve as an important bridge connecting terrestrial sample plots and satellite remote sensing. However, in practical livestock applications, they face challenges such as normalizing light variations and insufficient coupling with physical structure parameters. Simply relying on vegetation indices and phenological indicators is insufficient for accurate productivity calculation. Furthermore, while existing multifunctional hyperspectral monitoring equipment can output various vegetation indices and phenological information, it often struggles to simultaneously acquire key physical quantities highly correlated with productivity (such as plant height, site height, canopy thickness, or community structure parameters) in natural grassland scenarios, resulting in a gap between "spectral observation and productivity calculation."

[0009] Therefore, there is an urgent need for a vegetation productivity monitoring technology for natural grassland applications that can achieve low-cost, automated, sustainable, and high-frequency quantitative monitoring and integrated collection of key physical structural parameters of vegetation and multispectral ecological information at the pasture scale, thereby providing a reliable data foundation for grassland ecological health assessment, early warning, and refined grazing management. Summary of the Invention

[0010] In view of this, this application provides a method, monitoring device, electronic device, computer-readable storage medium and computer program product for precise monitoring of natural grassland vegetation productivity based on highly fusion inversion, so as to achieve high-precision, near real-time monitoring of natural grassland vegetation productivity.

[0011] One aspect of this application provides a method for precise monitoring of natural grassland productivity based on high-level fusion inversion, comprising: responding to a monitoring command, acquiring distance observation data within a target area collected by a ranging device, wherein the natural grassland within the target area contains multiple plants, and the distance observation data represents the distance between the ranging device and a target plant among the multiple plants; generating average height data of the multiple plants within the target area based on the distance observation data, target point information, and an empirical height formula, wherein the target point information represents a preset point on the target plant; and generating productivity data of the target area using a vegetation productivity fusion inversion model based on the average height data and a multispectral image of the target area. The vegetation productivity fusion inversion model combines the average height data with multispectral ecological indicators to establish a mapping relationship between vegetation physical structure parameters and productivity, thereby improving the accuracy of productivity estimation.

[0012] The present invention also provides corresponding monitoring devices, electronic devices, computer-readable storage media, and computer program products to implement the method.

[0013] According to an embodiment of this application, a processor generates productivity data for the target area based on the average height data and a multispectral image of the target area. This includes: performing data parsing processing on the multispectral image to obtain multiple observation indicators for the target area, wherein the multiple observation indicators characterize different vegetation ecological information within the target area; and generating the productivity data based on a preset productivity calculation formula, according to the average height data and the multiple observation indicators.

[0014] According to embodiments of this application, multiple observation indicators include phenological indicators, leaf area indicators, and cover indicators.

[0015] According to an embodiment of this application, the productivity data is generated based on a preset productivity calculation formula, according to the average height data and multiple observation indicators, including: calculating the productivity data based on the preset productivity calculation formula, according to the phenological indicators, the leaf area indicators, the coverage indicators, the average height data and multiple preset coefficients.

[0016] According to an embodiment of this application, the above-mentioned preset productivity calculation formula is as shown in formula (1):

[0017] (1)

[0018] in, Fresh weight, i.e., the aforementioned productivity data, This is the average height data. For coverage metrics, Leaf area index As a phenological indicator, , , , , All are preset coefficients.

[0019] According to an embodiment of this application, generating productivity data for the target area based on the aforementioned average height data and the multispectral image of the target area includes: processing the aforementioned average height data and the multispectral image of the target area using a natural grassland vegetation productivity fusion inversion model to obtain the aforementioned productivity data.

[0020] According to an embodiment of this application, the above-mentioned natural grassland vegetation productivity fusion inversion model is trained in the following manner: A training sample set corresponding to at least one training region is obtained, wherein the training sample set includes multiple training samples and label data corresponding to each training sample; the training samples include training height data and training multispectral images of the training region; the label data is productivity information obtained from actual field measurements; for each training sample, data parsing processing is performed on the training multispectral images to obtain multiple training indicators, wherein the multiple training indicators characterize different vegetation ecological information within the training region; the training samples and the multiple training indicators are input into an initial neural network to output predicted productivity information; the predicted productivity information and the label data corresponding to the training samples are input into a loss function to output a target loss result; the model parameters of the initial neural network are iteratively adjusted according to the target loss result to obtain the trained natural grassland vegetation productivity fusion inversion model.

[0021] The natural grassland vegetation productivity fusion inversion model is used to establish a nonlinear mapping relationship between physical structure parameters, multispectral ecological characteristics and measured productivity with average vegetation height as the core. Its training and inference processes are based on observable physical structure and ecological parameter inputs.

[0022] According to an embodiment of this application, the aforementioned distance observation data includes target horizontal distance information, target oblique distance, and target height information. The target horizontal distance information represents the horizontal distance between the ranging device and the target plant, the target oblique distance represents the oblique distance between the ranging device and the target location information, and the target height information represents the vertical distance between the ranging device and the ground.

[0023] According to an embodiment of this application, based on an empirical height formula, and based on the aforementioned distance observation data and target location information, average height data of multiple plants within the target area is generated, including: calculating point height information based on the aforementioned target location information, according to the aforementioned target height information, the aforementioned target horizontal distance information, and the aforementioned target diagonal distance, wherein the aforementioned point height information represents the distance between the aforementioned target location information and the ground; and calculating the aforementioned average height data based on the aforementioned empirical height formula and the aforementioned point height information.

[0024] According to an embodiment of this application, the above empirical height formula is as shown in formula (2):

[0025]

[0026] (2)

[0027] in, This is the average height data. For point height information, For target height information, The diagonal distance to the target. B is the target horizontal distance information, which is a coefficient determined based on the proportion of the target point's distance from the ground to the height of the target plant. A and C are preset coefficients for the relationship between the height of the target plant and its average height.

[0028] Another aspect of this application provides a natural grassland vegetation productivity monitoring device, comprising: an acquisition module, configured to acquire distance observation data collected by a ranging device within a target area in response to a monitoring command, wherein multiple plants are distributed within the target area, and the distance observation data characterizes the distance between the ranging device and a target plant among the multiple plants; a first generation module, configured to generate average height data of multiple plants within the target area based on an empirical height formula, using a processor according to the distance observation data and target point information, wherein the target point information characterizes a preset point on the target plant; and a second generation module, configured to generate productivity data of the target area based on the average height data and a multispectral image of the target area.

[0029] Another aspect of this application provides an electronic device comprising: one or more processors; and a memory for storing one or more programs, wherein, when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to perform the method as described above.

[0030] Another aspect of this application provides a computer-readable storage medium storing computer-executable instructions that, when executed, are used to implement the method described above.

[0031] Another aspect of this application provides a computer program product comprising computer-executable instructions which, when executed, are used to implement the method described above.

[0032] According to embodiments of this application, distance observation data within the target area and preset points on the target vegetation are collected by a ranging device. The average height data of the target area can be obtained by processing the distance observation data and target point information based on an empirical height formula. Therefore, productivity data of the target area can be generated based on the average height data and a multispectral image of the target area. Because the average height data and multispectral image of the target area are fully considered from both physical and ecological perspectives when calculating productivity data, the productivity data of the target area can be accurately calculated. This provides a timely and accurate data foundation for early warning of natural grassland ecological health and the formulation of forage utilization policies, contributing to the healthy development of grassland ecosystems. Attached Figure Description

[0033] The above and other objects, features and advantages of this application will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0034] Figure 1 An exemplary system architecture for monitoring natural grassland vegetation productivity according to embodiments of this application is shown;

[0035] Figure 2 A flowchart of a method for monitoring the productivity of natural grassland vegetation according to an embodiment of this application is shown;

[0036] Figure 3 A schematic diagram of training multispectral images according to an embodiment of this application is shown;

[0037] Figure 4 A schematic diagram illustrating the calculation of average height data according to an embodiment of this application is shown;

[0038] Figure 5 A schematic diagram of the equipment used in the natural grassland vegetation productivity monitoring method according to an embodiment of this application is shown;

[0039] Figure 6 A block diagram of a natural grassland vegetation productivity monitoring device according to an embodiment of this application is shown;

[0040] Figure 7 A block diagram of an electronic device suitable for implementing the methods described above, according to an embodiment of this application, is shown. Detailed Implementation

[0041] The embodiments of this application will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of this application. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of this application for ease of explanation. However, it will be apparent that one or more embodiments may be implemented without these specific details. Furthermore, descriptions of well-known structures and technologies are omitted in the following description to avoid unnecessarily obscuring the concepts of this application.

[0042] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of this application. The terms “comprising”, “including”, etc., as used herein indicate the presence of the stated features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0043] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0044] When using expressions such as "at least one of A, B and C", they should generally be interpreted in accordance with the meaning that is commonly understood by those skilled in the art (e.g., "a system having at least one of A, B and C" should include, but is not limited to, a system having A alone, a system having B alone, a system having C alone, a system having A and B, a system having A and C, a system having B and C, and / or a system having A, B and C, etc.).

[0045] Currently, there are problems such as insufficient algorithms for full-cycle productivity estimation at the scale of natural grasslands and pastures, and excessively long inversion intervals. Firstly, most studies utilize remote sensing indices for productivity inversion, such as NDPI, NDVI, EVI, and SAVI vegetation indices. These can effectively estimate the growth status of natural grassland vegetation over large areas and diverse climatic conditions. However, their accuracy often depends on years of accumulated climate data and the selection of environmental variables, and the time scale is limited by the acquisition cycle of the remote sensing indices. Based on this, remote sensing-process coupling methods strive to combine remote sensing data with mechanistic models. While this effectively improves the accuracy of biomass estimation, in practical applications, optimizing model parameters is difficult, and the complexity of the model often leads to decreased computational efficiency. In recent years, machine learning methods, by combining multiple data sources, have shown good predictive capabilities, but the black-box nature of their models also brings unexplained defects.

[0046] Currently, intelligent hyperspectral and productivity instruments are generally expensive and lack a direct pathway from observation to productivity measurement. For example, current multifunctional spectroscopic phenological monitoring cameras can generate RGB images, narrow-band spectrograms, and NDVI composite images, while simultaneously recording vegetation indices such as NDVI, GCC, RCC, BCC, and GVI (GI). However, they cannot measure physical indicators such as distance and thickness, thus failing to obtain physical index data related to vegetation growth, which are crucial factors directly influencing plant growth, such as plant height and density. Furthermore, they cannot directly measure canopy productivity, and models considering vegetation physical indices cannot be implemented using multifunctional spectroscopic phenological monitoring cameras.

[0047] In view of this, embodiments of this application provide a vegetation productivity monitoring method, a vegetation productivity monitoring device, and an electronic device. The method includes, in response to a monitoring command, acquiring distance observation data within a target area collected by a ranging device, wherein multiple plants are distributed in the natural grassland within the target area, and the distance observation data characterizes the distance between the ranging device and a target plant among the multiple plants; based on an empirical height formula, using a processor to generate average height data of multiple plants within the target area according to the distance observation data and target point information, wherein the target point information characterizes a preset point on the target plant; and generating productivity data of the target area based on the average height data and a multispectral image of the target area.

[0048] In the embodiments of this application, the collection, updating, analysis, processing, use, transmission, provision, disclosure, and storage of data (e.g., including but not limited to user personal information) comply with relevant laws and regulations, are used for legitimate purposes, and do not violate public order and good morals. In particular, necessary measures have been taken to prevent unauthorized access to user personal information data and to safeguard user personal information security and network security.

[0049] Figure 1 An exemplary system architecture 100 for applying vegetation productivity monitoring methods according to embodiments of this application is shown. It should be noted that... Figure 1 The examples shown are merely examples of system architectures that can be applied to the embodiments of this application, in order to help those skilled in the art understand the technical content of this application, but do not mean that the embodiments of this application cannot be used in other devices, systems, environments or scenarios.

[0050] like Figure 1As shown, the system architecture 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 serves as a medium for providing communication links between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired and / or wireless communication links, etc.

[0051] Users can use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 via the network 104 to receive or send messages, etc. Various communication client applications can be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as web browser applications, search applications, instant messaging tools, and / or email clients, etc. (for example only).

[0052] The first terminal device 101, the second terminal device 102, and the third terminal device 103 can be various electronic devices with displays and support web browsing, including but not limited to smartphones, tablets, laptops, and desktop computers.

[0053] Server 105 can be a server that provides various services, such as a backend management server that supports websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103 (this is just an example). The backend management server can analyze and process data such as received user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal devices.

[0054] It should be noted that the natural grassland vegetation productivity monitoring method provided in this application embodiment can generally be executed by server 105. Correspondingly, the natural grassland vegetation productivity monitoring device provided in this application embodiment can generally be installed in server 105. The natural grassland vegetation productivity monitoring method provided in this application embodiment can also be executed by a server or server cluster that is different from server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or server 105. Correspondingly, the natural grassland vegetation productivity monitoring device provided in this application embodiment can also be installed in a server or server cluster that is different from server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or server 105. Alternatively, the natural grassland vegetation productivity monitoring method provided in this application embodiment can also be executed by the first terminal device 101, the second terminal device 102, or the third terminal device 103, or by other terminal devices different from the first terminal device 101, the second terminal device 102, or the third terminal device 103. Accordingly, the natural grassland vegetation productivity monitoring device provided in this application embodiment can also be installed in the first terminal device 101, the second terminal device 102 or the third terminal device 103, or in other terminal devices different from the first terminal device 101, the second terminal device 102 or the third terminal device 103.

[0055] It should be understood that Figure 1 The number of terminal devices, networks, and servers is chosen solely based on implementation needs, and there can be any number of terminal devices, networks, and servers.

[0056] Figure 2 A flowchart of a method for monitoring the productivity of natural grassland vegetation according to an embodiment of this application is shown.

[0057] like Figure 2 As shown, the method for monitoring the productivity of natural grassland vegetation, performed by electronic equipment, includes operations S201 to S203.

[0058] In operation S201, in response to the monitoring command, distance observation data of the target area collected by the ranging device is acquired. The target area contains a natural grassland with multiple plants, and the distance observation data represents the distance between the ranging device and the target plant among the multiple plants.

[0059] In operation S202, based on the empirical height formula, the processor generates the average height data of multiple plants in the target area according to the distance observation data and target point information. The target point information represents the preset points on the target plants.

[0060] In operation S203, natural grassland vegetation productivity data for the target area is generated based on average height data and multispectral images of the target area.

[0061] According to an embodiment of this application, the monitoring instruction can be a corresponding operation entered by staff on an electronic device such as a mobile phone or computer, and the electronic device automatically generates the monitoring instruction in response to the operation.

[0062] According to embodiments of this application, the target area can be a place in animal husbandry that provides food for livestock such as cattle and sheep, for example, setting up fences on natural grasslands for rotational grazing, where various types of forage grasses can be distributed. The ranging device can be a laser ranging device equipped with a three-dimensional tilt sensor. The preset point is a representative measurement point pre-selected along the height direction of the aboveground part of the target plant, preferably located within the range of 1 / 2 to 4 / 5 of the target plant's height. Through calibration, a stable proportional relationship exists between this point and the overall height of the plant, and this proportional relationship is used to determine the coefficient B in the empirical height formula.

[0063] According to this embodiment, when the electronic device responds to the monitoring command, the ranging and acquisition device acquires distance observation data of the target area, including at least the horizontal distance and oblique distance of the target plant, as well as the distance of the ranging device relative to the ground.

[0064] According to an embodiment of this application, the processor processes distance observation data based on an empirical height formula and preset points set on the target plant to obtain the average height data of the vegetation in the target area. Based on the calculated average height data and the multispectral image of the target area collected by the plant multispectral instrument installed in the target area, the productivity data of the plants in the target area can be obtained. From the productivity data, it can be seen that the target area can provide the quantity or weight of food.

[0065] According to embodiments of this application, distance observation data within the target area and preset points on the target vegetation are collected by a ranging device. The average height data of the target area can be obtained by processing the distance observation data and target point information based on an empirical height formula. Therefore, productivity data of the target area can be generated based on the average height data and a multispectral image of the target area. Because the average height data and multispectral image of the target area are fully considered from both physical and ecological perspectives when calculating productivity data, the productivity data of the target area can be accurately calculated. This provides a timely and accurate data foundation for grassland ecological health early warning and forage utilization policy formulation, contributing to the healthy development of the grassland ecosystem.

[0066] According to an embodiment of this application, a processor is used to generate productivity data of a target area based on average height data and a multispectral image of the target area. This includes: performing data parsing processing on the multispectral image to obtain multiple observation indicators of the target area, wherein the multiple observation indicators characterize different vegetation ecological information within the target area; and generating productivity data based on a preset productivity calculation formula, according to the average height data and the multiple observation indicators.

[0067] According to an embodiment of this application, the multispectral image of the target area is further analyzed to extract multispectral ecological indicators such as phenological indicators, leaf area index, and vegetation coverage. The types of observation indicators are shown in Table 1.

[0068] Table 1

[0069] Observation indicators Indicator Explanation GVI Green vegetation index NDVI Normalized Difference Vegetation Index ROI_NDVI Normalized Difference Vegetation Index of Area of ​​Interest RCC Phenological index: R / (R+G+B) GCC Phenological index: G / (R+G+B) BCC Phenological index: B / (R+G+B) FVC Vegetation coverage (%) LAI Leaf area index, a key biophysical parameter characterizing vegetation canopy structure. L Laser ranging distance (m) T Temperature (°C) H Humidity (°)

[0070] R, G, and B are the color values ​​for the red (R), green (G), and blue (B) color channels, respectively.

[0071] According to an embodiment of this application, productivity data can be calculated based on a preset productivity fusion inversion calculation formula, using previously calculated average height data and some observation indicators in Table 1.

[0072] According to embodiments of this application, multiple observation indicators include phenological indicators, leaf area indicators, and cover indicators.

[0073] According to an embodiment of this application, productivity data is generated based on a preset productivity calculation formula, average height data, and multiple observation indicators, including: calculating productivity data based on a preset productivity calculation formula, phenological indicators, leaf area indicators, coverage indicators, average height data, and multiple preset coefficients.

[0074] According to the embodiments of this application, the phenological indicators used in this embodiment are... The index, calculated using the preset productivity formula shown in formula (1), is combined with the leaf area index. Coverage indicators Average height data Productivity data can be calculated using multiple preset coefficients. .

[0075] According to an embodiment of this application, the preset productivity calculation formula is as shown in formula (1):

[0076] (1)

[0077] in, Fresh weight, i.e., productivity data, This is the average height data. For coverage metrics, Leaf area index As a phenological indicator, , , , , All are preset coefficients.

[0078] In one specific embodiment, , , , , The specific values ​​can be set according to the geographical location of the target area. In this embodiment, we take... =3.02、 , , =3474.74、 The specific value of the preset coefficient in this embodiment is obtained by integrating the on-site sampling data of the target area during historical periods.

[0079] According to an embodiment of this application, the average height data and the multispectral ecological indicators are input into the vegetation productivity fusion inversion model, and vegetation productivity data of the target area is generated by calculation, including: using the natural grassland vegetation productivity fusion inversion model to process the average height data and the multispectral image of the target area to obtain productivity data.

[0080] According to embodiments of this application, the natural grassland vegetation productivity fusion inversion model can be obtained by training neural networks such as Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN).

[0081] According to an embodiment of this application, after inputting the average height data and the multispectral image of the target area into a trained natural grassland vegetation productivity fusion inversion model, the productivity data of the target area can be obtained.

[0082] Figure 3 A schematic diagram of training multispectral images according to an embodiment of this application is shown.

[0083] According to an embodiment of this application, the natural grassland vegetation productivity fusion inversion model is trained as follows: A training sample set corresponding to at least one training region is obtained, wherein the training sample set includes multiple training samples and label data corresponding to each training sample. The training samples include training height data and training multispectral images of the training region, and the label data is productivity information of the training region obtained from actual field measurements. For each training sample, the training multispectral image is processed by data parsing to obtain multiple training indicators, wherein the multiple training indicators characterize different vegetation ecological information within the training region. The training samples and multiple training indicators are input into an initial neural network to output predicted productivity information. The predicted productivity information and the label data corresponding to the training samples are input into a loss function to output a target loss result. The model parameters of the initial neural network are iteratively adjusted according to the target loss result to obtain the trained natural grassland vegetation productivity fusion inversion model.

[0084] According to embodiments of this application, the training area and the target area should ideally have the same geographical environment. Training samples can be the average height data of plants within the training area. The loss function can be the mean squared error (MSE), mean absolute error (MAE), Huber loss function, etc.

[0085] According to embodiments of this application, for such Figure 3 Each training multispectral image shown can be processed through data analysis to obtain multiple training indicators for that training region, such as phenological indicators. Indicators, leaf area indicators Coverage indicators etc., among which, Figure 3 (a) Figure 3 (b) Figure 3 (c) and Figure 3 (d) are multispectral images of the training region at different historical times.

[0086] According to an embodiment of this application, training samples and corresponding training indicators are input into an initial neural network to obtain predicted productivity information for the training area. The predicted productivity information and corresponding label data are then input into a loss function to obtain the target loss result. Based on this target loss result, the model parameters of the neural network are adjusted to obtain a trained natural grassland vegetation productivity fusion inversion model.

[0087] Figure 4 A schematic diagram illustrating the calculation of average height data according to an embodiment of this application is shown.

[0088] According to an embodiment of this application, the distance observation data includes target horizontal distance information, target oblique distance, and target height information. The target horizontal distance information represents the horizontal distance between the ranging device and the target plant, the target oblique distance represents the oblique distance between the ranging device and the target location information, and the target height information represents the vertical distance between the ranging device and the ground.

[0089] According to embodiments of this application, such as Figure 4 As shown, based on the empirical height formula, and according to distance observation data and target point information, the average height data of multiple plants in the target area is generated, including: calculating the point height information based on the target point information, the target height information, the target horizontal distance information, and the target diagonal distance, wherein the point height information represents the distance between the target point information and the ground; and calculating the average height data according to the empirical height formula and the point height information.

[0090] According to an embodiment of this application, a constant B can be determined based on the ratio between the distance between the target point and the ground and the height of the target plant. For example, when a pre-selected representative measurement point is located at 3 / 4 of the target plant, B can be 4 / 3.

[0091] According to an embodiment of this application, based on target height information Target horizontal distance information diagonal distance from the target Calculate point height information As shown in formula (2):

[0092] (2)

[0093] According to an embodiment of this application, based on the empirical height formula shown in formula (3) and the point height information Calculate average height data .

[0094] (3)

[0095] in, This is the average height data. For point height information, For target height information, The diagonal distance to the target. For the target horizontal distance information, B is a coefficient determined based on the proportion of the target point's distance from the ground to the target plant's height. A and C are preset coefficients relating the target plant's height to its average height. .

[0096] In one specific embodiment, the preset point is selected as 3 / 4 of the plant height, so coefficient B is set to 4 / 3. Through historical data fitting, coefficients A=0.5613 and C=0.0645 are obtained. When the distance measuring device height... When the horizontal distance D=4.1m, the average height data of the target area It can be calculated using formula (4):

[0097] (4)

[0098] According to embodiments of this application, real-time monitoring of vegetation productivity helps to understand the health status of grassland ecosystems, providing a scientific basis for ecological environmental protection and grassland resource management. In the breeding of new livestock and poultry breeds and the technological innovation of modern ranches, accurate vegetation productivity data can help decision-makers and managers adjust pastoral production management measures, improve grassland yield and quality, and promote the sustainable development of animal husbandry. Furthermore, the vegetation productivity monitoring method of this embodiment has broad application prospects and can be extended to the monitoring and management of any grassland area or ecosystem, providing scientific support for agricultural and livestock production, grassland management, and sustainable resource development.

[0099] Figure 5 A schematic diagram of the equipment used in the vegetation productivity monitoring method according to an embodiment of this application is shown.

[0100] According to embodiments of this application, such as Figure 5 As shown, the ranging device and multispectral camera can be integrated into the same observation platform or deployed separately. The observation platform has rotation and pitch functions, enabling multi-angle observation. It can achieve horizontal rotation angles of 0–260°, vertical multi-angle observations of -5–60°, and 20x zoom within a 100-meter range, among other functions. It can acquire true-color and multispectral high-definition images of the entire natural grassland vegetation cycle, all-weather, and, with the assistance of a communication module, also supports remote real-time video viewing. In addition, the device combination may also include environmental sensors (such as temperature and humidity sensors) and a communication module, forming a complete near-ground collaborative sensing node. It can realize real-time monitoring, periodic transmission, and standardized storage of key parameters such as grassland vegetation productivity at the pasture scale, improving the efficiency of natural grassland data collection and management.

[0101] According to the embodiments of this application, the vegetation productivity monitoring method of this application can reflect the spatiotemporal variation characteristics of vegetation productivity more timely and accurately. Its intelligent and networked monitoring method greatly improves monitoring efficiency and data management level, which is of great significance for large-scale and long-term vegetation productivity monitoring. It is also better adapted to the characteristics of grassland ecosystems and can provide strong support for the accurate monitoring and management of grassland vegetation productivity.

[0102] Figure 6 A block diagram of a vegetation productivity monitoring device according to an embodiment of this application is shown.

[0103] like Figure 6 As shown, the vegetation productivity monitoring device 600 includes an acquisition module 610, a first generation module 620, and a second generation module 630.

[0104] The acquisition module 610 is used to acquire distance observation data in the target area collected by the ranging device in response to the monitoring command. The target area contains a natural grassland with multiple plants, and the distance observation data represents the distance between the ranging device and the target plant among the multiple plants.

[0105] The first generation module 620 is used to generate average height data of multiple plants in the target area based on empirical height formulas and using the processor according to distance observation data and target point information. The target point information represents the preset points on the target plants.

[0106] The second generation module 630 is used to generate productivity data for the target area based on the average height data and the multispectral image of the target area.

[0107] According to embodiments of this application, distance observation data within a target area and preset locations on target vegetation are collected using a ranging device. The average height data of the target area can be obtained by processing the distance observation data and target location information based on an empirical height formula. Therefore, productivity data for the target area can be generated based on the average height data and a multispectral image of the target area. Because the average height data and multispectral image of the target area are fully considered from both physical and ecological perspectives when calculating productivity data, the productivity data of the target area can be accurately calculated. This provides a timely and accurate data foundation for vegetation health early warning and vegetation utilization policy formulation, contributing to the healthy development of grassland ecosystems.

[0108] According to an embodiment of this application, the second generation module 630 includes a parsing unit and a generation unit.

[0109] The parsing unit is used to perform data parsing processing on multispectral images to obtain multiple observation indicators of the target area. These multiple observation indicators characterize different vegetation ecological information within the target area.

[0110] The generation unit is used to generate productivity data based on a preset productivity calculation formula, average height data, and multiple observation indicators.

[0111] According to embodiments of this application, multiple observation indicators include phenological indicators, leaf area indicators, and cover indicators.

[0112] According to an embodiment of this application, the generation unit includes a computational subunit.

[0113] The calculation subunit is used to calculate productivity data based on a preset productivity calculation formula, according to phenological indicators, leaf area indicators, coverage indicators, average height data, and multiple preset coefficients.

[0114] According to an embodiment of this application, the second generation module 630 includes a obtaining unit.

[0115] The obtained unit is used to process the average height data and multispectral images of the target area using the natural grassland vegetation productivity fusion inversion model to obtain productivity data.

[0116] According to an embodiment of this application, the training device for the natural grassland vegetation productivity fusion inversion model includes a second acquisition module, an analysis module, a prediction module, a loss calculation module, and an adjustment module.

[0117] The second acquisition module is used to acquire a training sample set corresponding to at least one training region. The training sample set includes multiple training samples and label data corresponding to each training sample. The training samples include training height data and training multispectral images of the training region. The label data represents the productivity information obtained from the actual measurement of the training region.

[0118] The parsing module is used to perform data parsing processing on the training multispectral image for each training sample to obtain multiple training indicators, which represent different vegetation ecological information within the training area.

[0119] The prediction module is used to input training samples and multiple training metrics into the initial neural network and output predicted productivity information.

[0120] The loss calculation module is used to input the predicted productivity information and the label data corresponding to the training samples into the loss function and output the target loss result.

[0121] The adjustment module is used to iteratively adjust the model parameters of the initial neural network based on the target loss result, so as to obtain a trained natural grassland vegetation productivity fusion inversion model.

[0122] According to an embodiment of this application, the distance observation data includes target horizontal distance information, target oblique distance, and target height information. The target horizontal distance information represents the horizontal distance between the ranging device and the target plant, the target oblique distance represents the oblique distance between the ranging device and the target location information, and the target height information represents the vertical distance between the ranging device and the ground.

[0123] According to an embodiment of this application, the first generation module 620 includes a first computing unit and a second computing unit.

[0124] The first calculation unit is used to calculate the point height information based on the target point information, according to the target height information, the target horizontal distance information, and the target diagonal distance. The point height information represents the distance between the target point and the ground.

[0125] The second calculation unit is used to calculate the average height data based on the empirical height formula and the height information of the points.

[0126] Any one or more of the modules, submodules, units, and subunits according to the embodiments of this application, or at least part of the functions of any one or more of them, can be implemented in one module. Any one or more of the modules, submodules, units, and subunits according to the embodiments of this application can be implemented by dividing them into multiple modules. Any one or more of the modules, submodules, units, and subunits according to the embodiments of this application can be at least partially implemented as hardware circuits, such as field-programmable gate arrays (FPGAs), programmable logic arrays (PLAs), systems-on-a-chip, systems-on-a-substrate, systems-on-package, application-specific integrated circuits (ASICs), or implemented by hardware or firmware in any other reasonable manner by integrating or packaging circuits, or implemented in any one of software, hardware, and firmware, or in a suitable combination of any of these. Alternatively, one or more of the modules, submodules, units, and subunits according to the embodiments of this application can be at least partially implemented as computer program modules, which, when run, can perform corresponding functions.

[0127] For example, any plurality of the acquisition module 610, the first generation module 620, and the second generation module 630 can be combined into one module / unit / subunit, or any one of these modules / units / subunits can be split into multiple modules / units / subunits. Alternatively, at least part of the functionality of one or more of these modules / units / subunits can be combined with at least part of the functionality of other modules / units / subunits and implemented in one module / unit / subunit. According to embodiments of this application, at least one of the acquisition module 610, the first generation module 620, and the second generation module 630 can be at least partially implemented as hardware circuitry, such as a field-programmable gate array (FPGA), a programmable logic array (PLA), a system-on-a-chip, a system-on-a-substrate, a system-on-package, an application-specific integrated circuit (ASIC), or any other reasonable means of integrating or packaging the circuitry, or implemented in software, hardware, or firmware, or in any appropriate combination of any of these three implementation methods. Alternatively, at least one of the acquisition module 610, the first generation module 620, and the second generation module 630 can be implemented at least partially as a computer program module, which can perform corresponding functions when the computer program module is run.

[0128] It should be noted that the vegetation productivity monitoring device part in the embodiments of this application corresponds to the vegetation productivity monitoring method part in the embodiments of this application. The description of the vegetation productivity monitoring device part is specifically referred to in the vegetation productivity monitoring method part, and will not be repeated here.

[0129] Figure 7 A block diagram of an electronic device suitable for implementing the methods described above, according to an embodiment of this application, is shown. Figure 7 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0130] like Figure 7 As shown, an electronic device 700 according to an embodiment of this application includes a processor 701, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 702 or a program loaded from a storage portion 708 into a random access memory (RAM) 703. The processor 701 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 701 may also include onboard memory for caching purposes. The processor 701 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of this application.

[0131] RAM 703 stores various programs and data required for the operation of electronic device 700. Processor 701, ROM 702, and RAM 703 are interconnected via bus 704. Processor 701 executes various operations of the method flow according to embodiments of this application by executing programs in ROM 702 and / or RAM 703. It should be noted that the programs may also be stored in one or more memories other than ROM 702 and RAM 703. Processor 701 may also execute various operations of the method flow according to embodiments of this application by executing programs stored in said one or more memories.

[0132] According to embodiments of this application, the electronic device 700 may further include an input / output (I / O) interface 705, which is also connected to a bus 704. The electronic device 700 may also include one or more of the following components connected to the input / output (I / O) interface 705: an input section 706 including a keyboard, mouse, etc.; an output section 707 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 708 including a hard disk, etc.; and a communication section 709 including a network interface card such as a LAN card, modem, etc. The communication section 709 performs communication processing via a network such as the Internet. A drive 710 is also connected to the input / output (I / O) interface 705 as needed. A removable medium 711, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 710 as needed so that computer programs read from it can be installed into the storage section 708 as needed.

[0133] According to embodiments of this application, the method flow according to embodiments of this application can be implemented as a computer software program. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable storage medium, the computer program containing program code for performing the method shown in the flowchart. In such embodiments, the computer program can be downloaded and installed from a network via communication section 709, and / or installed from removable medium 711. When the computer program is executed by processor 701, it performs the functions defined in the system of embodiments of this application. According to embodiments of this application, the systems, devices, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0134] This application also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more programs, which, when executed, implement the method according to the embodiments of this application.

[0135] According to embodiments of this application, the computer-readable storage medium can be a non-volatile computer-readable storage medium. Examples include, but are not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0136] For example, according to embodiments of this application, a computer-readable storage medium may include the ROM 702 and / or RAM 703 described above and / or one or more memories other than ROM 702 and RAM 703.

[0137] Embodiments of this application also include a computer program product comprising a computer program containing program code for performing the methods provided in the embodiments of this application. When the computer program product is run on an electronic device, the program code is used to enable the electronic device to implement the methods provided in the embodiments of this application.

[0138] When the computer program is executed by the processor 701, it performs the functions defined in the system / apparatus of this application embodiment. According to the embodiments of this application, the systems, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0139] In one embodiment, the computer program may rely on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may also be transmitted and distributed in the form of signals over a network medium, and may be downloaded and installed via the communication section 709, and / or installed from a removable medium 711. The program code contained in the computer program can be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination thereof.

[0140] According to embodiments of this application, program code for executing the computer programs provided in the embodiments of this application can be written in any combination of one or more programming languages. Specifically, these computational programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, languages ​​such as Java, C++, Python, "C", or similar programming languages. The program code can be executed entirely on the user's computing device, partially on the user's device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0141] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, blocks represented by two adjacent points may actually be executed substantially in parallel, or they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions. Those skilled in the art will understand that the features described in the various embodiments of this application can be combined and / or combined in various ways, even if such combinations are not explicitly described in this application. In particular, without departing from the spirit and teachings of this application, the features described in the various embodiments of this application can be combined and / or combined in various ways. All such combinations and / or combinations fall within the scope of this application.

[0142] The embodiments of this application have been described above. However, these embodiments are merely illustrative and not intended to limit the scope of this application. Although various embodiments have been described above, this does not mean that the measures in the various embodiments cannot be used advantageously in combination. This application does not depart from its scope, and those skilled in the art can make various substitutions and modifications, all of which should fall within the scope of this application.

Claims

1. A high-precision monitoring method for natural grassland vegetation productivity based on highly fusion inversion, characterized in that, include: In response to a monitoring command, distance observation data collected by a ranging device within a target area is acquired. The target area contains a natural grassland with multiple plants. The distance observation data represents the distance between the ranging device and a target plant among the multiple plants. The distance observation data includes target horizontal distance information, target oblique distance, and target height information. The target horizontal distance information represents the horizontal distance between the ranging device and the target plant; the target oblique distance represents the oblique distance between the ranging device and the target location information; and the target height information represents the vertical distance between the ranging device and the ground. The processor calculates the point height information based on the distance observation data and target point information, and based on the target point information, the target height information, the target horizontal distance information, and the target diagonal distance. The point height information represents the distance between the target point and the ground, and the target point information represents a preset point on the target plant. Based on the empirical height formula and the height information of the points, the average height data of multiple plants in the target area is calculated; The multispectral image of the target area is processed by data analysis to obtain multiple observation indicators of the target area. The multiple observation indicators characterize different vegetation ecological information in the target area, including phenological indicators, leaf area indicators and coverage indicators. Based on a preset productivity calculation formula, the productivity data of the target area is calculated according to the phenological index, the leaf area index, the coverage index, the average height data, and multiple preset coefficients. The method further includes: processing the average height data and the multispectral image of the target area using a natural grassland vegetation productivity fusion inversion model to obtain the productivity data.

2. The method according to claim 1, characterized in that, The preset productivity calculation formula is shown in formula (1): （1） in, Fresh weight, i.e., the aforementioned productivity data, This is the average height data. For coverage metrics, Leaf area index As a phenological indicator, , , , , All are preset coefficients.

3. The method according to claim 1, characterized in that, The natural grassland vegetation productivity fusion inversion model was trained in the following way: Obtain a training sample set corresponding to at least one training region, wherein the training sample set includes multiple training samples and label data corresponding to each training sample, the training sample includes training height data and training multispectral image of the training region, and the label data is productivity information obtained from real field measurements; For each training sample, the training multispectral image is subjected to data parsing processing to obtain multiple training indicators, wherein the multiple training indicators characterize different vegetation ecological information within the training area; The training samples and multiple training metrics are input into an initial neural network, which outputs predicted productivity information. The predicted productivity information and the label data corresponding to the training samples are input into the loss function, and the target loss result is output. The model parameters of the initial neural network are iteratively adjusted based on the target loss result to obtain a trained natural grassland vegetation productivity fusion inversion model.

4. The method according to claim 1, characterized in that, The empirical height formula is shown in formula (2): （2） in, This is the average height data. For point height information, For target height information, The diagonal distance to the target. B is the target horizontal distance information, which is a coefficient determined based on the proportion of the target point's distance from the ground to the height of the target plant. A and C are preset coefficients for the relationship between the height of the target plant and its average height.

5. A device for monitoring the productivity of natural grassland vegetation, comprising: The acquisition module is used to acquire distance observation data within a target area collected by the ranging device in response to a monitoring command. The target area contains a natural grassland with multiple plants. The distance observation data represents the distance between the ranging device and a target plant among the multiple plants. The distance observation data includes target horizontal distance information, target oblique distance, and target height information. The target horizontal distance information represents the horizontal distance between the ranging device and the target plant. The target oblique distance represents the oblique distance between the ranging device and the target location information. The target height information represents the vertical distance between the ranging device and the ground. The first generation module is used to generate average height data of multiple plants in the target area based on an empirical height formula and using a processor according to the distance observation data and target point information, wherein the target point information represents a preset point on the target plant; The second generation module is used to generate productivity data for the target area based on the average height data and the multispectral image of the target area; The second generation module includes: The analysis unit is used to perform data analysis and processing on the multispectral image to obtain multiple observation indicators of the target area, wherein the multiple observation indicators characterize different vegetation ecological information within the target area; The generation unit is used to generate productivity data based on a preset productivity calculation formula, average height data, and multiple observation indicators. The generation unit includes: The calculation subunit is used to calculate productivity data based on a preset productivity calculation formula, according to phenological indicators, leaf area indicators, coverage indicators, average height data, and multiple preset coefficients. The second generation module includes: The unit is used to process the average height data and multispectral images of the target area using the natural grassland vegetation productivity fusion inversion model to obtain productivity data. The first generation module includes: The first calculation unit is used to calculate the point height information based on the target point information, according to the target height information, the target horizontal distance information, and the target diagonal distance. The point height information represents the distance between the target point and the ground. The second calculation unit is used to calculate the average height data based on the empirical height formula and the height information of the points.

6. An electronic device, comprising: One or more processors; Memory, used to store one or more programs. Wherein, when the one or more programs are executed by the one or more processors, the one or more processors implement the method of any one of claims 1 to 4.

Images