A smart forestry monitoring system and method based on edge computing

Abstract

The present invention proposes a smart forestry monitoring system and method based on edge computing, the method comprising: obtaining forest area terrain data, and obtaining a digital elevation model of the forest area terrain based on processing the obtained relevant data; storing the obtained digital elevation model in a cloud platform; analyzing the digital elevation model of the forest area terrain stored in the cloud platform through an artificial intelligence algorithm, determining the sensor deployment location based on the analysis result, and deploying the corresponding sensor based on the sensor deployment location. The raw data is collected by the sensor, the edge device is connected to the sensor, and the raw data collected by the sensor is obtained in real time; the real-time data collection and processing of the edge device and the sensor enables the monitoring data to be timely transmitted to the cloud platform for further analysis and processing.

Description

Technical Field

[0001] The present invention proposes a smart forestry monitoring system and method based on edge computing, which belongs to the technical field of smart forestry monitoring. Background Art

[0002] Smart forestry monitoring is an important means to achieve forestry modernization and is of great significance to improving forestry production efficiency and management level. However, traditional forestry monitoring methods have many problems and challenges, such as untimely data acquisition and inaccurate data analysis and processing. In order to solve these problems, researchers began to explore smart forestry monitoring methods based on edge computing.

[0003] In traditional forestry monitoring methods, data acquisition usually relies on manual inspections or decentralized sensor networks, which are not only inefficient but also difficult to achieve full coverage. At the same time, due to the complexity and dynamism of the forestry environment, traditional data analysis and processing methods also have many limitations, making it difficult to accurately identify the topographical features of forest areas and the dynamic changes in the ecological environment. In response to these problems, some researchers have begun to try to use edge computing technology to improve the efficiency and accuracy of forestry monitoring. Edge computing technology can sink computing tasks from the cloud to edge devices close to the data source, thereby shortening data transmission delays and improving the real-time and accuracy of data analysis and processing. Summary of the invention

[0004] The present invention provides an edge computing-based smart forestry monitoring system and method to solve the problems in the prior art of insufficient real-time monitoring, low efficiency in data storage and processing collected by sensors, low security of data transmission processing and storage, weak autonomous decision-making and response capabilities of the monitoring system, and unreasonable sensor settings leading to insufficient information collection:

[0005] The present invention proposes a smart forestry monitoring method based on edge computing, the method comprising:

[0006] S1: Acquire forest area topography-related data, and acquire a digital elevation model of the forest area topography based on processing the acquired relevant data; and store the acquired digital elevation model in a cloud platform;

[0007] S2: Analyze the digital elevation model of the forest area topography stored in the cloud platform through an artificial intelligence algorithm, determine the sensor deployment location according to the analysis result, and deploy corresponding sensors according to the sensor deployment location, collect raw data through the sensor, connect the edge device with the sensor, and obtain the raw data collected by the sensor in real time;

[0008] S3: The edge device preprocesses the acquired raw data, and performs data analysis and processing on the preprocessed raw data through a machine learning algorithm and a pattern recognition algorithm to obtain analysis and processing results. The edge device generates monitoring information based on the analysis and processing results, and transmits the monitoring information to the cloud platform through the Internet of Things;

[0009] S4: The cloud platform receives the monitoring information transmitted by the edge device, further analyzes and processes the monitoring information in combination with the digital elevation model through an artificial intelligence algorithm, obtains further analysis and processing results, and performs monitoring and management based on the further analysis and processing results.

[0010] Furthermore, the obtaining of forest area topography-related data, and obtaining a digital elevation model of the forest area topography based on processing the obtained relevant data; and storing the obtained digital elevation model in a cloud platform, includes:

[0011] S101: Determine the location information of the forest area to be monitored, and determine the method for acquiring the topography-related data of the forest area according to the location information of the forest area to be monitored;

[0012] S102: collecting relevant data of forest area topography and landforms according to the determined acquisition method, wherein the acquisition method includes aerial survey, ground survey and satellite remote sensing;

[0013] S103: pre-processing the collected data related to the forest topography to obtain the elevation data of the forest topography; and fusing different elevation data through a data fusion algorithm and a weighted average algorithm, wherein the pre-processing includes noise removal, adjustment processing and data correction;

[0014] S104: interpolating the discrete elevation data points into a continuous elevation surface through an interpolation algorithm, and generating a corresponding elevation data model;

[0015] S105: converting the interpolated continuous elevation data model into a grid format, wherein the grid format includes a regular grid and an irregular grid, wherein the regular grid includes a square grid and a rectangular grid, and the irregular grid includes a triangular grid;

[0016] S106: Smoothing and correcting the generated digital elevation model through filtering and interpolation, and removing local outliers and errors;

[0017] S107: exporting the generated digital elevation model into a data file in a standard format, wherein the format includes GeoTIFF and ASCII;

[0018] S108: Determine the format of the digital elevation model, and determine storage requirements according to the data size, access frequency, and expected growth data of the digital elevation model;

[0019] S109: dividing the digital elevation model into a series of spatial regions according to a grid division method according to the spatial range of the digital elevation model;

[0020] S110: creating a storage bucket in the cloud platform, and creating a corresponding storage directory structure according to the divided space area, grid ID and row and column numbers;

[0021] S111: Divide the divided digital elevation model into grids according to their grids, and store them in corresponding storage directories in sequence;

[0022] S112: establishing a backup and disaster recovery mechanism for the digital elevation model data stored in the storage directory, and setting usage permissions and access control policies;

[0023] S113: Add corresponding metadata information in the cloud platform, where the metadata information includes the spatial range, resolution, and update date of the grid;

[0024] S114: Regularly optimize and upgrade the cloud platform according to actual usage and demand changes.

[0025] Furthermore, the digital elevation model of the forest area topography stored in the cloud platform is analyzed by an artificial intelligence algorithm, the sensor deployment location is determined according to the analysis result, and the corresponding sensor is deployed according to the sensor deployment location, raw data is collected by the sensor, the edge device is connected to the sensor, and the raw data collected by the sensor is obtained in real time, including:

[0026] S201: Acquire digital elevation model data of forest area topography from a cloud platform, process and analyze the digital elevation model data through an artificial intelligence algorithm, and acquire relevant information of the forest area topography, the information including topography, height, slope, and hydrology;

[0027] S202: Determine the deployment location and quantity of sensors according to the information, and determine the type of sensors to be used according to the characteristics of the forest area and the information to be monitored;

[0028] S203: selecting a suitable location and scheme for sensor deployment according to the determined sensor location and quantity and the sensor types to be used; the sensor types include temperature and humidity sensors, fire risk index sensors, geological monitoring sensors, and image sensors;

[0029] S2031: The temperature and humidity sensor is used to monitor the climate and humidity changes in the forest area; the temperature and humidity sensor is deployed at a position 1.5-3m above the ground. The deployment rule of the temperature and humidity sensor is to set two temperature and humidity sensors for every 1250 square meters. Let the area of ​​the forest be S, and the number of temperature and humidity sensors set be N. If S-1250≤0, then N=2. If S-1250>0, then

[0030] S2032: The fire risk index sensor is used to assess the fire risk in the forest area. The fire risk index sensor is deployed in the forest area close to the fire source, flammable materials and the edge of the forest area. The number of the fire risk index sensors is not less than 16;

[0031] S2033: The geological monitoring sensor is used to detect soil moisture and soil stability. Twelve geological monitoring sensors are randomly deployed in the forest area and installed next to trees without direct contact with the trees.

[0032] S2034: The image sensor is used to monitor the growth status of vegetation and the situation of pests and diseases. The image sensor is installed in a representative location with a wide field of view, no obstacles, no direct sunlight, and the representative location includes forests of different types and woodlands of different ages. The number of image sensors installed is one per thousand square meters;

[0033] S204: After the sensor is deployed, an edge device is installed next to the sensor, and the edge device is connected to the sensor through a low-power wireless communication protocol, where the low-power wireless communication protocol includes LoRaWAN and NB-IoT;

[0034] S205: Collect raw data through sensors, and transmit the data collected by the sensors to edge devices in real time.

[0035] Furthermore, the edge device preprocesses the acquired raw data, and performs data analysis and processing on the preprocessed raw data through a machine learning algorithm and a pattern recognition algorithm to obtain analysis and processing results. The edge device generates monitoring information based on the analysis and processing results, and transmits the monitoring information to the cloud platform through the Internet of Things, including:

[0036] S301: The edge device detects the acquired raw data to remove outliers and removes noise data points through a filtering algorithm;

[0037] S302: compressing the original data by using a compression algorithm, normalizing and standardizing the data with different characteristics, and mapping the normalized and standardized data to the same magnitude;

[0038] S303: extracting features from the data mapped to the same magnitude, reducing data dimensions and retaining key information;

[0039] S304: Divide the feature-extracted data into multiple subtasks, and calculate the subtasks through parallel computing to obtain pre-processed data after feature extraction;

[0040] S305: Analyze and process the pre-processed data after feature extraction through a machine learning algorithm and a pattern recognition algorithm to obtain corresponding information, wherein the corresponding information includes forest area conditions, ecological environment, and vegetation generation conditions;

[0041] S306: Generate monitoring information based on the results of data analysis and processing, the monitoring information including temperature and humidity ratio, fire risk level, vegetation coverage, soil moisture and stability, and pest and disease risk;

[0042] S307: The edge device transmits the monitoring information to the cloud platform in real time through the Internet of Things, and encrypts the transmitted monitoring information using an encryption protocol;

[0043] S308: Monitor the connection status between the edge device and the cloud platform by adopting a heartbeat mechanism.

[0044] Furthermore, the edge device preprocesses the acquired raw data, and performs data analysis and processing on the preprocessed raw data through a machine learning algorithm and a pattern recognition algorithm to obtain analysis and processing results. The edge device generates monitoring information based on the analysis and processing results, and transmits the monitoring information to the cloud platform through the Internet of Things, including:

[0045] S401: The cloud platform receives the monitoring information transmitted by the edge device, and creates a storage directory in the storage bucket according to the specific situation of the monitoring information, wherein the specific situation of the monitoring information includes the name of the monitoring point and the time period;

[0046] S402: After the storage directory is created, the received monitoring information is uploaded to the corresponding storage directory;

[0047] S403: Extract and select features from data using machine learning algorithms based on the characteristics and objectives of forestry monitoring;

[0048] S404: establishing a management and monitoring model through the feature extraction result, an artificial intelligence algorithm and a digital elevation model, and training and verifying the established management and monitoring model through historical monitoring data;

[0049] S405: further analyzing the newly collected monitoring data using the trained management monitoring model to obtain further analysis results, wherein the further analysis includes trend analysis, anomaly detection, and pattern recognition;

[0050] S406: Perform monitoring and management based on the analysis results, including making decisions, optimizing forestry management strategies, generating reports, and providing real-time monitoring data visualization.

[0051] The present invention proposes a smart forestry monitoring system based on edge computing, the system comprising:

[0052] Digital elevation model building module: obtains forest area topography-related data, and obtains a digital elevation model of the forest area topography based on processing the obtained relevant data; stores the obtained digital elevation model in a cloud platform;

[0053] Raw data acquisition module: Analyze the digital elevation model of the forest area topography stored in the cloud platform through artificial intelligence algorithms, determine the sensor deployment location according to the analysis results, and deploy corresponding sensors according to the sensor deployment location, collect raw data through sensors, connect edge devices with sensors, and obtain raw data collected by sensors in real time;

[0054] Monitoring information generation module: The edge device pre-processes the acquired raw data, and performs data analysis and processing on the pre-processed raw data through machine learning algorithms and pattern recognition algorithms to obtain analysis and processing results. The edge device generates monitoring information based on the analysis and processing results, and transmits the monitoring information to the cloud platform through the Internet of Things;

[0055] Monitoring and management module: The cloud platform receives the monitoring information transmitted by the edge device, combines the digital elevation model with the artificial intelligence algorithm, further analyzes and processes the monitoring information, obtains further analysis and processing results, and performs monitoring and management based on the further analysis and processing results.

[0056] Furthermore, the digital elevation model building module includes:

[0057] Method determination module: determine the location information of the forest area to be monitored, and determine the method for obtaining the topography-related data of the forest area according to the location information of the forest area to be monitored;

[0058] Topography and geomorphology acquisition module: collects relevant data of forest topography and geomorphology according to the determined acquisition method, and the acquisition method includes aerial survey, ground survey and satellite remote sensing;

[0059] Data preprocessing module: preprocess the collected data related to the forest area topography and obtain the elevation data of the forest area topography; and fuse different elevation data through data fusion algorithm and weighted average algorithm, and the preprocessing includes noise removal, adjustment processing and data correction;

[0060] Interpolation algorithm execution module: through the interpolation algorithm, the discrete elevation data points are interpolated into a continuous elevation surface and the corresponding elevation data model is generated;

[0061] A grid format conversion module: converting the interpolated continuous elevation data model into a grid format, wherein the grid format includes a regular grid and an irregular grid, wherein the regular grid includes a square grid and a rectangular grid, and the irregular grid includes a triangular grid;

[0062] Smoothing correction module: Smoothing and correcting the generated digital elevation model through filtering and interpolation, and removing local outliers and errors;

[0063] Format export module: export the generated digital elevation model to a data file in a standard format, including GeoTIFF and ASCII;

[0064] Storage requirement determination module: determining the format of the digital elevation model, and determining storage requirements according to the data size, access frequency and expected growth data of the digital elevation model;

[0065] A grid division module: according to the spatial range of the digital elevation model, the digital elevation model is divided into a series of spatial regions according to a grid division method;

[0066] Bucket creation module: creates a bucket in the cloud platform, and creates a corresponding storage directory structure based on the divided space area, grid ID, and row and column numbers;

[0067] Storage corresponding module: divide the divided digital elevation model into grids according to their grids and store them in the corresponding storage directory in sequence;

[0068] Permission setting module: establish backup and disaster recovery mechanisms for digital elevation model data stored in the storage directory, and set usage permissions and access control policies;

[0069] Update module: adding corresponding metadata information in the cloud platform, wherein the metadata information includes the spatial range, resolution and update date of the grid;

[0070] Optimization and upgrade module: regularly optimize and upgrade the cloud platform according to actual usage and changes in demand.

[0071] Furthermore, the raw data acquisition module includes:

[0072] Relevant information acquisition module: obtains digital elevation model data of forest area topography from the cloud platform, processes and analyzes the digital elevation model data through artificial intelligence algorithms, and obtains relevant information of the forest area topography, including topography, height, slope and hydrology;

[0073] Sensor type determination module: determines the deployment location and quantity of sensors according to the information, and determines the type of sensors to be used according to the characteristics of the forest area and the information to be monitored;

[0074] Installation plan determination module: selects appropriate locations and plans for sensor deployment according to the determined sensor locations, quantity, and sensor types to be used; the sensor types include temperature and humidity sensors, fire risk index sensors, geological monitoring sensors, and image sensors;

[0075] Temperature and humidity sensor module: The temperature and humidity sensor is used to monitor the climate and humidity changes in the forest area; the temperature and humidity sensor is deployed at a position 1.5-3m above the ground. The deployment rule of the temperature and humidity sensor is to set two temperature and humidity sensors for every 1250 square meters. Let the forest area be S, and the number of temperature and humidity sensors set be N. If S-1250≤0, then N=2, if S-1250>0, then

[0076] Fire risk index sensor module: The fire risk index sensor is used to assess the fire risk in the forest area. The fire risk index sensor is deployed in the forest area close to the fire source, flammable materials and the edge of the forest area. The number of the fire risk index sensors is not less than 16;

[0077] Geological monitoring sensor module: The geological monitoring sensor is used to detect soil moisture and soil stability. Twelve geological monitoring sensors are randomly deployed in the forest area and installed next to trees without direct contact with the trees.

[0078] Image sensor module: The image sensor is used to monitor the growth status of vegetation and the situation of pests and diseases. The image sensor is installed in a representative location with a wide field of view, no obstacles, no direct sunlight, and the representative location includes forests of different types and woodlands of different ages. The number of image sensors installed is one per thousand square meters;

[0079] Edge device installation module: After the sensor is deployed, an edge device is installed next to the sensor, and the edge device is connected to the sensor through a low-power wireless communication protocol, which includes LoRaWAN and NB-IoT;

[0080] Real-time data transmission module: collects raw data through sensors and transmits the data collected by sensors to edge devices in real time.

[0081] Furthermore, the monitoring information generation module includes:

[0082] Data filtering module: The edge device detects the acquired raw data to remove outliers and removes noise data points through filtering algorithms;

[0083] Data compression module: compresses the original data through a compression algorithm, normalizes and standardizes the data with different characteristics, and maps the normalized and standardized data to the same magnitude;

[0084] Feature extraction module: extracting features from the data mapped to the same magnitude, reducing data dimensions and retaining key information;

[0085] Subtask division module: divides the feature-extracted data into multiple subtasks, calculates the subtasks through parallel computing, and obtains pre-processed data after feature extraction;

[0086] Feature analysis and processing module: analyzes and processes the pre-processed data after feature extraction through machine learning algorithms and pattern recognition algorithms to obtain corresponding information, including forest conditions, ecological environment, and vegetation generation conditions;

[0087] Monitoring information generation module: generates monitoring information based on the results of data analysis and processing, and the monitoring information includes temperature and humidity ratio, fire risk level, vegetation coverage, soil moisture and stability, and pest and disease risk;

[0088] Transmission encryption module: The edge device transmits the monitoring information to the cloud platform in real time through the Internet of Things, and uses an encryption protocol to encrypt the transmitted monitoring information;

[0089] Status monitoring module: monitors the connection status between the edge device and the cloud platform by adopting a heartbeat mechanism.

[0090] Furthermore, the monitoring management module includes:

[0091] Cloud platform receiving module: The cloud platform receives the monitoring information transmitted by the edge device and creates a storage directory in the storage bucket according to the specific situation of the monitoring information, wherein the specific situation of the monitoring information includes the name of the monitoring point and the time period;

[0092] Information storage corresponding module: after creating the storage directory, the received monitoring information is uploaded to the corresponding storage directory;

[0093] Machine learning module: Extract and select features of data through machine learning algorithms according to the characteristics and objectives of forestry monitoring;

[0094] Management and monitoring model building module: Based on the results of feature extraction, the management and monitoring model is built through artificial intelligence algorithms combined with digital elevation models, and the established management and monitoring model is trained and verified through historical monitoring data;

[0095] Model detection module: Use the trained management monitoring model to further analyze the newly collected monitoring data to obtain further analysis results, including trend analysis, anomaly detection and pattern recognition;

[0096] Monitoring and management implementation module: monitoring and management are carried out according to the analysis results, including making decisions, optimizing forestry management strategies, generating reports and providing real-time monitoring data visualization.

[0097] Beneficial effects of the present invention: The above technical solution can quickly obtain digital elevation models and monitoring information by obtaining forest area topography-related data, and performing preprocessing and analysis. Real-time data acquisition and processing of edge devices and sensors enable monitoring data to be transmitted to the cloud platform in a timely manner for further analysis and processing; the digital elevation model is analyzed by artificial intelligence algorithms to determine the deployment location and number of sensors. According to the characteristics of the forest area and monitoring requirements, the appropriate sensor type is selected to ensure that the monitoring coverage is wide and effective; the edge device performs preprocessing on the collected raw data, and applies machine learning algorithms and pattern recognition algorithms to perform data analysis and processing to obtain accurate analysis results. Through feature extraction, trend analysis, anomaly detection and pattern recognition, detailed monitoring information such as forest area conditions, ecological environment and vegetation generation can be obtained; intelligent monitoring management is realized by establishing management monitoring models and training and verification of historical monitoring data. Decision-making and forestry management strategy optimization are carried out according to the analysis results, reports are generated and real-time monitoring data visualization is provided to improve the management level and effect of forestry monitoring; the digital elevation model and monitoring information are stored in the cloud platform, and the effective management of data and the satisfaction of storage requirements are ensured by dividing spatial areas and creating corresponding storage directory structures. At the same time, backup and disaster recovery mechanisms are established, and encryption protocols are used to protect data transmission and access to ensure data security and reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0098] Figure 1 This is a step diagram of a smart forestry monitoring method based on edge computing described in the present invention. DETAILED DESCRIPTION

[0099] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0100] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. The embodiments described are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0101] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0102] One embodiment of the present invention provides a smart forestry monitoring method based on edge computing, the method comprising:

[0103] S1: Acquire forest area topography-related data, and acquire a digital elevation model of the forest area topography based on processing the acquired relevant data; and store the acquired digital elevation model in a cloud platform;

[0104] S2: Analyze the digital elevation model of the forest area topography stored in the cloud platform through an artificial intelligence algorithm, determine the sensor deployment location according to the analysis result, and deploy corresponding sensors according to the sensor deployment location, collect raw data through the sensor, connect the edge device with the sensor, and obtain the raw data collected by the sensor in real time;

[0105] S3: The edge device preprocesses the acquired raw data, and performs data analysis and processing on the preprocessed raw data through a machine learning algorithm and a pattern recognition algorithm to obtain analysis and processing results. The edge device generates monitoring information based on the analysis and processing results, and transmits the monitoring information to the cloud platform through the Internet of Things;

[0106] S4: The cloud platform receives the monitoring information transmitted by the edge device, further analyzes and processes the monitoring information in combination with the digital elevation model through an artificial intelligence algorithm, obtains further analysis and processing results, and performs monitoring and management based on the further analysis and processing results.

[0107] The working principle of the above technical solution is as follows: first, obtain the relevant data of the forest area topography, which can be obtained through surveying, remote sensing or laser radar and other technical means. Then, process the acquired data to establish a digital elevation model of the forest area topography. The obtained digital elevation model is stored in the cloud platform for subsequent analysis and processing; on the cloud platform, the stored digital elevation model is analyzed by artificial intelligence algorithm. Through the analysis results, the sensor deployment location is determined, and a suitable location is selected to monitor the environment and potential risks of the forest area. According to the deployment location, the corresponding sensor is installed in the forest area and connected to the edge device to obtain the raw data collected by the sensor in real time; the edge device receives the raw data collected by the sensor and preprocesses it. This includes data cleaning, noise filtering, data normalization and other operations. Then, the preprocessed data is analyzed and processed by machine learning algorithms and pattern recognition algorithms to obtain analysis results. The edge device generates monitoring information based on these results and transmits it to the cloud platform through the Internet of Things; the cloud platform receives the monitoring information transmitted by the edge device, and combines the digital elevation model to further analyze and process the monitoring information using artificial intelligence algorithm. On this basis, more in-depth analysis and processing results are generated. Based on these results, monitoring and management, such as environmental assessment and risk early warning, are carried out to provide effective forest monitoring and management decision support.

[0108] The effects of the above technical solutions are: by collecting data in real time through sensors, and using artificial intelligence algorithms for processing and analysis, real-time monitoring of the forest environment can be achieved. It helps to detect abnormal situations and potential risks in a timely manner so that appropriate measures can be taken to manage and respond; by establishing a digital elevation model and adopting a machine learning algorithm, the solution can provide high-precision analysis results; it can help forest managers to more accurately understand the topographic characteristics of the forest area, as well as the trends of environmental changes and potential risks, and provide a scientific basis for decision-making; the cloud platform combines digital elevation models and monitoring information to generate in-depth analysis and processing results through artificial intelligence algorithms; it can be used for decision support in environmental assessment, risk warning and other aspects. Forest managers can formulate effective management strategies based on these results to improve the sustainable development and protection level of forest areas; by integrating digital elevation models, sensor data collection and artificial intelligence algorithm analysis and processing, the overall management is optimized. Forest managers can conduct centralized monitoring and management on the cloud platform to improve management efficiency and resource utilization; through real-time monitoring and analysis and processing, the solution can detect potential risks and environmental problems early and provide early warning information; it helps forest managers to take timely measures to reduce losses and impacts and improve emergency response capabilities.

[0109] In one embodiment of the present invention, the step of acquiring forest area topography-related data, and acquiring a digital elevation model of the forest area topography based on processing the acquired related data; and storing the acquired digital elevation model in a cloud platform comprises:

[0110] S101: Determine the location information of the forest area to be monitored, and determine the method for acquiring the topographic data of the forest area according to the location information of the forest area to be monitored; the relevant data includes elevation measurement data, laser radar scanning data, aerial remote sensing images and satellite images;

[0111] S102: Collect relevant data on the topography of the forest area according to the determined acquisition method, which includes aerial survey, ground survey and satellite remote sensing; for example, for a forest area located in a mountainous area, aerial survey can be used to obtain the elevation data and related information such as vegetation of the forest area, and ground survey is required for accurate details of each point; for forest areas in plain areas, satellite remote sensing technology can be used to quickly obtain relevant topography data. Use these different methods to acquire data, and select the appropriate method according to the specific situation to achieve the best effect.

[0112] S103: pre-processing the collected data related to the forest topography to obtain the elevation data of the forest topography; and fusing different elevation data through a data fusion algorithm and a weighted average algorithm, wherein the pre-processing includes noise removal, adjustment processing and data correction;

[0113] S104: interpolating the discrete elevation data points into a continuous elevation surface through an interpolation algorithm, and generating a corresponding elevation data model;

[0114] S105: converting the interpolated continuous elevation data model into a grid format, wherein the grid format includes a regular grid and an irregular grid, wherein the regular grid includes a square grid and a rectangular grid, and the irregular grid includes a triangular grid;

[0115] S106: Smoothing and correcting the generated digital elevation model through filtering and interpolation, and removing local outliers and errors;

[0116] S107: exporting the generated digital elevation model into a data file in a standard format, wherein the format includes GeoTIFF and ASCII;

[0117] S108: Determine the format of the digital elevation model, and determine the storage requirements according to the data size, access frequency and expected growth data of the digital elevation model; for example: assuming that the digital elevation model of a certain area is stored in a raster format, with a spatial resolution of 10 meters, a coverage area of ​​100km×100km, and an elevation value of 2 bytes (16 bits) for each grid cell, then its data size is 10000×10000×2=200MB. If the digital elevation model needs to acquire and process data in real time, a cloud storage solution with high bandwidth and low latency can be selected. If the data growth is expected to be large in the next few years, a highly scalable object-oriented storage or distributed file system can be selected for storage.

[0118] S109: dividing the digital elevation model into a series of spatial regions according to a grid division method according to the spatial range of the digital elevation model;

[0119] S110: creating a storage bucket in the cloud platform, and creating a corresponding storage directory structure according to the divided space area, grid ID and row and column numbers;

[0120] S111: Divide the divided digital elevation model into grids according to their grids, and store them in corresponding storage directories in sequence;

[0121] S112: establishing a backup and disaster recovery mechanism for the digital elevation model data stored in the storage directory, and setting usage permissions and access control policies;

[0122] S113: Add corresponding metadata information in the cloud platform, where the metadata information includes the spatial range, resolution, and update date of the grid;

[0123] S114: Regularly optimize and upgrade the cloud platform according to actual usage and demand changes. For example, regularly perform security reinforcement and patch updates on the cloud platform based on the latest security threats and vulnerability intelligence. Timely update security patches for operating systems and software, strengthen the configuration of network firewalls and intrusion detection systems, conduct penetration tests and security audits, etc. to ensure the security of the cloud platform.

[0124] The working principle of the above technical solution is: determine the location of the monitored forest area, and decide the method of obtaining relevant data according to the location; collect relevant data on the topography of the forest area according to the determined acquisition method, including elevation measurement data, lidar scanning data, aerial remote sensing images and satellite images, and data collection can be carried out through aerial measurement, ground survey and satellite remote sensing; pre-process the collected forest area topography related data, including noise removal, adjustment processing and data correction operations to obtain accurate elevation data; perform data fusion and weighted average processing on different elevation data, and then use interpolation algorithms to interpolate discrete elevation data points into continuous elevation surfaces to generate elevation data models; convert the interpolated continuous elevation data model into a grid format, including regular grids and irregular grids, such as square, rectangular and triangular grids; smooth and correct the generated digital elevation model through filtering and interpolation algorithms to remove Local outliers and errors can be eliminated to improve the quality and accuracy of data; the generated digital elevation model can be exported to data files in standard formats, such as GeoTIFF and ASCII, and the storage requirements can be determined based on the format, size, access frequency and expected growth data of the digital elevation model; the digital elevation model can be divided into a series of spatial areas according to its spatial range, and storage buckets and corresponding storage directory structures can be created in the cloud platform to store the digital elevation model in sequence according to grid ID and row and column numbers; a backup and disaster recovery mechanism can be established to ensure the security and reliability of digital elevation model data, and usage permissions and access control policies can be set to protect the confidentiality and integrity of the data; metadata information related to the digital elevation model can be added to the cloud platform, including spatial range, resolution, update date and other information, to facilitate subsequent management and retrieval; according to actual usage and changes in demand, the cloud platform can be regularly optimized and upgraded to improve system performance and user experience.

[0125] The effects of the above technical solutions are as follows: through a variety of acquisition methods, including elevation measurement data, lidar scanning data, aerial remote sensing images and satellite images, the forest area topography-related data obtained are more comprehensive; by processing the acquired data through preprocessing, data fusion and interpolation algorithms, the digital elevation model of the forest area topography can be accurately obtained to improve the accuracy and precision of the data; the digital elevation model is stored in the cloud platform, and the digital elevation model data can be easily managed and retrieved by dividing the spatial area and creating a storage directory structure; a backup and disaster recovery mechanism is established to ensure the security and reliability of the digital elevation model data, and at the same time, usage permissions and access control policies are set to protect the confidentiality and integrity of the data; metadata information related to the digital elevation model is added to the cloud platform, including spatial range, resolution and update date, etc., to facilitate subsequent management and retrieval; according to actual usage and changes in demand, the cloud platform is regularly optimized and upgraded to improve system performance and user experience.

[0126] In one embodiment of the present invention, the digital elevation model of the forest area topography stored in the cloud platform is analyzed by an artificial intelligence algorithm, a sensor deployment location is determined according to the analysis result, and corresponding sensors are deployed according to the sensor deployment location, raw data is collected by the sensor, an edge device is connected to the sensor, and raw data collected by the sensor is obtained in real time, including:

[0127] S201: Acquire digital elevation model data of forest area topography from a cloud platform, process and analyze the digital elevation model data through an artificial intelligence algorithm, and acquire relevant information of the forest area topography, the information including topography, height, slope, and hydrology;

[0128] S202: Determine the deployment location and quantity of sensors according to the information, and determine the type of sensors to be used according to the characteristics of the forest area and the information to be monitored;

[0129] S203: selecting a suitable location and scheme for sensor deployment according to the determined sensor location and quantity and the sensor types to be used; the sensor types include temperature and humidity sensors, fire risk index sensors, geological monitoring sensors, and image sensors;

[0130] S2031: The temperature and humidity sensor is used to monitor the climate and humidity changes in the forest area; the temperature and humidity sensor is deployed at a position 1.5-3m above the ground. The deployment rule of the temperature and humidity sensor is to set two temperature and humidity sensors for every 1250 square meters. Let the area of ​​the forest be S, and the number of temperature and humidity sensors set be N. If S-1250≤0, then N=2. If S-1250>0, then

[0131] S2032: The fire risk index sensor is used to assess the fire risk in the forest area. The fire risk index sensor is deployed in the forest area near the fire source, flammable materials and the edge of the forest area. The number of the fire risk index sensors is not less than 16; for example: for the potential fire source, the sensor can be deployed in the area where fires often occur, such as camps, barbecue areas, camping grounds or outdoor fire areas. For areas near flammable vegetation, the sensor can be arranged near the vegetation area that is easy to burn, such as dry grass, forest edges or bushes. For monitoring the entrance and exit areas, the sensor can be set at the entrance and exit of the forest area to monitor the fire risk of people entering and leaving, so as to take timely measures to prevent the spread of fire.

[0132] S2033: The geological monitoring sensor is used to detect soil moisture and soil stability. Twelve geological monitoring sensors are randomly deployed in the forest area and installed next to trees without direct contact with the trees.

[0133] S2034: The image sensor is used to monitor the growth status of vegetation and the situation of pests and diseases. The image sensor is installed in a representative location with a wide field of view, no obstacles, no direct sunlight, and the representative location includes forests of different types and woodlands of different ages. The number of image sensors installed is one per thousand square meters;

[0134] S204: After the sensor is deployed, an edge device is installed next to the sensor, and the edge device is connected to the sensor through a low-power wireless communication protocol, where the low-power wireless communication protocol includes LoRaWAN and NB-IoT;

[0135] S205: Collect raw data through sensors, and transmit the data collected by the sensors to edge devices in real time.

[0136] The working principle of the above technical solution is: obtain digital elevation model data of forest area topography through the cloud platform, use artificial intelligence algorithms to process and analyze the data, and extract relevant information such as terrain, height, slope and hydrology; according to the characteristics of the forest area and monitoring needs, determine the type of sensors to be used, including temperature and humidity sensors, fire risk index sensors, geological monitoring sensors and image sensors; based on the determined sensor type and monitoring target, determine the deployment location and number of sensors. Temperature and humidity sensors are deployed at a distance of 1.5-3m from the ground according to the rules, with two sensors set up for every 1,250 square meters; fire risk index sensors are deployed near fire sources, flammable materials and the edge of forest areas, with a number of no less than 16; 12 geological monitoring sensors are randomly deployed, placed next to trees in the forest area, and are not in direct contact with the trees; image sensors are installed in open locations at a density of one per thousand square meters, such as different types of forests and woodlands of different ages; edge devices are installed next to the sensors, and low-power wireless communication protocols (such as LoRaWAN and NB-IoT) are used to connect the edge devices to the sensors to realize data collection and transmission; the sensors collect environmental data through edge devices, and transmit the collected data to the edge devices in real time to form a real-time monitoring system.

[0137] The effects of the above technical solutions are as follows: by processing and analyzing the digital elevation model data through artificial intelligence algorithms, relevant information such as forest terrain, height, slope, and hydrology can be obtained, providing detailed terrain data for forest management and planning; according to the terrain information obtained and monitoring needs, the deployment location and number of sensors are determined to ensure that the monitoring coverage is wide and accurate. Different types of sensors (temperature and humidity, fire risk index, geological monitoring, and images) are deployed in appropriate locations to comprehensively monitor the climate, fire risk, soil conditions, and vegetation growth status of the forest area; through the connection between edge devices and sensors, real-time transmission of sensor collected data is realized. In this way, climate change, fire risk, soil conditions, and vegetation conditions in the forest area can be understood in a timely manner, and the data can be analyzed and processed; combining the data of multiple sensors, the fire risk, soil moisture, stability, and vegetation conditions in the forest area can be more accurately assessed. This helps to detect abnormal situations in a timely manner and improve the monitoring capabilities of natural disasters and ecological environment; through automated data collection and transmission, the workload of manual collection and input is reduced, saving time and labor costs; using low-power wireless communication protocols (such as LoRaWAN and NB-IoT) to ensure data security and reliability during transmission.

[0138] In one embodiment of the present invention, the edge device preprocesses the acquired raw data, and performs data analysis and processing on the preprocessed raw data through a machine learning algorithm and a pattern recognition algorithm to obtain analysis and processing results. The edge device generates monitoring information based on the analysis and processing results, and transmits the monitoring information to the cloud platform through the Internet of Things, including:

[0139] S301: The edge device detects the acquired raw data to remove outliers and removes noise data points through a filtering algorithm;

[0140] S302: compressing the original data by using a compression algorithm, normalizing and standardizing the data with different characteristics, and mapping the normalized and standardized data to the same magnitude;

[0141] S303: extracting features from the data mapped to the same magnitude, reducing data dimensions and retaining key information;

[0142] S304: Divide the feature-extracted data into multiple subtasks, and calculate the subtasks through parallel computing to obtain pre-processed data after feature extraction;

[0143] S305: Analyze and process the pre-processed data after feature extraction through a machine learning algorithm and a pattern recognition algorithm to obtain corresponding information, wherein the corresponding information includes forest area conditions, ecological environment, and vegetation generation conditions;

[0144] S306: Generate monitoring information based on the results of data analysis and processing, the monitoring information including temperature and humidity ratio, fire risk level, vegetation coverage, soil moisture and stability, and pest and disease risk;

[0145] S307: The edge device transmits the monitoring information to the cloud platform in real time through the Internet of Things, and encrypts the transmitted monitoring information using an encryption protocol;

[0146] S308: Monitor the connection status between the edge device and the cloud platform by adopting a heartbeat mechanism.

[0147] The working principle of the above technical solution is as follows: the edge device performs outlier detection and filtering algorithm processing on the acquired raw data. By detecting outliers and applying filtering algorithms, outliers and noise data points can be removed to improve the accuracy and reliability of the data; the raw data is compressed and normalized. The compression algorithm is used to compress the data to reduce the storage and transmission costs of the data. At the same time, the data with different features are normalized and standardized, and they are mapped to the same magnitude to facilitate subsequent analysis and processing; the data mapped to the same magnitude is feature extracted. Through the feature extraction algorithm, key information is extracted from the data, the dimension of the data is reduced, and important features are retained for subsequent calculation and analysis; multiple subtasks are divided and parallel calculations are performed. The feature-extracted data is divided into multiple subtasks, and these subtasks are processed by parallel calculation to quickly obtain the pre-processed data after feature extraction; the pre-processed data is analyzed and processed using machine learning algorithms and pattern recognition algorithms. Through machine learning and pattern recognition algorithms, the pre-processed data is analyzed to extract relevant information such as forest conditions, ecological environment, and vegetation generation; monitoring information is generated and encrypted for transmission. According to the results of data analysis and processing, corresponding monitoring information is generated, such as temperature and humidity ratio, fire risk level, vegetation coverage, soil moisture and stability, and pest and disease risk. The edge device transmits this monitoring information to the cloud platform in real time through the Internet of Things, and uses encryption protocols to encrypt the transmitted data to ensure data security and privacy; monitor the connection status. The heartbeat mechanism is used to monitor the connection status between the edge device and the cloud platform to ensure the stability and reliability of the connection.

[0148] The effect of the above technical solution is: the edge device removes outliers and noise data points in the original data through detection and filtering algorithms, thereby improving the accuracy and reliability of the data. The compression algorithm compresses the original data, reducing the storage and transmission costs. Similarly, normalization and standardization processing maps data with different features to the same magnitude, facilitating subsequent calculations and analysis; feature extraction reduces the dimension of the data and retains key information, improving the efficiency and accuracy of data processing; through parallel computing, the processing of data after feature extraction can be accelerated to improve computing efficiency; machine learning algorithms and pattern recognition algorithms analyze and process the pre-processed data after feature extraction to extract relevant information, such as forest conditions, ecological environment, and vegetation generation; the generated monitoring information includes temperature and humidity ratio, fire risk level, vegetation coverage, soil moisture and stability, and pest and disease risks, which can provide comprehensive forest monitoring information. The edge device transmits the monitoring information to the cloud platform in real time through the Internet of Things, and uses encryption protocols to encrypt the transmitted monitoring information to ensure the security and privacy of the data. The heartbeat mechanism is used to monitor the connection status between the edge device and the cloud platform to ensure the stability and reliability of the connection.

[0149] In one embodiment of the present invention, the edge device preprocesses the acquired raw data, and performs data analysis and processing on the preprocessed raw data through a machine learning algorithm and a pattern recognition algorithm to obtain analysis and processing results. The edge device generates monitoring information based on the analysis and processing results, and transmits the monitoring information to the cloud platform through the Internet of Things, including:

[0150] S401: The cloud platform receives the monitoring information transmitted by the edge device, and creates a storage directory in the storage bucket according to the specific situation of the monitoring information, wherein the specific situation of the monitoring information includes the name of the monitoring point and the time period;

[0151] S402: After the storage directory is created, the received monitoring information is uploaded to the corresponding storage directory;

[0152] S403: According to the characteristics and objectives of forestry monitoring, feature extraction and selection of data are performed through machine learning algorithms; for example, taking tree growth monitoring as an example, by installing IoT devices to monitor forest growth in real time in the forest area, and extracting and selecting features of monitoring data through machine learning algorithms, the accuracy and efficiency of monitoring data can be improved. Based on the results of data analysis, decision makers can conduct more refined management and planning of forest resources and promote sustainable development of forestry.

[0153] S404: establishing a management and monitoring model through the feature extraction result, an artificial intelligence algorithm and a digital elevation model, and training and verifying the established management and monitoring model through historical monitoring data;

[0154] S405: further analyzing the newly collected monitoring data using the trained management monitoring model to obtain further analysis results, wherein the further analysis includes trend analysis, anomaly detection, and pattern recognition;

[0155] S406: Perform monitoring and management based on the analysis results, including making decisions, optimizing forestry management strategies, generating reports, and providing real-time monitoring data visualization.

[0156] The working principle of the above technical solution is as follows: after the cloud platform receives the monitoring information transmitted by the edge device, it creates a corresponding storage directory in the storage bucket according to the specific situation of the monitoring information (monitoring point name and time period); after creating the storage directory, the cloud platform uploads the received monitoring information to the corresponding storage directory to realize data archiving and storage; according to the characteristics and goals of forestry monitoring, the cloud platform uses machine learning algorithms to extract and select features from the data to extract valuable features; using the results of feature extraction and digital elevation models, the cloud platform establishes a management monitoring model. The model is trained and verified by using historical monitoring data to ensure the accuracy and reliability of the model; using the trained management monitoring model, the cloud platform further analyzes the newly collected monitoring data. This includes trend analysis, anomaly detection and pattern recognition to obtain more in-depth analysis results; based on the analysis results, the cloud platform performs monitoring management, including making decisions, optimizing forestry management strategies, generating reports and providing visualization of real-time monitoring data. This enables users to understand the status of the forest area in a timely manner and take corresponding management measures.

[0157] The effects of the above technical solutions are as follows: the cloud platform receives and processes the monitoring information transmitted by the edge device, automatically creates a storage directory and uploads the data, reducing the possibility of manual operation and errors; by creating a storage directory in the storage bucket and uploading the monitoring information to the corresponding directory, effective management and archiving of the monitoring data are achieved, which is convenient for subsequent access and use; the data is feature extracted and selected through machine learning algorithms, and representative and important features are screened out from massive data, which improves the efficiency and accuracy of data analysis; a management and monitoring model is established by combining artificial intelligence algorithms with digital elevation models, which can more accurately reflect the situation of forestry monitoring and provide a basis for subsequent analysis and decision-making; the newly collected monitoring data is further analyzed using the trained management and monitoring model, and trend analysis, anomaly detection and pattern recognition are performed to obtain more in-depth analysis results and reveal the potential information and laws in the data; monitoring management is carried out based on the analysis results, including making decisions, optimizing forestry management strategies, generating reports and providing visualization of real-time monitoring data, which helps decision makers to conduct forestry management and planning more scientifically.

[0158] One embodiment of the present invention is a smart forestry monitoring system based on edge computing, the system comprising:

[0159] Digital elevation model building module: obtains forest area topography-related data, and obtains a digital elevation model of the forest area topography based on processing the obtained relevant data; stores the obtained digital elevation model in a cloud platform;

[0160] Raw data acquisition module: Analyze the digital elevation model of the forest area topography stored in the cloud platform through artificial intelligence algorithms, determine the sensor deployment location according to the analysis results, and deploy corresponding sensors according to the sensor deployment location, collect raw data through sensors, connect edge devices with sensors, and obtain raw data collected by sensors in real time;

[0161] Monitoring information generation module: The edge device pre-processes the acquired raw data, and performs data analysis and processing on the pre-processed raw data through machine learning algorithms and pattern recognition algorithms to obtain analysis and processing results. The edge device generates monitoring information based on the analysis and processing results, and transmits the monitoring information to the cloud platform through the Internet of Things;

[0162] Monitoring and management module: The cloud platform receives the monitoring information transmitted by the edge device, combines the digital elevation model with the artificial intelligence algorithm, further analyzes and processes the monitoring information, obtains further analysis and processing results, and performs monitoring and management based on the further analysis and processing results.

[0163] The working principle of the above technical solution is as follows: first, obtain the relevant data of the forest area topography, which can be obtained through surveying, remote sensing or laser radar and other technical means. Then, process the acquired data to establish a digital elevation model of the forest area topography. The obtained digital elevation model is stored in the cloud platform for subsequent analysis and processing; on the cloud platform, the stored digital elevation model is analyzed by artificial intelligence algorithm. Through the analysis results, the sensor deployment location is determined, and a suitable location is selected to monitor the environment and potential risks of the forest area. According to the deployment location, the corresponding sensor is installed in the forest area and connected to the edge device to obtain the raw data collected by the sensor in real time; the edge device receives the raw data collected by the sensor and preprocesses it. This includes data cleaning, noise filtering, data normalization and other operations. Then, the preprocessed data is analyzed and processed by machine learning algorithms and pattern recognition algorithms to obtain analysis results. The edge device generates monitoring information based on these results and transmits it to the cloud platform through the Internet of Things; the cloud platform receives the monitoring information transmitted by the edge device, and combines the digital elevation model to further analyze and process the monitoring information using artificial intelligence algorithm. On this basis, more in-depth analysis and processing results are generated. Based on these results, monitoring and management, such as environmental assessment and risk early warning, are carried out to provide effective forest monitoring and management decision support.

[0164] The effects of the above technical solutions are: by collecting data in real time through sensors, and using artificial intelligence algorithms for processing and analysis, real-time monitoring of the forest environment can be achieved. It helps to detect abnormal situations and potential risks in a timely manner so that appropriate measures can be taken to manage and respond; by establishing a digital elevation model and adopting a machine learning algorithm, the solution can provide high-precision analysis results; it can help forest managers to more accurately understand the topographic characteristics of the forest area, as well as the trends of environmental changes and potential risks, and provide a scientific basis for decision-making; the cloud platform combines digital elevation models and monitoring information to generate in-depth analysis and processing results through artificial intelligence algorithms; it can be used for decision support in environmental assessment, risk warning and other aspects. Forest managers can formulate effective management strategies based on these results to improve the sustainable development and protection level of forest areas; by integrating digital elevation models, sensor data collection and artificial intelligence algorithm analysis and processing, the overall management is optimized. Forest managers can conduct centralized monitoring and management on the cloud platform to improve management efficiency and resource utilization; through real-time monitoring and analysis and processing, the solution can detect potential risks and environmental problems early and provide early warning information; it helps forest managers to take timely measures to reduce losses and impacts and improve emergency response capabilities.

[0165] In one embodiment of the present invention, the digital elevation model building module includes:

[0166] Method determination module: determine the location information of the forest area to be monitored, and determine the method for obtaining the topographic and geomorphic related data of the forest area according to the location information of the forest area to be monitored; the related data include elevation measurement data, laser radar scanning data, aerial remote sensing images and satellite images;

[0167] Topography and geomorphology acquisition module: collects relevant data on forest topography and geomorphology according to the determined acquisition method, which includes aerial survey, ground survey and satellite remote sensing; for example, for a forest area located in a mountainous area, aerial survey can be used to obtain the elevation data and vegetation and other related information of the forest area, and ground survey is required for the accurate details of each point; for forest areas in plain areas, satellite remote sensing technology can be used to quickly obtain relevant topography and geomorphology data. Use these different methods to acquire data, and choose the appropriate method according to the specific situation to achieve the best effect.

[0168] Data preprocessing module: preprocess the collected data related to the forest area topography and obtain the elevation data of the forest area topography; and fuse different elevation data through data fusion algorithm and weighted average algorithm, and the preprocessing includes noise removal, adjustment processing and data correction;

[0169] Interpolation algorithm execution module: through the interpolation algorithm, the discrete elevation data points are interpolated into a continuous elevation surface and the corresponding elevation data model is generated;

[0170] A grid format conversion module: converting the interpolated continuous elevation data model into a grid format, wherein the grid format includes a regular grid and an irregular grid, wherein the regular grid includes a square grid and a rectangular grid, and the irregular grid includes a triangular grid;

[0171] Smoothing correction module: Smoothing and correcting the generated digital elevation model through filtering and interpolation, and removing local outliers and errors;

[0172] Format export module: export the generated digital elevation model to a data file in a standard format, including GeoTIFF and ASCII;

[0173] Storage requirement determination module: Determine the format of the digital elevation model, and determine the storage requirement based on the data size, access frequency and expected growth data of the digital elevation model; for example: assuming that the digital elevation model of a certain area is stored in a raster format, with a spatial resolution of 10 meters, a coverage area of ​​100km×100km, and the elevation value of each grid unit is 2 bytes (16 bits), then its data size is 10000×10000×2=200MB. If the digital elevation model needs to acquire and process data in real time, a cloud storage solution with high bandwidth and low latency can be selected. If the data growth is expected to be large in the next few years, a highly scalable object-oriented storage or distributed file system can be selected for storage.

[0174] A grid division module: according to the spatial range of the digital elevation model, the digital elevation model is divided into a series of spatial regions according to a grid division method;

[0175] Bucket creation module: creates a bucket in the cloud platform, and creates a corresponding storage directory structure based on the divided space area, grid ID, and row and column numbers;

[0176] Storage corresponding module: divide the divided digital elevation model into grids according to their grids and store them in the corresponding storage directory in sequence;

[0177] Permission setting module: establish backup and disaster recovery mechanisms for digital elevation model data stored in the storage directory, and set usage permissions and access control policies;

[0178] Update module: adding corresponding metadata information in the cloud platform, wherein the metadata information includes the spatial range, resolution and update date of the grid;

[0179] Optimization and upgrade module: Regularly optimize and upgrade the cloud platform according to actual usage and demand changes. For example, regularly perform security reinforcement and patch updates on the cloud platform based on the latest security threats and vulnerability intelligence. Timely update security patches for operating systems and software, strengthen the configuration of network firewalls and intrusion detection systems, conduct penetration tests and security audits, etc. to ensure the security of the cloud platform.

[0180] The working principle of the above technical solution is: determine the location of the monitored forest area, and decide the method of obtaining relevant data according to the location; collect relevant data on the topography of the forest area according to the determined acquisition method, including elevation measurement data, lidar scanning data, aerial remote sensing images and satellite images, and data collection can be carried out through aerial measurement, ground survey and satellite remote sensing; pre-process the collected forest area topography related data, including noise removal, adjustment processing and data correction operations to obtain accurate elevation data; perform data fusion and weighted average processing on different elevation data, and then use interpolation algorithms to interpolate discrete elevation data points into continuous elevation surfaces to generate elevation data models; convert the interpolated continuous elevation data model into a grid format, including regular grids and irregular grids, such as square, rectangular and triangular grids; smooth and correct the generated digital elevation model through filtering and interpolation algorithms to remove Local outliers and errors can be eliminated to improve the quality and accuracy of data; the generated digital elevation model can be exported to data files in standard formats, such as GeoTIFF and ASCII, and the storage requirements can be determined based on the format, size, access frequency and expected growth data of the digital elevation model; the digital elevation model can be divided into a series of spatial areas according to its spatial range, and storage buckets and corresponding storage directory structures can be created in the cloud platform to store the digital elevation model in sequence according to grid ID and row and column numbers; a backup and disaster recovery mechanism can be established to ensure the security and reliability of digital elevation model data, and usage permissions and access control policies can be set to protect the confidentiality and integrity of the data; metadata information related to the digital elevation model can be added to the cloud platform, including spatial range, resolution, update date and other information, to facilitate subsequent management and retrieval; according to actual usage and changes in demand, the cloud platform can be regularly optimized and upgraded to improve system performance and user experience.

[0181] The effects of the above technical solutions are as follows: through a variety of acquisition methods, including elevation measurement data, lidar scanning data, aerial remote sensing images and satellite images, the forest area topography-related data obtained are more comprehensive; by processing the acquired data through preprocessing, data fusion and interpolation algorithms, the digital elevation model of the forest area topography can be accurately obtained to improve the accuracy and precision of the data; the digital elevation model is stored in the cloud platform, and the digital elevation model data can be easily managed and retrieved by dividing the spatial area and creating a storage directory structure; a backup and disaster recovery mechanism is established to ensure the security and reliability of the digital elevation model data, and at the same time, usage permissions and access control policies are set to protect the confidentiality and integrity of the data; metadata information related to the digital elevation model is added to the cloud platform, including spatial range, resolution and update date, etc., to facilitate subsequent management and retrieval; according to actual usage and changes in demand, the cloud platform is regularly optimized and upgraded to improve system performance and user experience.

[0182] In one embodiment of the present invention, the raw data acquisition module includes:

[0183] Relevant information acquisition module: obtains digital elevation model data of forest area topography from the cloud platform, processes and analyzes the digital elevation model data through artificial intelligence algorithms, and obtains relevant information of the forest area topography, including topography, height, slope and hydrology;

[0184] Sensor type determination module: determines the deployment location and quantity of sensors according to the information, and determines the type of sensors to be used according to the characteristics of the forest area and the information to be monitored;

[0185] Installation plan determination module: selects appropriate locations and plans for sensor deployment according to the determined sensor locations, quantity, and sensor types to be used; the sensor types include temperature and humidity sensors, fire risk index sensors, geological monitoring sensors, and image sensors;

[0186] Temperature and humidity sensor module: The temperature and humidity sensor is used to monitor the climate and humidity changes in the forest area; the temperature and humidity sensor is deployed at a position 1.5-3m above the ground. The deployment rule of the temperature and humidity sensor is to set two temperature and humidity sensors for every 1250 square meters. Let the forest area be S, and the number of temperature and humidity sensors set be N. If S-1250≤0, then N=2, if S-1250>0, then

[0187] Fire risk index sensor module: The fire risk index sensor is used to assess the fire risk in the forest area. The fire risk index sensor is deployed in the forest area near the fire source, flammable materials and the edge of the forest area. The number of the fire risk index sensors is not less than 16; for example: for potential fire sources, sensors can be deployed in areas where fires often occur, such as camps, barbecue areas, camping sites or outdoor fire areas. For areas near flammable vegetation, sensors can be arranged near easily burned vegetation areas, such as dry grass, forest edges or bushes. For monitoring entrance and exit areas, sensors can be set at the entrance and exit of the forest area to monitor the fire risk of people entering and leaving, so that timely measures can be taken to prevent the spread of fire.

[0188] Geological monitoring sensor module: The geological monitoring sensor is used to detect soil moisture and soil stability. Twelve geological monitoring sensors are randomly deployed in the forest area and installed next to trees without direct contact with the trees.

[0189] Image sensor module: The image sensor is used to monitor the growth status of vegetation and the situation of pests and diseases. The image sensor is installed in a representative location with a wide field of view, no obstacles, no direct sunlight, and the representative location includes forests of different types and woodlands of different ages. The number of image sensors installed is one per thousand square meters;

[0190] Edge device installation module: After the sensor is deployed, an edge device is installed next to the sensor, and the edge device is connected to the sensor through a low-power wireless communication protocol, which includes LoRaWAN and NB-IoT;

[0191] Real-time data transmission module: collects raw data through sensors and transmits the data collected by sensors to edge devices in real time.

[0192] The working principle of the above technical solution is: obtain digital elevation model data of forest area topography through the cloud platform, use artificial intelligence algorithms to process and analyze the data, and extract relevant information such as terrain, height, slope and hydrology; according to the characteristics of the forest area and monitoring needs, determine the type of sensors to be used, including temperature and humidity sensors, fire risk index sensors, geological monitoring sensors and image sensors; based on the determined sensor type and monitoring target, determine the deployment location and number of sensors. Temperature and humidity sensors are deployed at a distance of 1.5-3m from the ground according to the rules, with two sensors set up for every 1,250 square meters; fire risk index sensors are deployed near fire sources, flammable materials and the edge of forest areas, with a number of no less than 16; 12 geological monitoring sensors are randomly deployed, placed next to trees in the forest area, and are not in direct contact with the trees; image sensors are installed in open locations at a density of one per thousand square meters, such as different types of forests and woodlands of different ages; edge devices are installed next to the sensors, and low-power wireless communication protocols (such as LoRaWAN and NB-IoT) are used to connect the edge devices to the sensors to realize data collection and transmission; the sensors collect environmental data through edge devices, and transmit the collected data to the edge devices in real time to form a real-time monitoring system.

[0193] The effects of the above technical solutions are as follows: by processing and analyzing the digital elevation model data through artificial intelligence algorithms, relevant information such as forest terrain, height, slope, and hydrology can be obtained, providing detailed terrain data for forest management and planning; according to the terrain information obtained and monitoring needs, the deployment location and number of sensors are determined to ensure that the monitoring coverage is wide and accurate. Different types of sensors (temperature and humidity, fire risk index, geological monitoring, and images) are deployed in appropriate locations to comprehensively monitor the climate, fire risk, soil conditions, and vegetation growth status of the forest area; through the connection between edge devices and sensors, real-time transmission of sensor collected data is realized. In this way, climate change, fire risk, soil conditions, and vegetation conditions in the forest area can be understood in a timely manner, and the data can be analyzed and processed; combining the data of multiple sensors, the fire risk, soil moisture, stability, and vegetation conditions in the forest area can be more accurately assessed. This helps to detect abnormal situations in a timely manner and improve the monitoring capabilities of natural disasters and ecological environment; through automated data collection and transmission, the workload of manual collection and input is reduced, saving time and labor costs; using low-power wireless communication protocols (such as LoRaWAN and NB-IoT) to ensure data security and reliability during transmission.

[0194] In one embodiment of the present invention, the monitoring information generation module includes:

[0195] Data filtering module: The edge device detects the acquired raw data to remove outliers and removes noise data points through filtering algorithms;

[0196] Data compression module: compresses the original data through a compression algorithm, normalizes and standardizes the data with different characteristics, and maps the normalized and standardized data to the same magnitude;

[0197] Feature extraction module: extracting features from the data mapped to the same magnitude, reducing data dimensions and retaining key information;

[0198] Subtask division module: divides the feature-extracted data into multiple subtasks, calculates the subtasks through parallel computing, and obtains pre-processed data after feature extraction;

[0199] Feature analysis and processing module: analyzes and processes the pre-processed data after feature extraction through machine learning algorithms and pattern recognition algorithms to obtain corresponding information, including forest conditions, ecological environment, and vegetation generation conditions;

[0200] Monitoring information generation module: generates monitoring information based on the results of data analysis and processing, and the monitoring information includes temperature and humidity ratio, fire risk level, vegetation coverage, soil moisture and stability, and pest and disease risk;

[0201] Transmission encryption module: The edge device transmits the monitoring information to the cloud platform in real time through the Internet of Things, and uses an encryption protocol to encrypt the transmitted monitoring information;

[0202] Status monitoring module: monitors the connection status between the edge device and the cloud platform by adopting a heartbeat mechanism.

[0203] The working principle of the above technical solution is as follows: the edge device performs outlier detection and filtering algorithm processing on the acquired raw data. By detecting outliers and applying filtering algorithms, outliers and noise data points can be removed to improve the accuracy and reliability of the data; the raw data is compressed and normalized. The compression algorithm is used to compress the data to reduce the storage and transmission costs of the data. At the same time, the data with different features are normalized and standardized, and they are mapped to the same magnitude to facilitate subsequent analysis and processing; the data mapped to the same magnitude is feature extracted. Through the feature extraction algorithm, key information is extracted from the data, the dimension of the data is reduced, and important features are retained for subsequent calculation and analysis; multiple subtasks are divided and parallel calculations are performed. The feature-extracted data is divided into multiple subtasks, and these subtasks are processed by parallel calculation to quickly obtain the pre-processed data after feature extraction; the pre-processed data is analyzed and processed using machine learning algorithms and pattern recognition algorithms. Through machine learning and pattern recognition algorithms, the pre-processed data is analyzed to extract relevant information such as forest conditions, ecological environment, and vegetation generation; monitoring information is generated and encrypted for transmission. According to the results of data analysis and processing, corresponding monitoring information is generated, such as temperature and humidity ratio, fire risk level, vegetation coverage, soil moisture and stability, and pest and disease risk. The edge device transmits this monitoring information to the cloud platform in real time through the Internet of Things, and uses encryption protocols to encrypt the transmitted data to ensure data security and privacy; monitor the connection status. The heartbeat mechanism is used to monitor the connection status between the edge device and the cloud platform to ensure the stability and reliability of the connection.

[0204] The effect of the above technical solution is: the edge device removes outliers and noise data points in the original data through detection and filtering algorithms, thereby improving the accuracy and reliability of the data. The compression algorithm compresses the original data, reducing the storage and transmission costs. Similarly, normalization and standardization processing maps data with different features to the same magnitude, facilitating subsequent calculations and analysis; feature extraction reduces the dimension of the data and retains key information, improving the efficiency and accuracy of data processing; through parallel computing, the processing of data after feature extraction can be accelerated to improve computing efficiency; machine learning algorithms and pattern recognition algorithms analyze and process the pre-processed data after feature extraction to extract relevant information, such as forest conditions, ecological environment, and vegetation generation; the generated monitoring information includes temperature and humidity ratio, fire risk level, vegetation coverage, soil moisture and stability, and pest and disease risks, which can provide comprehensive forest monitoring information. The edge device transmits the monitoring information to the cloud platform in real time through the Internet of Things, and uses encryption protocols to encrypt the transmitted monitoring information to ensure the security and privacy of the data. The heartbeat mechanism is used to monitor the connection status between the edge device and the cloud platform to ensure the stability and reliability of the connection.

[0205] In one embodiment of the present invention, the monitoring management module includes:

[0206] Cloud platform receiving module: The cloud platform receives the monitoring information transmitted by the edge device and creates a storage directory in the storage bucket according to the specific situation of the monitoring information, wherein the specific situation of the monitoring information includes the name of the monitoring point and the time period;

[0207] Information storage corresponding module: after creating the storage directory, the received monitoring information is uploaded to the corresponding storage directory;

[0208] Machine learning module: According to the characteristics and objectives of forestry monitoring, feature extraction and selection of data are performed through machine learning algorithms; for example, taking tree growth monitoring as an example, by installing IoT devices to monitor forest growth in real time in the forest area, and extracting and selecting features of monitoring data through machine learning algorithms, the accuracy and efficiency of monitoring data can be improved. Based on the results of data analysis, decision makers can manage and plan forest resources more finely and promote the sustainable development of forestry.

[0209] Management and monitoring model building module: Based on the results of feature extraction, the management and monitoring model is built through artificial intelligence algorithms combined with digital elevation models, and the established management and monitoring model is trained and verified through historical monitoring data;

[0210] Model detection module: Use the trained management monitoring model to further analyze the newly collected monitoring data to obtain further analysis results, including trend analysis, anomaly detection and pattern recognition;

[0211] Monitoring and management implementation module: monitoring and management are carried out according to the analysis results, including making decisions, optimizing forestry management strategies, generating reports and providing real-time monitoring data visualization.

[0212] The working principle of the above technical solution is as follows: after the cloud platform receives the monitoring information transmitted by the edge device, it creates a corresponding storage directory in the storage bucket according to the specific situation of the monitoring information (monitoring point name and time period); after creating the storage directory, the cloud platform uploads the received monitoring information to the corresponding storage directory to realize data archiving and storage; according to the characteristics and goals of forestry monitoring, the cloud platform uses machine learning algorithms to extract and select features from the data to extract valuable features; using the results of feature extraction and digital elevation models, the cloud platform establishes a management monitoring model. The model is trained and verified by using historical monitoring data to ensure the accuracy and reliability of the model; using the trained management monitoring model, the cloud platform further analyzes the newly collected monitoring data. This includes trend analysis, anomaly detection and pattern recognition to obtain more in-depth analysis results; based on the analysis results, the cloud platform performs monitoring management, including making decisions, optimizing forestry management strategies, generating reports and providing visualization of real-time monitoring data. This enables users to understand the status of the forest area in a timely manner and take corresponding management measures.

[0213] The effects of the above technical solutions are as follows: the cloud platform receives and processes the monitoring information transmitted by the edge device, automatically creates a storage directory and uploads the data, reducing the possibility of manual operation and errors; by creating a storage directory in the storage bucket and uploading the monitoring information to the corresponding directory, effective management and archiving of the monitoring data are achieved, which is convenient for subsequent access and use; the data is feature extracted and selected through machine learning algorithms, and representative and important features are screened out from massive data, which improves the efficiency and accuracy of data analysis; a management and monitoring model is established by combining artificial intelligence algorithms with digital elevation models, which can more accurately reflect the situation of forestry monitoring and provide a basis for subsequent analysis and decision-making; the newly collected monitoring data is further analyzed using the trained management and monitoring model, and trend analysis, anomaly detection and pattern recognition are performed to obtain more in-depth analysis results and reveal the potential information and laws in the data; monitoring management is carried out based on the analysis results, including making decisions, optimizing forestry management strategies, generating reports and providing visualization of real-time monitoring data, which helps decision makers to conduct forestry management and planning more scientifically.

[0214] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A smart forestry monitoring method based on edge computing, characterized in that: The method comprises: Acquire the forest area topography-related data, and acquire the digital elevation model of the forest area topography based on the acquired relevant data; store the acquired digital elevation model in the cloud platform; The digital elevation model of the forest area topography stored in the cloud platform is analyzed by an artificial intelligence algorithm, the sensor deployment location is determined according to the analysis result, and the corresponding sensor is deployed according to the sensor deployment location, the raw data is collected by the sensor, the edge device is connected to the sensor, and the raw data collected by the sensor is obtained in real time; The edge device preprocesses the acquired raw data, and performs data analysis and processing on the preprocessed raw data through machine learning algorithms and pattern recognition algorithms to obtain analysis and processing results. The edge device generates monitoring information based on the analysis and processing results, and transmits the monitoring information to the cloud platform through the Internet of Things; The cloud platform receives the monitoring information transmitted by the edge device, further analyzes and processes the monitoring information in combination with the digital elevation model through an artificial intelligence algorithm, obtains further analysis and processing results, and performs monitoring and management according to the further analysis and processing results; The step of obtaining the forest area topography-related data and obtaining the digital elevation model of the forest area topography based on the obtained relevant data; and storing the obtained digital elevation model in a cloud platform includes: Determine the location information of the forest area to be monitored, and determine the method for obtaining the topographic and geomorphic data of the forest area according to the location information of the forest area to be monitored; Collecting relevant data on the topography of the forest area according to the determined acquisition method, which includes aerial survey, ground survey and satellite remote sensing; Preprocessing the collected data related to the forest area topography to obtain the elevation data of the forest area topography; fusing different elevation data through data fusion algorithm and weighted average algorithm, wherein the preprocessing includes noise removal, adjustment processing and data correction; Through the interpolation algorithm, the discrete elevation data points are interpolated into a continuous elevation surface, and the corresponding elevation data model is generated; Converting the interpolated continuous elevation data model into a grid format, wherein the grid format includes a regular grid and an irregular grid, wherein the regular grid includes a square grid and a rectangular grid, and the irregular grid includes a triangular grid; The generated digital elevation model is smoothed and corrected through filtering and interpolation, and local outliers and errors are removed; Exporting the generated digital elevation model to a data file in a standard format, including GeoTIFF and ASCII; Determining the format of the digital elevation model and determining storage requirements based on the data size, access frequency, and expected growth data of the digital elevation model; According to the spatial range of the digital elevation model, the digital elevation model is divided into a series of spatial regions according to a grid division method; Create a storage bucket in the cloud platform, and create a corresponding storage directory structure based on the divided space area, grid ID, and row and column numbers; Divide the divided digital elevation model into grids and store them in corresponding storage directories in sequence; For the digital elevation model data stored in the storage directory, establish a backup and disaster recovery mechanism, and set usage permissions and access control policies; Add corresponding metadata information in the cloud platform, wherein the metadata information includes the spatial range, resolution and update date of the grid; Regularly optimize and upgrade the cloud platform according to actual usage and changes in demand; The digital elevation model of the forest area topography stored in the cloud platform is analyzed by an artificial intelligence algorithm, the sensor deployment location is determined according to the analysis result, and the corresponding sensor is deployed according to the sensor deployment location, raw data is collected by the sensor, the edge device is connected to the sensor, and the raw data collected by the sensor is obtained in real time, including: Acquire digital elevation model data of forest area topography from the cloud platform, process and analyze the digital elevation model data through an artificial intelligence algorithm, and obtain relevant information of the forest area topography, wherein the relevant information of the forest area topography includes topography, height, slope, and hydrology; Determine the deployment location and quantity of sensors based on the relevant information of the forest area terrain, and determine the type of sensors to be used based on the characteristics of the forest area and the information to be monitored; According to the determined sensor locations, quantity and types of sensors to be used, select appropriate locations and plans for sensor deployment; the sensor types include temperature and humidity sensors, fire risk index sensors, geological monitoring sensors and image sensors; The temperature and humidity sensor is used to monitor the climate and humidity changes in the forest area; the temperature and humidity sensor is deployed at a position 1.5-3m above the ground. The deployment rule of the temperature and humidity sensor is to set two temperature and humidity sensors for every 1250 square meters. Let the forest area be S, and the number of temperature and humidity sensors set be N. If S-1250≤0, then N=2, if S-1250>0, then ; The fire risk index sensor is used to assess the fire risk in the forest area. The fire risk index sensor is deployed in the forest area close to the fire source, flammable materials and the edge of the forest area. The number of the fire risk index sensors is not less than 16; The geological monitoring sensors are used to detect soil moisture and soil stability. 12 of the geological monitoring sensors are randomly deployed in the forest area and installed next to trees without direct contact with the trees. The image sensor is used to monitor the growth status of vegetation and the situation of pests and diseases. The image sensor is installed in a representative location with a wide field of view, no obstacles, no direct sunlight, and the representative location includes forests of different types and woodlands of different ages. The number of image sensors installed is one per thousand square meters; After the sensor is deployed, an edge device is installed next to the sensor, and the edge device is connected to the sensor through a low-power wireless communication protocol, which includes LoRaWAN and NB-IoT; The raw data is collected through sensors and transmitted to edge devices in real time.

2. According to the smart forestry monitoring method based on edge computing as described in claim 1, The feature is that the edge device pre-processes the acquired raw data, and performs data analysis and processing on the pre-processed raw data through a machine learning algorithm and a pattern recognition algorithm to obtain analysis and processing results. The edge device generates monitoring information based on the analysis and processing results, and transmits the monitoring information to the cloud platform through the Internet of Things, including: The edge device detects the acquired raw data to remove outliers and removes noise data points through a filtering algorithm; Compressing the original data by using a compression algorithm, normalizing and standardizing the data with different characteristics, and mapping the normalized and standardized data to the same magnitude; Extracting features from the data mapped to the same magnitude to reduce data dimensions and retain key information; Divide the feature-extracted data into multiple subtasks, and calculate the subtasks through parallel computing to obtain pre-processed data after feature extraction; Analyze and process the pre-processed data after feature extraction through machine learning algorithms and pattern recognition algorithms to obtain corresponding information, including forest conditions, ecological environment, and vegetation generation conditions; Generate monitoring information based on the results of data analysis and processing, the monitoring information including temperature and humidity ratio, fire risk level, vegetation coverage, soil moisture and stability, and pest and disease risk; The edge device transmits the monitoring information to the cloud platform in real time through the Internet of Things, and encrypts the transmitted monitoring information using an encryption protocol; The connection status between the edge device and the cloud platform is monitored by adopting a heartbeat mechanism.

3. According to the edge computing-based smart forestry monitoring method of claim 1, it is characterized in that: The edge device preprocesses the acquired raw data, and performs data analysis and processing on the preprocessed raw data through a machine learning algorithm and a pattern recognition algorithm to obtain analysis and processing results. The edge device generates monitoring information based on the analysis and processing results, and transmits the monitoring information to the cloud platform through the Internet of Things, including: The cloud platform receives the monitoring information transmitted by the edge device, and creates a storage directory in the storage bucket according to the specific situation of the monitoring information, wherein the specific situation of the monitoring information includes the name of the monitoring point and the time period; After the storage directory is created, the received monitoring information is uploaded to the corresponding storage directory; According to the characteristics and objectives of forestry monitoring, feature extraction and selection of data are performed through machine learning algorithms; Based on the results of feature extraction, a management and monitoring model is established by combining artificial intelligence algorithms with digital elevation models, and the established management and monitoring model is trained and verified through historical monitoring data; Using the trained management monitoring model to further analyze the newly collected monitoring data to obtain further analysis results, wherein the further analysis includes trend analysis, anomaly detection, and pattern recognition; Monitoring and management are carried out based on the analysis results, including making decisions, optimizing forestry management strategies, generating reports and providing real-time monitoring data visualization.

4. A smart forestry monitoring system based on edge computing, characterized in that: The system comprises: Digital elevation model building module: obtains forest area topography-related data, and obtains a digital elevation model of the forest area topography based on processing the obtained relevant data; stores the obtained digital elevation model in a cloud platform; Raw data acquisition module: Analyze the digital elevation model of the forest area topography stored in the cloud platform through artificial intelligence algorithms, determine the sensor deployment location according to the analysis results, and deploy corresponding sensors according to the sensor deployment location, collect raw data through sensors, connect edge devices with sensors, and obtain raw data collected by sensors in real time; Monitoring information generation module: The edge device pre-processes the acquired raw data, and performs data analysis and processing on the pre-processed raw data through machine learning algorithms and pattern recognition algorithms to obtain analysis and processing results. The edge device generates monitoring information based on the analysis and processing results, and transmits the monitoring information to the cloud platform through the Internet of Things; Monitoring and management module: the cloud platform receives the monitoring information transmitted by the edge device, further analyzes and processes the monitoring information in combination with the digital elevation model through an artificial intelligence algorithm, obtains further analysis and processing results, and performs monitoring and management according to the further analysis and processing results; The digital elevation model building module includes: Method determination module: determine the location information of the forest area to be monitored, and determine the method for obtaining the topography-related data of the forest area according to the location information of the forest area to be monitored; Topography and geomorphology acquisition module: collects relevant data of forest topography and geomorphology according to the determined acquisition method, and the acquisition method includes aerial survey, ground survey and satellite remote sensing; Data preprocessing module: preprocess the collected data related to the forest area topography and obtain the elevation data of the forest area topography; and fuse different elevation data through data fusion algorithm and weighted average algorithm, and the preprocessing includes noise removal, adjustment processing and data correction; Interpolation algorithm execution module: through the interpolation algorithm, the discrete elevation data points are interpolated into a continuous elevation surface and the corresponding elevation data model is generated; A grid format conversion module: converting the interpolated continuous elevation data model into a grid format, wherein the grid format includes a regular grid and an irregular grid, wherein the regular grid includes a square grid and a rectangular grid, and the irregular grid includes a triangular grid; Smoothing correction module: Smoothing and correcting the generated digital elevation model through filtering and interpolation, and removing local outliers and errors; Format export module: export the generated digital elevation model to a data file in a standard format, including GeoTIFF and ASCII; Storage requirement determination module: determining the format of the digital elevation model, and determining storage requirements according to the data size, access frequency and expected growth data of the digital elevation model; A grid division module: according to the spatial range of the digital elevation model, the digital elevation model is divided into a series of spatial regions according to a grid division method; Bucket creation module: creates a bucket in the cloud platform, and creates a corresponding storage directory structure based on the divided space area, grid ID, and row and column numbers; Storage corresponding module: divide the divided digital elevation model into grids according to their grids and store them in the corresponding storage directory in sequence; Permission setting module: establish backup and disaster recovery mechanisms for digital elevation model data stored in the storage directory, and set usage permissions and access control policies; Update module: adding corresponding metadata information in the cloud platform, wherein the metadata information includes the spatial range, resolution and update date of the grid; Optimization and upgrade module: regularly optimize and upgrade the cloud platform according to actual usage and demand changes; The raw data acquisition module comprises: Relevant information acquisition module: obtains digital elevation model data of forest area topography from the cloud platform, processes and analyzes the digital elevation model data through artificial intelligence algorithms, and obtains relevant information of the forest area topography, including topography, height, slope and hydrology; Sensor type determination module: determines the deployment location and quantity of sensors according to the relevant information of the forest area terrain, and determines the type of sensors to be used according to the characteristics of the forest area and the information to be monitored; Installation plan determination module: selects appropriate locations and plans for sensor deployment according to the determined sensor locations, quantity, and sensor types to be used; the sensor types include temperature and humidity sensors, fire risk index sensors, geological monitoring sensors, and image sensors; Temperature and humidity sensor module: The temperature and humidity sensor is used to monitor the climate and humidity changes in the forest area; the temperature and humidity sensor is deployed at a position 1.5-3m above the ground. The deployment rule of the temperature and humidity sensor is to set two temperature and humidity sensors for every 1250 square meters. Let the forest area be S, and the number of temperature and humidity sensors set be N. If S-1250≤0, then N=2; if S-1250>0, then ; Fire risk index sensor module: The fire risk index sensor is used to assess the fire risk in the forest area. The fire risk index sensor is deployed in the forest area close to the fire source, flammable materials and the edge of the forest area. The number of the fire risk index sensors is not less than 16; Geological monitoring sensor module: The geological monitoring sensor is used to detect soil moisture and soil stability. Twelve geological monitoring sensors are randomly deployed in the forest area and installed next to trees without direct contact with the trees. Image sensor module: The image sensor is used to monitor the growth status of vegetation and the situation of pests and diseases. The image sensor is installed in a representative location with a wide field of view, no obstacles, no direct sunlight, and the representative location includes forests of different types and woodlands of different ages. The number of image sensors installed is one per thousand square meters; Edge device installation module: After the sensor is deployed, an edge device is installed next to the sensor, and the edge device is connected to the sensor through a low-power wireless communication protocol, which includes LoRaWAN and NB-IoT; Real-time data transmission module: collects raw data through sensors and transmits the data collected by sensors to edge devices in real time.

5. According to claim 4, a smart forestry monitoring system based on edge computing is characterized in that: The monitoring information generation module includes: Data filtering module: The edge device detects the acquired raw data to remove outliers and removes noise data points through filtering algorithms; Data compression module: compresses the original data through a compression algorithm, normalizes and standardizes the data with different characteristics, and maps the normalized and standardized data to the same magnitude; Feature extraction module: extracting features from the data mapped to the same magnitude, reducing data dimensions and retaining key information; Subtask division module: divides the feature-extracted data into multiple subtasks, calculates the subtasks through parallel computing, and obtains pre-processed data after feature extraction; Feature analysis and processing module: analyzes and processes the pre-processed data after feature extraction through machine learning algorithms and pattern recognition algorithms to obtain corresponding information, including forest conditions, ecological environment, and vegetation generation conditions; Monitoring information generation module: generates monitoring information based on the results of data analysis and processing, and the monitoring information includes temperature and humidity ratio, fire risk level, vegetation coverage, soil moisture and stability, and pest and disease risk; Transmission encryption module: The edge device transmits the monitoring information to the cloud platform in real time through the Internet of Things, and uses an encryption protocol to encrypt the transmitted monitoring information; Status monitoring module: monitors the connection status between the edge device and the cloud platform by adopting a heartbeat mechanism.

6. According to claim 4, a smart forestry monitoring system based on edge computing is characterized in that: The monitoring management module includes: Cloud platform receiving module: The cloud platform receives the monitoring information transmitted by the edge device and creates a storage directory in the storage bucket according to the specific situation of the monitoring information, wherein the specific situation of the monitoring information includes the name of the monitoring point and the time period; Information storage corresponding module: after creating the storage directory, upload the received monitoring information to the corresponding storage directory; Machine learning module: Extract and select features of data through machine learning algorithms according to the characteristics and objectives of forestry monitoring; Management and monitoring model building module: Based on the results of feature extraction, the management and monitoring model is built through artificial intelligence algorithms combined with digital elevation models, and the established management and monitoring model is trained and verified through historical monitoring data; Model detection module: Use the trained management monitoring model to further analyze the newly collected monitoring data to obtain further analysis results, including trend analysis, anomaly detection and pattern recognition; Monitoring and management implementation module: monitoring and management are carried out according to the analysis results, including making decisions, optimizing forestry management strategies, generating reports and providing real-time monitoring data visualization.

Images