A method and system for creating a forestry information database
By preprocessing and optimizing multi-source forestry information data, extracting spatial and temporal features, performing space and time-based collaborative analysis and data sharding processing, the problem of insufficient data integration and analysis in the existing technology is solved, and the flexibility of forestry information database is realized and scalability.
Patent Information
- Application Number
- CN202411263101.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-10
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2044-09-10
AI Technical Summary
The existing technology lacks effective variation intensity analysis and data sharding processing mechanisms when processing large-scale dynamic data, and traditional databases lack space-time analysis capabilities when processing dynamic changes and time-series data, resulting in flexibility and scalability in forestry information database creation. Lower sex.
By obtaining multi-source forestry information data sets, data preprocessing and quality optimization are performed, spatial features and timing characteristics are extracted, multi-dimensional feature spatial matrix is constructed, space-time collaborative analysis is performed, data change intensity analysis and sharding processing is performed, and finally constructed on the cloud platform Shared database.
It realizes the standardization and optimization of data, improves the accuracy and completeness of data, supports refined forest management and spatiotemporal dynamic analysis, improves data processing efficiency and storage capabilities, and realizes cross-platform sharing and collaborative operation of data.
Smart Images

Figure CN118779303B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of database construction, and in particular to a method and system for establishing a forestry information database. Background Art
[0002] The earliest forestry data management systems usually used traditional paper records and simple spreadsheets. Although this method can record basic data, it is difficult to process large-scale data sets and is not conducive to data sharing and analysis. With the development of computer technology, database management systems (DBMS) began to be introduced into forestry data management, greatly improving the efficiency of data storage, retrieval and management. The rise of geographic information systems (GIS) made spatial analysis of forestry data possible. GIS technology can process data related to geographical location, such as forest cover and tree species distribution, and provide support for forestry decision-making through spatial analysis tools. During this period, forestry information databases began to integrate remote sensing data, meteorological data and ecological data, realizing the integration of multi-source data. With the development of big data technology and cloud computing, the functions of forestry information databases have been further expanded. Big data technology allows the processing of massive data, and cloud computing provides flexible storage and computing resources. Modern forestry information databases can not only store structured data, but also process unstructured data such as satellite images and drone data, realizing real-time data updates and analysis. However, existing technologies often lack effective change intensity analysis and data sharding processing mechanisms when processing large-scale dynamic data. At the same time, traditional databases often lack effective spatiotemporal analysis capabilities when processing dynamic changes and time series data, which leads to low flexibility and scalability in the creation of forestry information databases. Summary of the invention
[0003] Based on this, it is necessary to provide a method for creating a forestry information database to solve at least one of the above technical problems.
[0004] To achieve the above object, a method for creating a forestry information database comprises the following steps:
[0005] Step S1: obtaining a multi-source forestry information data set; performing data preprocessing on the multi-source forestry information data set to generate standard multi-source forestry information data; performing data quality optimization on the standard multi-source forestry information data to generate standard multi-source forestry optimized data;
[0006] Step S2: extracting spatial features from standard multi-source forestry optimization data to generate forestry spatial feature data; connecting forestry spatial feature data to forest units to generate topologically connected forest unit data; analyzing attribute features of forestry spatial grid unit data to generate forestry spatial unit attribute feature data; performing multi-attribute virtual unit feature fusion on forestry spatial unit attribute feature data through topologically connected forest unit data to generate a forestry spatial virtual forest unit set;
[0007] Step S3: extracting time dimension features from the forestry space virtual forest unit set to obtain forestry time series feature data; constructing a multidimensional feature space matrix for the forestry time series feature data and the forestry space virtual forest unit set to generate a multidimensional feature space matrix; performing spatiotemporal collaborative analysis on standard multi-source forestry optimization data based on the multidimensional feature space matrix to generate dynamic spatiotemporal forest data;
[0008] Step S4: Perform data change intensity analysis on the dynamic spatiotemporal forest data to generate dynamic spatiotemporal forest data change intensity data; perform data frequency conversion slicing on the dynamic spatiotemporal forest data based on the dynamic spatiotemporal forest data change intensity data to generate dynamic forest change data slicing; perform data clustering on the dynamic forest change data slicing to generate a forest distributed data storage cluster;
[0009] Step S5: Perform data signature identification on the forest distributed data storage cluster to generate forest distributed storage cluster label data; upload the forest distributed storage cluster label data to the cloud platform to build a shared database, generate a forestry information sharing database, and perform cross-platform operations on forestry information data.
[0010] The present invention standardizes data from different sources through preprocessing to ensure that the data format is consistent, which is convenient for subsequent integration and analysis. Optimize data quality, reduce errors and missing values, and improve the accuracy and completeness of data, thereby providing reliable basic data for decision-making. Standardized and optimized data are easy to process and analyze quickly, improving overall work efficiency. Through spatial feature extraction and forest unit connection, the spatial layout and characteristics of the forest are accurately depicted, supporting refined management. Attribute feature analysis and virtual unit feature fusion integrate multi-dimensional data to generate comprehensive forestry spatial feature data, providing a basis for complex data analysis. Through topological connection and attribute feature analysis, forest units can be better understood and classified, and the accuracy of classification and management can be improved. The extracted time series feature data can reveal the dynamics of forest changes over time and help identify trends and cyclical changes. The constructed multi-dimensional feature space matrix makes it possible to conduct a comprehensive analysis of spatiotemporal data, provides dynamic spatiotemporal forest data, and helps to better understand the spatiotemporal patterns of forest changes. Dynamic spatiotemporal forest data can support more accurate forest management and protection strategies and improve the scientificity and effectiveness of decision-making. Change intensity analysis can reveal the significance of forest changes and identify key change areas. Data frequency conversion sharding and cluster processing can effectively manage and store large-scale data, improve data processing speed and efficiency, and reduce storage pressure. Data sharding and clustering make fine-grained forest change analysis possible and provide more accurate forest distribution information. Data signature identification ensures the authenticity and integrity of the data and prevents data tampering and loss. The forestry information sharing database built on the cloud platform realizes cross-platform data sharing, improves data accessibility and the efficiency of cross-organizational collaboration. The shared database can provide unified data support for various forestry applications and promote more innovative application scenarios, such as forest monitoring and ecological assessment. Therefore, the present invention solves the problems of data integration, insufficient analysis and sharing in the creation of forestry information databases through standardized data processing, spatial feature extraction, spatiotemporal data analysis, data change processing and cloud platform sharing, and improves the flexibility and scalability of forestry information database creation.
[0011] Preferably, step S1 comprises the following steps:
[0012] Step S11: Acquire a multi-source forestry information dataset;
[0013] Step S12: performing data denoising on the multi-source forestry information data set to generate multi-source forestry information denoised data;
[0014] Step S13: performing data relationship analysis on the multi-source forestry information denoised data to generate multi-source forestry data association information; performing outlier correction on the multi-source forestry information denoised data based on the multi-source forestry data association information to generate multi-source forestry information data;
[0015] Step S14: filling missing values in the multi-source forestry information data to generate multi-source forestry information filling data; standardizing the multi-source forestry information filling data to generate standard multi-source forestry information data;
[0016] Step S15: Optimize the data quality of the standard multi-source forestry information data to generate standard multi-source forestry optimized data.
[0017] The present invention eliminates noise and outliers in the data through denoising and outlier correction, thereby ensuring the accuracy of forestry information data. This provides a reliable data basis for subsequent analysis. The missing value filling process effectively handles the missing items in the data, avoids analysis bias caused by incomplete data, and enhances the integrity of the data. Through the relationship analysis of multi-source forestry information data, the correlation between data is identified and utilized, the inherent connection and consistency of the data are improved, and the foundation is laid for complex analysis and modeling. Through the standardized processing of data, the scale differences between multi-source data are resolved, so that data from different sources can be uniformly processed and compared, and the availability of data is improved. The final data quality optimization step ensures that the generated standard multi-source forestry optimization data is of high quality, providing more reliable and accurate data support for applications in fields such as forestry management, decision support, and ecological research.
[0018] Preferably, step S15 comprises the following steps:
[0019] Step S151: Perform preliminary forestry feature screening on standard multi-source forestry information data to obtain preliminary forestry feature data;
[0020] Step S152: recursively eliminating duplicate features from the preliminary forestry feature data to generate forestry feature simplified data; performing feature importance evaluation on the forestry feature simplified data to generate forestry feature importance evaluation data;
[0021] Step S153: performing information gain feature selection on the forestry feature simplified data according to the forestry feature importance evaluation data to generate information gain feature data;
[0022] Step S154: Based on the data entropy method and the signal-to-noise ratio method, the data quality of the standard multi-source forestry information data is optimized according to the information gain characteristic data to generate standard multi-source forestry optimized data.
[0023] The present invention effectively reduces redundant features, retains the most representative forestry features, and improves the simplicity and processing efficiency of data through preliminary forestry feature screening and recursive repeated feature elimination. Importance assessment of streamlined forestry feature data helps to identify the most critical features for forestry information analysis, and provide more targeted information for subsequent data processing and decision-making. Based on feature importance assessment, the information gain feature selection method is adopted to further optimize feature data, so that the selected features can maximize the data interpretation ability and analysis effect. Combined with the data entropy method and the signal-to-noise ratio method, the information gain feature data is optimized for data quality, eliminating noise and invalid information in the data, and ensuring that the final generated standard multi-source forestry optimization data has higher quality. The final generated standard multi-source forestry optimization data not only has high quality and high reliability, but also has stronger explanatory and analytical value, which helps to improve the value of forestry information data in practical applications and provide more accurate data support for forestry management, ecological monitoring and other fields.
[0024] Preferably, step S2 comprises the following steps:
[0025] Step S21: extracting spatial features from standard multi-source forestry optimization data to generate forestry spatial feature data;
[0026] Step S22: dividing the forestry spatial feature data into spatial unit grids to generate forestry spatial grid unit data;
[0027] Step S23: performing forestry space unit topological relationship analysis on the forestry space grid unit data to generate forestry space unit topological relationship data; performing forest unit connection on the forestry space grid unit data according to the forestry space unit topological relationship data to generate topological connection forest unit data;
[0028] Step S24: Analyze the forestry space grid unit data for attribute characteristics to generate forestry space unit attribute characteristic data, wherein the attribute characteristic analysis includes ecological characteristic analysis, terrain characteristic analysis, and vegetation coverage characteristic analysis; perform multi-attribute virtual unit feature fusion on the forestry space unit attribute characteristic data by topologically linking the forest unit data to generate a forestry space virtual forest unit set.
[0029] The present invention extracts spatial features of standard multi-source forestry optimization data, and the generated forestry spatial feature data can accurately reflect the spatial distribution and structural characteristics of forest resources, providing a solid foundation for subsequent analysis and decision-making. By dividing the spatial unit grid of forestry spatial feature data, the generated forestry spatial grid unit data provides partition support for the refined management of forest resources, making forestry management and ecological monitoring more accurate. By analyzing the topological relationship of forestry spatial grid unit data, the generated forestry spatial unit topological relationship data can identify the spatial relationship between forest units, reveal the spatial structure and connectivity of forest resources, and provide deeper spatial information for forest management. By analyzing the multidimensional attributes of the ecological characteristics, terrain characteristics, and vegetation coverage characteristics of forestry spatial grid unit data, the generated attribute feature data can comprehensively characterize the multivariate attributes of forest resources and support more detailed forest resource evaluation and decision-making. By using topological connection forest unit data to fuse forestry spatial unit attribute feature data, the generated forestry spatial virtual forest unit set can not only reflect the spatial structure of forest resources, but also comprehensively reflect its multivariate attribute characteristics such as ecology, terrain and vegetation, providing virtualized integrated data for the comprehensive evaluation and management of forest resources. The generated forestry space virtual forest unit set provides comprehensive data support for forest resource management, ecological protection, forest health monitoring and other aspects, and improves the breadth and effectiveness of data in practical applications.
[0030] Preferably, step S3 comprises the following steps:
[0031] Step S31: extracting time dimension features from the forestry space virtual forest unit set to obtain forestry time series feature data;
[0032] Step S32: extracting environmental dimension features from forestry time series feature data using high-resolution remote sensing technology to obtain forestry environmental feature data;
[0033] Step S33: constructing a multidimensional feature space matrix according to the forestry time series feature data, the forestry environment feature data and the forestry space virtual forest unit set to generate a multidimensional feature space matrix;
[0034] Step S34: construct a forest dynamic monitoring model based on the multi-dimensional feature space matrix to generate a forest dynamic monitoring model; use the forest dynamic monitoring model to conduct spatiotemporal collaborative analysis of standard multi-source forestry optimization data to generate dynamic spatiotemporal forest data.
[0035] The present invention extracts the time dimension features of the forestry space virtual forest unit set, and the generated forestry time series feature data can capture the dynamic changes of forest resources, providing a key basis for subsequent time series analysis and prediction. The environmental dimension features of forestry time series feature data are extracted using high-resolution remote sensing technology, and the generated forestry environmental feature data can finely reflect the complexity and diversity of the forest environment, including climate, soil, water resources and other aspects, providing important data support for the comprehensive analysis of forest health and ecological environment. The constructed multidimensional feature space matrix integrates the feature data of multiple dimensions of time, environment and space to form a comprehensive and three-dimensional feature space. This matrix can fully express the complexity of forest resources and provide a solid data structure for in-depth analysis and modeling. Through the forest dynamic monitoring model constructed based on the multidimensional feature space matrix, accurate dynamic monitoring of forest resources can be carried out. This model can identify subtle trends and abnormal situations of forest changes, and provide real-time monitoring means for forest protection and management. The forest dynamic monitoring model is used to perform spatiotemporal collaborative analysis on standard multi-source forestry optimization data, and the generated dynamic spatiotemporal forest data can fully show the coordinated changes of forest resources in time and space. This in-depth analysis helps to understand the long-term evolution and future trends of forest resources, and provides a basis for formulating scientific forest management strategies. The generated dynamic spatiotemporal forest data provides data support for intelligent forest resource management, enabling real-time monitoring and intelligent decision-making of forest resources, and improving the scientificity and efficiency of forest management.
[0036] Preferably, step S34 includes the following steps:
[0037] Step S341: performing forest change analysis on the multi-dimensional feature space matrix to generate forest change analysis data;
[0038] Step S342: Divide the forest change analysis data into data sets to generate a model training set and a model test set; perform model training on the model training set using a graph neural network algorithm to generate a forest dynamic monitoring pre-model; perform model testing on the forest dynamic monitoring pre-model using the model test set to generate a forest dynamic monitoring model;
[0039] Step S343: importing the standard multi-source forestry optimization data into the forest dynamic monitoring model to predict forest dynamic changes and generate forest dynamic change prediction data;
[0040] Step S344: Perform spatiotemporal dynamic pattern analysis on standard multi-source forestry optimization data based on forest dynamic change prediction data to generate dynamic spatiotemporal forest data.
[0041] The present invention analyzes forest changes in a multidimensional feature space matrix, and the generated forest change analysis data can reveal the changing trends of forests at different time and space scales, providing rich change information and laying a solid foundation for predicting and monitoring forest dynamic changes. The model training set is trained using the graph neural network algorithm, and the generated forest dynamic monitoring pre-model can effectively capture the complex patterns in forest changes. Subsequently, the model is tested with the model test set to ensure the accuracy and reliability of the model, and the forest dynamic monitoring model finally generated has good prediction performance. By importing standard multi-source forestry optimization data into the forest dynamic monitoring model, the generated forest dynamic change prediction data can identify the future change trend of forest resources in advance, help prevent and respond to potential forest degradation or environmental changes, and provide forward-looking decision support for forest management. Based on the forest dynamic change prediction data, the spatiotemporal dynamic pattern analysis can reveal the laws of coordinated changes in forest resources in time and space, and the generated dynamic spatiotemporal forest data provides a comprehensive and detailed perspective for the overall management of forest resources, and improves the scientificity and systematicness of forest management. The application of the graph neural network algorithm enables the forest dynamic monitoring model to have a higher level of intelligence in capturing spatial relationships and dynamic changes. The continuous optimization and testing of the model ensures its stability and accuracy in processing complex forest data, thus providing reliable technical support for dynamic forest monitoring. The generated dynamic spatiotemporal forest data provides a powerful tool for real-time monitoring and management of forest resources, helping managers to detect anomalies and adjust strategies in a timely manner to ensure the sustainable development of forest resources.
[0042] Preferably, step S4 comprises the following steps:
[0043] Step S41: performing data change intensity analysis on the dynamic spatiotemporal forest data to generate dynamic spatiotemporal forest data change intensity data; performing data frequency conversion slicing on the dynamic spatiotemporal forest data based on the dynamic spatiotemporal forest data change intensity data to generate dynamic forest change data slicing;
[0044] Step S42: Distribute and store the dynamic forest change data slices to generate dynamic forest change storage data; perform data backup on the dynamic forest change storage data to generate dynamic forest change storage backup data;
[0045] Step S43: performing load balancing access request configuration according to the dynamic forest change storage data and the dynamic forest change storage backup data to generate load balancing access request configuration data; performing distributed index establishment on the dynamic forest change storage backup data to generate distributed index configuration data;
[0046] Step S44: clustering the dynamic forest change storage data using the load balancing access request configuration data and the distributed index configuration data to generate a forest distributed data storage cluster.
[0047] The present invention analyzes the data change intensity of dynamic spatiotemporal forest data, and the generated dynamic spatiotemporal forest data change intensity data can accurately reflect the frequency and intensity of forest resource changes, help identify the most significant changes in the area and time period, and provide a scientific basis for subsequent data processing. Based on the data change intensity, the data frequency conversion sharding can reasonably shard the dynamic forest change data to ensure the optimization and balance of data storage. The sharded data is stored through distributed storage technology, and the generated dynamic forest change storage data can make full use of storage resources to improve the efficiency and reliability of data storage. By backing up the dynamic forest change storage data, the generated dynamic forest change storage backup data provides a guarantee for data security. Even in the event of data damage or loss, the backup data can be quickly restored to ensure the integrity and availability of the forest change data. The load balancing access request configuration performed according to the dynamic forest change storage data and the backup data can optimize the efficiency of data access, prevent server overload or performance bottlenecks, and ensure smooth operation under high concurrency. Through the distributed index configuration, the generated distributed index configuration data can speed up the data retrieval speed, improve the system response capability, and make it possible to perform fast and accurate queries in massive forest data. The resulting forest distributed data storage cluster enhances the scalability and fault tolerance of the system by utilizing load balancing and distributed indexing technologies. The cluster can support the storage and processing of large-scale forest data, adapt to the growing data demand, and provide stable performance guarantees.
[0048] Preferably, step S41 includes the following steps:
[0049] Step S411: Obtain overall shard capacity data; perform data status screening on dynamic spatiotemporal forest data to obtain spatiotemporal forest dynamic change data and spatiotemporal forest static data;
[0050] Step S412: performing a change data difference analysis on the spatiotemporal forest dynamic change data to generate a spatiotemporal forest dynamic change data difference; performing a change spectrum analysis on the spatiotemporal forest change data based on the spatiotemporal forest dynamic change difference to generate spatiotemporal forest change spectrum data;
[0051] Step S413: quantify the dynamic change intensity through the difference of the spatiotemporal forest dynamic change data and the spatiotemporal forest change spectrum data to generate the spatiotemporal forest dynamic change intensity data; perform dynamic data slicing on the spatiotemporal forest dynamic change data through the spatiotemporal forest dynamic change intensity data to generate the spatiotemporal forest dynamic data slicing; perform static data slicing on the spatiotemporal forest static data to generate the spatiotemporal forest static data slicing;
[0052] Step S414: Optimize the shard space of the overall shard capacity data based on the spatiotemporal forest dynamic data shards and the spatiotemporal forest static data shards to generate dynamic forest change data shards.
[0053] The present invention accurately divides the data into spatiotemporal forest dynamic change data and spatiotemporal forest static data by state screening of dynamic spatiotemporal forest data, thereby ensuring the pertinence and processing efficiency of the data in the subsequent processing steps. By performing difference analysis and spectrum analysis on the spatiotemporal forest dynamic change data, the generated spatiotemporal forest dynamic change difference and spectrum data can reveal the subtle differences and frequency characteristics in the forest change process, and provide a solid data basis for the quantification of the dynamic change intensity. The spatiotemporal forest dynamic change difference and spectrum data are intensity quantified through step S413, and the generated spatiotemporal forest dynamic change intensity data can accurately reflect the intensity and impact range of forest changes. This quantification process enhances the understanding of forest changes and provides a basis for further data slicing operations. The dynamic data slicing based on the spatiotemporal forest dynamic change intensity data and the slicing processing of static data make the data slicing process more refined and rational. The generated spatiotemporal forest dynamic data slicing and static data slicing not only optimize the data storage structure, but also improve the efficiency and flexibility of data access. By optimizing the shard space for the overall shard capacity data, the dynamic forest change data shards generated can fully utilize storage resources, avoid space waste, and optimize the performance of the storage system. This optimization not only improves the efficiency of data storage, but also provides guarantees for the scalability and stability of the system.
[0054] Preferably, step S5 comprises the following steps:
[0055] Step S51: encrypting data in the forest distributed data storage cluster to generate forest distributed storage encrypted cluster data;
[0056] Step S52: digitally sign and identify the forest distributed storage encrypted cluster data to generate forest distributed storage cluster label data;
[0057] Step S53: upload the forest distributed storage cluster label data to the cloud platform to build a shared database, generate a forestry information shared database, and perform cross-platform operations on forestry information data.
[0058] The present invention encrypts the forest distributed data storage cluster, and the generated forest distributed storage encrypted cluster data can effectively prevent data leakage and unauthorized access, thereby ensuring the security of forestry information data during storage and transmission. This process improves the confidentiality of the data and ensures the security of sensitive forest data. By digitally signing the forest distributed storage encrypted cluster data, the generated forest distributed storage cluster label data can ensure the source traceability and integrity of the data. This digital signature can not only verify the authenticity of the data, but also provide a warning when the data is changed, which helps to maintain the integrity and consistency of the data. The forestry information change storage database is uploaded to the cloud platform through step S53, and the generated forestry information sharing database is constructed to support cross-platform operation of data. This not only improves the sharing and collaborative working capabilities of forestry information data, but also makes data access and operation between different platforms more convenient and efficient. The construction of the forestry information sharing database supports cross-platform operation operations, which means that the system can expand and integrate data from different sources in a larger range. This capability not only improves the flexibility and scalability of the system, but also provides a broader space for future forestry information management and monitoring. By encrypting, identifying and sharing the data on the cloud platform, step S5 makes data management more efficient and standardized. In particular, by building a shared database, unified management and operation of data on different platforms is achieved, providing a more integrated and efficient solution for forestry information management.
[0059] In this specification, a system for creating a forestry information database is provided, which is used to execute the above-mentioned method for creating a forestry information database. The system for creating a forestry information database includes:
[0060] A data quality optimization module is used to obtain a multi-source forestry information data set; perform data preprocessing on the multi-source forestry information data set to generate standard multi-source forestry information data; perform data quality optimization on the standard multi-source forestry information data to generate standard multi-source forestry optimized data;
[0061] The data space connection module is used to extract spatial features from standard multi-source forestry optimization data to generate forestry space feature data; to connect forestry space feature data to forest units to generate topological connection forest unit data; to analyze the attribute features of forestry space grid unit data to generate forestry space unit attribute feature data; to fuse multi-attribute virtual unit features of forestry space unit attribute feature data through topological connection forest unit data to generate forestry space virtual forest unit set;
[0062] The data-time coordination module is used to extract the time dimension features of the forestry space virtual forest unit set to obtain forestry time series feature data; construct a multidimensional feature space matrix for the forestry time series feature data and the forestry space virtual forest unit set to generate a multidimensional feature space matrix; conduct spatiotemporal coordination analysis on the standard multi-source forestry optimization data based on the multidimensional feature space matrix to generate dynamic spatiotemporal forest data;
[0063] The data storage sharding module is used to analyze the data change intensity of dynamic spatiotemporal forest data and generate dynamic spatiotemporal forest data change intensity data; perform data frequency conversion sharding on dynamic spatiotemporal forest data based on the dynamic spatiotemporal forest data change intensity data to generate dynamic forest change data shards; perform data clustering on dynamic forest change data shards to generate forest distributed data storage clusters;
[0064] The database construction module is used to sign and identify the forest distributed data storage cluster and generate forest distributed storage cluster label data; upload the forest distributed storage cluster label data to the cloud platform for shared database construction, generate a forestry information sharing database, and perform cross-platform operations on forestry information data.
[0065] The beneficial effect of the present invention is that by acquiring multi-source forestry information data sets and performing data preprocessing and quality optimization, the data format can be unified, the data quality can be improved, and the accuracy and reliability of subsequent analysis can be ensured. Through spatial feature extraction, forest unit connection and attribute feature analysis, the generated forestry space virtual forest unit set can comprehensively capture the spatial and attribute information of forestry data, provide detailed spatial feature descriptions, and support accurate forest management and decision-making. Temporal feature extraction and multi-dimensional feature space matrix construction of the forestry space virtual forest unit set can realize spatiotemporal collaborative analysis, and the generated dynamic spatiotemporal forest data helps to identify and analyze forest change trends and patterns, and improve dynamic management capabilities. Through data change intensity analysis, data frequency segmentation and data cluster processing, large-scale dynamic forest data can be effectively managed and stored, data processing efficiency can be improved, and more fine-grained forest change analysis can be supported. Data signature identification is performed on the forest distributed data storage cluster and uploaded to the cloud platform to build a shared database, realizing data security identification and cross-platform sharing, improving data accessibility and collaborative operation capabilities, and promoting information sharing and application. Therefore, the present invention solves the problems of data integration, insufficient analysis and sharing in the creation of forestry information database through standardized data processing, spatial feature extraction, spatiotemporal data analysis, data change processing and cloud platform sharing, and improves the flexibility and scalability of forestry information database creation. BRIEF DESCRIPTION OF THE DRAWINGS
[0066] Figure 1A schematic diagram of the steps of a method for creating a forestry information database;
[0067] Figure 2 for Figure 1 Detailed implementation steps of step S2 in the flowchart;
[0068] Figure 3 for Figure 1 Detailed implementation steps of step S3 in FIG.
[0069] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION
[0070] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.
[0071] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.
[0072] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.
[0073] To achieve this, please refer to Figures 1 to 3 , a method for creating a forestry information database, the method comprising the following steps:
[0074] Step S1: obtaining a multi-source forestry information data set; performing data preprocessing on the multi-source forestry information data set to generate standard multi-source forestry information data; performing data quality optimization on the standard multi-source forestry information data to generate standard multi-source forestry optimized data;
[0075] Step S2: extracting spatial features from standard multi-source forestry optimization data to generate forestry spatial feature data; connecting forestry spatial feature data to forest units to generate topologically connected forest unit data; analyzing attribute features of forestry spatial grid unit data to generate forestry spatial unit attribute feature data; performing multi-attribute virtual unit feature fusion on forestry spatial unit attribute feature data through topologically connected forest unit data to generate a forestry spatial virtual forest unit set;
[0076] Step S3: extracting time dimension features from the forestry space virtual forest unit set to obtain forestry time series feature data; constructing a multidimensional feature space matrix for the forestry time series feature data and the forestry space virtual forest unit set to generate a multidimensional feature space matrix; performing spatiotemporal collaborative analysis on standard multi-source forestry optimization data based on the multidimensional feature space matrix to generate dynamic spatiotemporal forest data;
[0077] Step S4: Perform data change intensity analysis on the dynamic spatiotemporal forest data to generate dynamic spatiotemporal forest data change intensity data; perform data frequency conversion slicing on the dynamic spatiotemporal forest data based on the dynamic spatiotemporal forest data change intensity data to generate dynamic forest change data slicing; perform data clustering on the dynamic forest change data slicing to generate a forest distributed data storage cluster;
[0078] Step S5: Perform data signature identification on the forest distributed data storage cluster to generate forest distributed storage cluster label data; upload the forest distributed storage cluster label data to the cloud platform to build a shared database, generate a forestry information sharing database, and perform cross-platform operations on forestry information data.
[0079] The present invention standardizes data from different sources through preprocessing to ensure that the data format is consistent, which is convenient for subsequent integration and analysis. Optimize data quality, reduce errors and missing values, and improve the accuracy and completeness of data, thereby providing reliable basic data for decision-making. Standardized and optimized data are easy to process and analyze quickly, improving overall work efficiency. Through spatial feature extraction and forest unit connection, the spatial layout and characteristics of the forest are accurately depicted, supporting refined management. Attribute feature analysis and virtual unit feature fusion integrate multi-dimensional data to generate comprehensive forestry spatial feature data, providing a basis for complex data analysis. Through topological connection and attribute feature analysis, forest units can be better understood and classified, and the accuracy of classification and management can be improved. The extracted time series feature data can reveal the dynamics of forest changes over time and help identify trends and cyclical changes. The constructed multi-dimensional feature space matrix makes it possible to conduct a comprehensive analysis of spatiotemporal data, provides dynamic spatiotemporal forest data, and helps to better understand the spatiotemporal patterns of forest changes. Dynamic spatiotemporal forest data can support more accurate forest management and protection strategies and improve the scientificity and effectiveness of decision-making. Change intensity analysis can reveal the significance of forest changes and identify key change areas. Data frequency conversion sharding and cluster processing can effectively manage and store large-scale data, improve data processing speed and efficiency, and reduce storage pressure. Data sharding and clustering make fine-grained forest change analysis possible and provide more accurate forest distribution information. Data signature identification ensures the authenticity and integrity of the data and prevents data tampering and loss. The forestry information sharing database built on the cloud platform realizes cross-platform data sharing, improves data accessibility and the efficiency of cross-organizational collaboration. The shared database can provide unified data support for various forestry applications and promote more innovative application scenarios, such as forest monitoring and ecological assessment. Therefore, the present invention solves the problems of data integration, insufficient analysis and sharing in the creation of forestry information databases through standardized data processing, spatial feature extraction, spatiotemporal data analysis, data change processing and cloud platform sharing, and improves the flexibility and scalability of forestry information database creation.
[0080] In the embodiment of the present invention, reference Figure 1 The above is a schematic diagram of the steps of a method for creating a forestry information database of the present invention. In this example, the method for creating a forestry information database includes the following steps:
[0081] Step S1: obtaining a multi-source forestry information data set; performing data preprocessing on the multi-source forestry information data set to generate standard multi-source forestry information data; performing data quality optimization on the standard multi-source forestry information data to generate standard multi-source forestry optimized data;
[0082] In the embodiment of the present invention, various data sources are determined, such as remote sensing images, ground survey data, meteorological data, forest resource survey data, etc. Remote sensing image data of forest areas are obtained through satellites or drones. Data such as trees, biodiversity, soil types, etc. are collected from field surveys. Meteorological related data such as temperature, precipitation, humidity, etc. are collected. Information affecting forests, such as land use data, policy data, etc., is obtained. Data obtained from different sources are sorted and summarized to form a preliminary multi-source forestry information data set. Noise and outliers in the data are identified and removed, and noise and outliers in the data are identified and removed. Data from different sources are aligned to the same coordinate system and time frame, and the data is converted into a unified format and structure for subsequent processing. Missing values in the data are interpolated or processed, and the data is standardized to conform to a unified scale and range. Through the above processing steps, a pre-processed standardized data set is generated to provide basic data for subsequent data quality optimization. The accuracy of the data is evaluated, potential errors are identified and corrected, the integrity of the data is checked, all necessary information is ensured to be included, and the consistency of the data is verified to ensure that the information in different data sources is consistent with each other. Combine data from different sources to enhance the comprehensiveness and accuracy of the data, select features that have a significant impact on the analysis, and remove redundant or irrelevant features. Smooth the data to reduce noise and fluctuations, compress the data to reduce storage space and processing time, and generate the final optimized data set through data quality optimization steps to improve data quality and availability.
[0083] Step S2: extracting spatial features from standard multi-source forestry optimization data to generate forestry spatial feature data; connecting forestry spatial feature data to forest units to generate topologically connected forest unit data; analyzing attribute features of forestry spatial grid unit data to generate forestry spatial unit attribute feature data; performing multi-attribute virtual unit feature fusion on forestry spatial unit attribute feature data through topologically connected forest unit data to generate a forestry spatial virtual forest unit set;
[0084] In an embodiment of the present invention, standard multi-source forestry optimization data is collected, including multiple data sources such as terrain, vegetation, and climate. Features are extracted using spatial analysis techniques, including but not limited to: terrain undulations, vegetation coverage, forest density, soil type, etc. Spatial data processing tools, such as GIS (geographic information system), are used to extract spatial features from multi-source data, generate forestry spatial feature data, and record the geographical and environmental features of each area or unit. The forestry spatial feature data is divided into forest units (such as different forest blocks or plots). Adjacent or related forest units are connected through topological analysis to generate the topological structure of the forest unit. Topological data processing methods (such as Voronoi diagrams and Delaunay triangulations) are used to define the spatial relationship between units, and topologically connected forest unit data are generated, including the connection relationship and interaction information between forest units. Attribute data of each forestry spatial grid unit is collected, such as tree species, tree age, growth rate, soil moisture, etc. Perform statistical analysis, feature extraction and data mining on the attribute data of grid units, identify important attribute features, use analytical tools (such as statistical software, machine learning models) to process and analyze these attribute data, generate forestry space unit attribute feature data, and record the attribute feature information of each spatial unit. Combine the forestry space unit attribute feature data with the topological connection forest unit data to perform feature fusion. Use data fusion techniques such as principal component analysis (PCA), multidimensional scaling analysis (MDS) or deep learning models to integrate attribute data from different sources. Generate virtual forest units through fusion results. These virtual units represent forest unit sets under different attributes and spatial relationships. Use virtual unit generation technology (such as virtual reality modeling, simulation software) to construct virtual forest unit sets and generate forestry space virtual forest unit sets for further analysis, simulation or decision support.
[0085] Step S3: extracting time dimension features from the forestry space virtual forest unit set to obtain forestry time series feature data; constructing a multidimensional feature space matrix for the forestry time series feature data and the forestry space virtual forest unit set to generate a multidimensional feature space matrix; performing spatiotemporal collaborative analysis on standard multi-source forestry optimization data based on the multidimensional feature space matrix to generate dynamic spatiotemporal forest data;
[0086] In an embodiment of the present invention, relevant data of a forestry space virtual forest unit set at different time points are collected, including growth status, climate conditions, seasonal changes, etc. Time series analysis methods are applied to extract time dimension features. The method may include: decomposing data into trends, seasonality, periodicity, and residuals, smoothing time series, eliminating short-term fluctuations, clustering time series, identifying different time patterns, and extracting forestry time series feature data, such as growth rate changes, interannual fluctuations, seasonal changes, etc. The forestry time series feature data and the data of the forestry space virtual forest unit set are integrated together to construct a feature matrix. Including the fusion of time dimension features and space dimension features. Select an appropriate multidimensional feature representation method, such as: mapping high-dimensional feature data to low-dimensional space, reducing the dimension and extracting the main features, decomposing high-dimensional data, extracting key features, and constructing a multidimensional feature space matrix, including feature data of time and space dimensions. Use a spatiotemporal analysis model to collaboratively analyze the multidimensional feature space matrix with standard multi-source forestry optimization data. Analytical methods can include: modeling the impact of temporal and spatial characteristics on forestry data, analyzing the impact of temporal and spatial interactions on forestry data, and simulating the dynamic behavior of forest systems under different temporal and spatial conditions. Generate dynamic spatiotemporal forest data, including forest data that changes in temporal and spatial dimensions. Apply data visualization tools (such as dynamic maps and time-series animations) to display the generated data to support dynamic analysis and decision-making.
[0087] Step S4: Perform data change intensity analysis on the dynamic spatiotemporal forest data to generate dynamic spatiotemporal forest data change intensity data; perform data frequency conversion slicing on the dynamic spatiotemporal forest data based on the dynamic spatiotemporal forest data change intensity data to generate dynamic forest change data slicing; perform data clustering on the dynamic forest change data slicing to generate a forest distributed data storage cluster;
[0088] In an embodiment of the present invention, by collecting time series records of dynamic spatiotemporal forest data, including forest cover changes, vegetation growth, climate change, etc. Calculate the data change intensity, that is, the degree of change of forest data in different time and space dimensions. The method may include: comparing data at different time points, identifying change areas and intensities, calculating data differences between adjacent time points or regions, measuring the rate and frequency of change, such as the change in forest cover each year, and generating change intensity data, including the magnitude, frequency and trend of change. Determine the strategy and standard for data sharding, such as based on time intervals, change intensity thresholds or spatial regions. According to the change intensity data, the dynamic spatiotemporal forest data is divided into different fragments (shards), which represent specific time periods, change intensity ranges or geographical regions, and the original data is segmented according to predefined sharding rules using data segmentation tools and algorithms to generate dynamic forest change data shards, and record the change characteristics and related attributes of each shard. Determine the strategy and standard for clustering, such as based on data similarity, change intensity, geographical location, etc. Use a data clustering algorithm to cluster the data shards. Common methods include: dividing data shards into K clusters, each cluster contains data shards with similar characteristics, clustering data shards with similar density based on density, generating a hierarchical structure of data shards for hierarchical analysis and storage. Creating a forest distributed data storage cluster, using a distributed database or data storage system (such as Hadoop, NoSQL database) to manage and store these clusters, and configuring a distributed storage system to ensure efficient data access and processing.
[0089] Step S5: Perform data signature identification on the forest distributed data storage cluster to generate forest distributed storage cluster label data; upload the forest distributed storage cluster label data to the cloud platform to build a shared database, generate a forestry information sharing database, and perform cross-platform operations on forestry information data.
[0090] In an embodiment of the present invention, a unique signature is generated for each data storage cluster. Common methods include: using a hash function (such as SHA-256) to generate a unique fingerprint of the data, and digitally signing the data to ensure the integrity and authenticity of the data. Generate forest distributed storage cluster label data based on the data signature. These labels should include: a unique identifier for each cluster, such as data creation time, data source, data type, etc., such as a hash value or a digital signature, for verifying the integrity of the data. Store the generated label data together with the actual data cluster to ensure that the data can be quickly found and verified when accessing the data. Arrange the forest distributed storage cluster label data into a format suitable for uploading, such as JSON, XML or CSV. Select a suitable cloud platform (such as AWS, Azure, Google Cloud) for data upload. Use the tools and services provided by the cloud platform (such as cloud storage services, data transfer tools) to upload label data and actual data clusters. Create a shared database on the cloud platform. The following technologies can be used: such as Amazon RDS, Azure SQL Database or Google Cloud SQL, such as Amazon S3, Azure Data Lake Storage or Google Cloud Storage, for storing and managing large-scale data sets. Configure database permissions and access control to ensure data security and privacy protection, and import label data into the database for management and query during the sharing process. Ensure that the shared database on the cloud platform can be integrated with other systems or platforms, including: providing RESTful or GraphQL APIs to allow different systems to access and operate the database, and setting up data interfaces or middleware to enable data exchange between different platforms. Implement cross-platform data operation tasks such as data query, update, analysis, and reporting, and use analytical tools on the cloud platform (such as Amazon Athena, Google BigQuery) for data processing and analysis. Provide access rights to all parties, allowing users to share and collaborate on data according to their needs, ensure the security of data sharing, and implement necessary encryption and authentication measures.
[0091] Preferably, step S1 comprises the following steps:
[0092] Step S11: Acquire a multi-source forestry information dataset;
[0093] Step S12: performing data denoising on the multi-source forestry information data set to generate multi-source forestry information denoised data;
[0094] Step S13: performing data relationship analysis on the multi-source forestry information denoised data to generate multi-source forestry data association information; performing outlier correction on the multi-source forestry information denoised data based on the multi-source forestry data association information to generate multi-source forestry information data;
[0095] Step S14: filling missing values in the multi-source forestry information data to generate multi-source forestry information filling data; standardizing the multi-source forestry information filling data to generate standard multi-source forestry information data;
[0096] Step S15: Optimize the data quality of the standard multi-source forestry information data to generate standard multi-source forestry optimized data.
[0097] In an embodiment of the present invention, a multi-source forestry information data set is formed by collecting raw data from different forestry information sources (such as satellite remote sensing data, ground sensor data, drone image data, etc.). The data set contains various types of forestry information, such as tree species distribution, forest health status, soil moisture, etc. The noise in the data set is processed by a denoising algorithm (such as Gaussian filtering, mean filtering, frequency domain filtering, etc.). The noise comes from sensor errors, environmental interference during data acquisition, etc. The result of denoising is a relatively pure data set, which is called multi-source forestry information denoised data. Data mining techniques (such as correlation analysis, cluster analysis, etc.) are used to perform relationship analysis on the denoised data to identify potential relationships and patterns between the data. These relationship information provides a basis for subsequent outlier correction. Then, based on these data relationships, the outliers in the denoised data are corrected, and the outliers are caused by data acquisition errors, abnormal environmental conditions, etc. The corrected data set is called multi-source forestry information data. Filling processing is performed on the missing data in the data set. The filling method may include interpolation, mean filling, prediction model filling, etc. The filled data set is called multi-source forestry information filled data. Subsequently, the filled data set is standardized so that data from different data sources can be compared and analyzed on the same scale. Standardization processing includes mean normalization, maximum and minimum value normalization, Z-score standardization, etc., and finally standard multi-source forestry information data is generated. In this step, the quality of the data is further improved, and the accuracy and completeness of the data are optimized through methods such as data cleaning, duplicate data deletion, and data consistency check. The optimized data set is called standard multi-source forestry optimized data, which has the characteristics of high quality and consistency and can be used for subsequent forestry information analysis and decision support.
[0098] Preferably, step S15 comprises the following steps:
[0099] Step S151: Perform preliminary forestry feature screening on standard multi-source forestry information data to obtain preliminary forestry feature data;
[0100] Step S152: recursively eliminating duplicate features from the preliminary forestry feature data to generate forestry feature simplified data; performing feature importance evaluation on the forestry feature simplified data to generate forestry feature importance evaluation data;
[0101] Step S153: performing information gain feature selection on the forestry feature simplified data according to the forestry feature importance evaluation data to generate information gain feature data;
[0102] Step S154: Based on the data entropy method and the signal-to-noise ratio method, the data quality of the standard multi-source forestry information data is optimized according to the information gain characteristic data to generate standard multi-source forestry optimized data.
[0103] In an embodiment of the present invention, the features in the standard multi-source forestry information data are preliminarily screened to exclude irrelevant or redundant features. The screening criteria can be based on domain knowledge, preliminary data analysis results, or use an automated feature selection method (such as a filtering method). The result is preliminary forestry feature data, which has potential relevance and importance to the analysis target. The preliminary forestry feature data is processed using a recursive feature elimination algorithm, and the least important features are gradually deleted until a feature subset is obtained. This process selects features through model performance evaluation (such as cross-validation) to ultimately generate forestry feature simplified data. The obtained forestry feature simplified data is subjected to feature importance evaluation. A variety of methods can be used for evaluation, such as feature importance scoring based on a tree model (such as a random forest, a gradient boosting machine) or other statistical indicators. The generated forestry feature importance evaluation data provides information on the contribution of each feature to the target variable. Based on the feature importance evaluation data, an information gain algorithm is applied to further select the most informative features. Information gain evaluates the contribution of features to the classification target, and selects features with the largest information gain to optimize the data set. This step generates information gain feature data, which contains the most useful features for the target analysis. Calculate the entropy value (Entropy) of the information gain feature data, and use the entropy value to analyze the degree of confusion and information content of the feature. A lower entropy value indicates that the feature has a higher amount of information about the target variable, which helps to optimize the quality of the data. Use entropy to optimize the standard multi-source forestry information data. Evaluate the signal-to-noise ratio (SNR) of the feature. A high SNR indicates that the feature contains more useful information in the data. By calculating the SNR of the feature, select features with a high SNR to optimize data quality. Combining the results of data entropy and SNR, perform the final data quality optimization on the standard multi-source forestry information data to generate standard multi-source forestry optimized data. The optimized data set will be more refined and can effectively support subsequent analysis and decision-making.
[0104] As an example of the present invention, refer to Figure 2As shown, in this example, step S2 includes:
[0105] Step S21: extracting spatial features from standard multi-source forestry optimization data to generate forestry spatial feature data;
[0106] Step S22: dividing the forestry spatial feature data into spatial unit grids to generate forestry spatial grid unit data;
[0107] Step S23: performing forestry space unit topological relationship analysis on the forestry space grid unit data to generate forestry space unit topological relationship data; performing forest unit connection on the forestry space grid unit data according to the forestry space unit topological relationship data to generate topological connection forest unit data;
[0108] Step S24: Analyze the forestry space grid unit data for attribute characteristics to generate forestry space unit attribute characteristic data, wherein the attribute characteristic analysis includes ecological characteristic analysis, terrain characteristic analysis, and vegetation coverage characteristic analysis; perform multi-attribute virtual unit feature fusion on the forestry space unit attribute characteristic data by topologically linking the forest unit data to generate a forestry space virtual forest unit set.
[0109] In an embodiment of the present invention, spatially related features such as geographic coordinates, spatial distribution patterns, spatial correlation, etc. are extracted from standard multi-source forestry optimization data. Spatial analysis techniques (such as spatial statistical analysis, spatial autocorrelation analysis, etc.) are used to identify and extract features. These spatial features may include the spatial shape, distribution density, spatial variability, etc. of the region to generate forestry spatial feature data. The extracted forestry spatial feature data is divided into uniform grid cells at a certain spatial resolution. Grid division can create spatial grids based on rasterization techniques (such as grid segmentation, spatial subdivision, etc.). These grid cells divide the entire study area into small, consistent and comparable areas. The generated forestry spatial grid cell data represents the spatial feature information in each grid cell. The topological relationship between spatial grid cells is analyzed, including adjacency, connectivity, interaction between regions, etc. Topological analysis methods (such as graph theory, spatial network analysis, etc.) are used to generate forestry spatial unit topological relationship data. This data describes the connection and relationship between each grid cell in space. Based on the topological relationship data, the spatial grid cells are connected to reflect the spatial distribution and structure of the actual forest unit. Connections can include spatially adjacent units, functionally related units, etc. Generate topologically connected forest unit data, which shows how the units in the forest area are connected and interact with each other. Evaluate the ecological attributes in the grid cells, such as species diversity, biomass, ecosystem service functions, etc., analyze terrain characteristics such as altitude, slope, aspect, etc., which affect the forestry environment in the area, evaluate characteristics such as vegetation coverage and vegetation type distribution, and understand the vegetation status and changes in the forest area. Combine the attribute feature data with the topologically connected forest unit data to perform feature fusion of multivariate attribute virtual units. Integrate the feature information of different attributes through data fusion techniques (such as principal component analysis, multidimensional scaling analysis, etc.) to generate a forestry space virtual forest unit set. This set of data represents a virtual forest unit that integrates spatial topology and attribute characteristics, providing a comprehensive understanding of the forest area.
[0110] As an example of the present invention, refer to Figure 3 As shown, in this example, step S3 includes:
[0111] Step S31: extracting time dimension features from the forestry space virtual forest unit set to obtain forestry time series feature data;
[0112] Step S32: extracting environmental dimension features from forestry time series feature data using high-resolution remote sensing technology to obtain forestry environmental feature data;
[0113] Step S33: constructing a multidimensional feature space matrix according to the forestry time series feature data, the forestry environment feature data and the forestry space virtual forest unit set to generate a multidimensional feature space matrix;
[0114] Step S34: construct a forest dynamic monitoring model based on the multi-dimensional feature space matrix to generate a forest dynamic monitoring model; use the forest dynamic monitoring model to conduct spatiotemporal collaborative analysis of standard multi-source forestry optimization data to generate dynamic spatiotemporal forest data.
[0115] In an embodiment of the present invention, time-related features are extracted from the time series data in the forestry space virtual forest unit set. Including but not limited to: such as vegetation growth cycle, seasonal changes, interannual changes, etc., analyzing the trends of forest coverage, tree species distribution changes in the time series, and identifying abnormal events in the time series (such as forest fires, pest outbreaks, etc.). These features will generate forestry time series feature data to describe the changes and dynamics of forests in the time dimension. High-resolution remote sensing technology is used to extract environmental-related features from forestry time series feature data. Remote sensing technology can provide detailed spatial and environmental information, including: spatiotemporal changes in environmental variables such as soil moisture, temperature, and precipitation, such as normalized vegetation index (NDVI), enhanced vegetation index (EVI), etc., which are used to evaluate vegetation health and coverage, and analyze the impact of environmental factors on forest time series changes. These data will generate forestry environmental feature data and provide detailed information on forest environmental changes. Integrate forestry time series feature data, forestry environmental feature data, and data from forestry space virtual forest unit sets to construct a multidimensional feature space matrix. The specific steps include: integrating the time dimension features, environmental dimension features and spatial features into a unified feature space. Create a multidimensional matrix in which each dimension corresponds to a feature category (time, environment, space, etc.). Data fusion techniques (such as principal component analysis, multidimensional scaling analysis, etc.) can be used to effectively integrate and express these features. Standardize the features of each dimension to ensure the uniformity and comparability of the matrix. The generated multidimensional feature space matrix will provide basic data for the subsequent dynamic monitoring model. Based on the multidimensional feature space matrix, a forest dynamic monitoring model is constructed. The model may include: time series prediction models (ARIMA, LSTM, etc.), spatial regression models, etc., which are used to predict and analyze the spatiotemporal dynamics of forests. Combined with environmental feature data, a model is constructed to analyze the impact of environmental factors on forest dynamics and detect abnormal forest events (such as abnormal changes, deterioration of health, etc.). The forest dynamic monitoring model is used to conduct spatiotemporal collaborative analysis of standard multi-source forestry optimization data to evaluate and predict the dynamic changes of forests. The analysis results will generate dynamic spatiotemporal forest data to describe the comprehensive changes of forests in time and space. These data can be used for applications such as forest management, ecological protection, and disaster warning.
[0116] Preferably, step S34 includes the following steps:
[0117] Step S341: performing forest change analysis on the multi-dimensional feature space matrix to generate forest change analysis data;
[0118] Step S342: Divide the forest change analysis data into data sets to generate a model training set and a model test set; perform model training on the model training set using a graph neural network algorithm to generate a forest dynamic monitoring pre-model; perform model testing on the forest dynamic monitoring pre-model using the model test set to generate a forest dynamic monitoring model;
[0119] Step S343: importing the standard multi-source forestry optimization data into the forest dynamic monitoring model to predict forest dynamic changes and generate forest dynamic change prediction data;
[0120] Step S344: Perform spatiotemporal dynamic pattern analysis on standard multi-source forestry optimization data based on forest dynamic change prediction data to generate dynamic spatiotemporal forest data.
[0121] In an embodiment of the present invention, a multidimensional feature space matrix is deeply analyzed to identify the change pattern of the forest. This includes: identifying the change trend of forest cover, vegetation type and biomass, detecting specific changes in the forest (such as tree growth, death, pests and diseases, etc.) by comparing data at different time points, analyzing the impact of environmental factors, climate change, etc. on forest changes, and the generated forest change analysis data will describe the changes in the forest in time and space, and provide basic information for subsequent modeling. The forest change analysis data is divided into a training set and a test set. Generally, the training set is used for model training, and the test set is used to evaluate the performance of the model. Random partitioning or cross-validation methods can be used to ensure the representativeness of the data and the generalization ability of the model. The training set is trained using a graph neural network (GNN) algorithm. GNN can process graph structured data and is suitable for processing nodes (such as forest units) and edges (such as adjacent relationships) in forest spatial networks, including feature embedding, message passing, graph convolution and other steps. The parameters of the model are optimized to minimize the prediction error, thereby generating a forest dynamic monitoring pre-model. The trained forest dynamic monitoring pre-model is tested using a test set. Evaluate the performance of the model (such as accuracy, recall rate, F1 score, etc.), adjust the model parameters or optimize the model structure according to the test results, and finally generate a forest dynamic monitoring model. Input the standard multi-source forestry optimization data into the constructed forest dynamic monitoring model to predict dynamic changes. Based on the model, conduct spatiotemporal dynamic analysis of the input data to predict future forest changes, including growth trends, coverage changes, environmental impacts, etc. The generated forest dynamic change prediction data will reflect the future dynamic situation and change trends of the forest. Based on the forest dynamic change prediction data, conduct a comprehensive analysis of the standard multi-source forestry optimization data. The analysis includes: identifying the spatiotemporal patterns of forest changes, such as seasonal changes, long-term trends, etc., analyzing the patterns of forest dynamic changes, such as growth areas, degradation areas, etc., and identifying abnormal change patterns in the prediction data (such as emergencies, abnormal growth, etc.). Integrate the results of spatiotemporal dynamic pattern analysis to generate comprehensive dynamic spatiotemporal forest data. These data describe the comprehensive changes of forests in time and space and can be used for further analysis and decision support.
[0122] Preferably, step S4 comprises the following steps:
[0123] Step S41: performing data change intensity analysis on the dynamic spatiotemporal forest data to generate dynamic spatiotemporal forest data change intensity data; performing data frequency conversion slicing on the dynamic spatiotemporal forest data based on the dynamic spatiotemporal forest data change intensity data to generate dynamic forest change data slicing;
[0124] Step S42: Distribute and store the dynamic forest change data slices to generate dynamic forest change storage data; perform data backup on the dynamic forest change storage data to generate dynamic forest change storage backup data;
[0125] Step S43: performing load balancing access request configuration according to the dynamic forest change storage data and the dynamic forest change storage backup data to generate load balancing access request configuration data; performing distributed index establishment on the dynamic forest change storage backup data to generate distributed index configuration data;
[0126] Step S44: clustering the dynamic forest change storage data using the load balancing access request configuration data and the distributed index configuration data to generate a forest distributed data storage cluster.
[0127] In an embodiment of the present invention, by analyzing the change intensity in the dynamic spatiotemporal forest data, for example, by calculating the change rate, change amplitude and other indicators of the data. Based on the change intensity calculation result, the dynamic spatiotemporal forest data change intensity data is generated to describe the change intensity of the data in time and space. According to the data change intensity data, the dynamic spatiotemporal forest data is subjected to variable frequency sharding. The specific steps include: designing a sharding strategy based on the change intensity, such as frequent sharding in high-change areas and less sharding in low-change areas. Dividing the dynamic spatiotemporal forest data into multiple data shards according to the strategy to form dynamic forest change data shards. These shards are convenient for storage, processing and analysis. Selecting a suitable distributed storage system (such as Hadoop HDFS, Apache Cassandra, Amazon S3, etc.) to store the dynamic forest change data shards. Distributedly storing the dynamic forest change data shards on multiple storage nodes to generate dynamic forest change storage data. This method improves the reliability and access efficiency of data storage. Designing a data backup strategy, such as regular backup, incremental backup, etc. Backing up the dynamic forest change storage data to generate dynamic forest change storage backup data to ensure the security and reliability of the data. Designing a load balancing strategy to distribute access requests to the dynamic forest change storage data. For example, load balancing is performed based on factors such as the geographic location of the request, the popularity of the data, and the access frequency. Generate load balancing access request configuration data and apply it to the load balancer (such as Nginx, HAProxy, etc.) to achieve balanced access to the data storage system. Design a distributed indexing strategy based on the characteristics of the data (such as data type, query mode, etc.). Establish a distributed index for the dynamic forest change storage backup data and generate distributed index configuration data. This index will optimize data query and retrieval efficiency. Design a data cluster architecture based on the load balancing access request configuration data and the distributed index configuration data. Integrate the dynamic forest change storage data, load balancing configuration, and distributed index configuration into a data cluster. The cluster can include multiple storage nodes, load balancers, and index services to provide efficient data storage and access services. Configure cluster management tools (such as Kubernetes, Apache Mesos, etc.) to monitor, manage, and maintain the cluster to ensure the stability and performance of the cluster.
[0128] Preferably, step S41 includes the following steps:
[0129] Step S411: Obtain overall shard capacity data; perform data status screening on dynamic spatiotemporal forest data to obtain spatiotemporal forest dynamic change data and spatiotemporal forest static data;
[0130] Step S412: performing a change data difference analysis on the spatiotemporal forest dynamic change data to generate a spatiotemporal forest dynamic change data difference; performing a change spectrum analysis on the spatiotemporal forest change data based on the spatiotemporal forest dynamic change difference to generate spatiotemporal forest change spectrum data;
[0131] Step S413: quantify the dynamic change intensity through the difference of the spatiotemporal forest dynamic change data and the spatiotemporal forest change spectrum data to generate the spatiotemporal forest dynamic change intensity data; perform dynamic data slicing on the spatiotemporal forest dynamic change data through the spatiotemporal forest dynamic change intensity data to generate the spatiotemporal forest dynamic data slicing; perform static data slicing on the spatiotemporal forest static data to generate the spatiotemporal forest static data slicing;
[0132] Step S414: Optimize the shard space of the overall shard capacity data based on the spatiotemporal forest dynamic data shards and the spatiotemporal forest static data shards to generate dynamic forest change data shards.
[0133] In an embodiment of the present invention, the overall storage requirements are determined, including the size of the data and the capacity of each shard. The data can be obtained according to the historical change records of the data, the storage demand prediction, and the like. The spatiotemporal forest dynamic change data are screened out from the dynamic spatiotemporal forest data, including the temporal change part, such as the vegetation coverage change, the tree growth, and the like. The spatiotemporal forest static data are screened out from the dynamic spatiotemporal forest data, including the spatial static part, such as the topography of the forest area, the original vegetation coverage, and the like. The spatiotemporal forest dynamic change data are subjected to difference calculation to analyze the changes of the data between different time points. This includes: calculating the amount of change between consecutive time points, calculating the amount of change between different spatial positions, forming the difference of the spatiotemporal forest dynamic change data, and describing the change amplitude and trend of the data. Based on the difference of the spatiotemporal forest dynamic change, Fourier transform or other spectrum analysis methods are used to extract the spectrum characteristics of the data, form the spatiotemporal forest change spectrum data, and describe the frequency components and their intensity of the changes. The intensity of the dynamic change is quantified by combining the difference of the spatiotemporal forest dynamic change data and the change spectrum data. This can be done by: calculating the statistics of the intensity of change, such as average intensity, maximum intensity, etc., mapping the intensity data to the data sharding strategy, describing the intensity of dynamic changes as the basis for sharding. Sharding the spatiotemporal forest dynamic change data based on the spatiotemporal forest dynamic change intensity data, dividing the data into different sharding units for easy storage and processing, and sharding the spatiotemporal forest static data according to the characteristics of static data. Based on the overall sharding capacity data, spatial optimization is performed on the spatiotemporal forest dynamic data sharding and static data sharding. Including: adjusting the size and number of shards to meet the overall capacity requirements and storage efficiency, optimizing the layout of shards in the storage system to improve data access speed and storage resource utilization, forming the final dynamic forest change data sharding, and ensuring the best data storage and processing effects.
[0134] Preferably, step S5 comprises the following steps:
[0135] Step S51: encrypting data in the forest distributed data storage cluster to generate forest distributed storage encrypted cluster data;
[0136] Step S52: digitally sign and identify the forest distributed storage encrypted cluster data to generate forest distributed storage cluster label data;
[0137] Step S53: upload the forest distributed storage cluster label data to the cloud platform to build a shared database, generate a forestry information shared database, and perform cross-platform operations on forestry information data.
[0138] In an embodiment of the present invention, by selecting a suitable data encryption algorithm, such as the Advanced Encryption Standard (AES), RSA or other public key encryption algorithms. All data in the forest distributed data storage cluster is encrypted. This includes: generating and managing encryption keys to ensure secure storage and access to the keys. Encrypting the data using the selected encryption algorithm to generate forest distributed storage encrypted cluster data. The encrypted data will prevent unauthorized access and data leakage. Generate a private key and a public key pair for digital signature. Digitally sign the forest distributed storage encrypted cluster data. The specific steps include: using a hash algorithm (such as SHA-256) to generate a hash value for the encrypted data. Signing the hash value with a private key to generate a digital signature. The digital signature is used together with the encrypted data to generate forest distributed storage cluster tag data. This tag data is used to verify the integrity and source of the data. Select a suitable cloud platform service provider (such as AWS, Azure, Google Cloud) for database upload and sharing. Ensure that the data format of the forestry information change storage database is compatible with the cloud platform. Upload the forestry information change storage database to the cloud platform. Data transfer tools or APIs can be used for uploading. Create a shared database instance on the cloud platform and configure database access permissions. Set up data sharing strategies to ensure data availability and consistency during cross-platform operations. Perform cross-platform data operations, including reading, writing, and updating data, through the configured shared database, and use the collaborative function of the cloud platform to support multiple users or systems to collaborate and analyze forestry information data.
[0139] In this specification, a system for creating a forestry information database is provided, which is used to execute the above-mentioned method for creating a forestry information database. The system for creating a forestry information database includes:
[0140] A data quality optimization module is used to obtain a multi-source forestry information data set; perform data preprocessing on the multi-source forestry information data set to generate standard multi-source forestry information data; perform data quality optimization on the standard multi-source forestry information data to generate standard multi-source forestry optimized data;
[0141] The data space connection module is used to extract spatial features from standard multi-source forestry optimization data to generate forestry space feature data; to connect forestry space feature data to forest units to generate topological connection forest unit data; to analyze the attribute features of forestry space grid unit data to generate forestry space unit attribute feature data; to fuse multi-attribute virtual unit features of forestry space unit attribute feature data through topological connection forest unit data to generate forestry space virtual forest unit set;
[0142] The data-time coordination module is used to extract the time dimension features of the forestry space virtual forest unit set to obtain forestry time series feature data; construct a multidimensional feature space matrix for the forestry time series feature data and the forestry space virtual forest unit set to generate a multidimensional feature space matrix; conduct spatiotemporal coordination analysis on the standard multi-source forestry optimization data based on the multidimensional feature space matrix to generate dynamic spatiotemporal forest data;
[0143] The data storage sharding module is used to analyze the data change intensity of dynamic spatiotemporal forest data and generate dynamic spatiotemporal forest data change intensity data; perform data frequency conversion sharding on dynamic spatiotemporal forest data based on the dynamic spatiotemporal forest data change intensity data to generate dynamic forest change data shards; perform data clustering on dynamic forest change data shards to generate forest distributed data storage clusters;
[0144] The database construction module is used to sign and identify the forest distributed data storage cluster and generate forest distributed storage cluster label data; upload the forest distributed storage cluster label data to the cloud platform for shared database construction, generate a forestry information sharing database, and perform cross-platform operations on forestry information data.
[0145] The beneficial effect of the present invention is that by acquiring multi-source forestry information data sets and performing data preprocessing and quality optimization, the data format can be unified, the data quality can be improved, and the accuracy and reliability of subsequent analysis can be ensured. Through spatial feature extraction, forest unit connection and attribute feature analysis, the generated forestry space virtual forest unit set can comprehensively capture the spatial and attribute information of forestry data, provide detailed spatial feature descriptions, and support accurate forest management and decision-making. Temporal feature extraction and multi-dimensional feature space matrix construction of the forestry space virtual forest unit set can realize spatiotemporal collaborative analysis, and the generated dynamic spatiotemporal forest data helps to identify and analyze forest change trends and patterns, and improve dynamic management capabilities. Through data change intensity analysis, data frequency segmentation and data cluster processing, large-scale dynamic forest data can be effectively managed and stored, data processing efficiency can be improved, and more fine-grained forest change analysis can be supported. Data signature identification is performed on the forest distributed data storage cluster and uploaded to the cloud platform to build a shared database, realizing data security identification and cross-platform sharing, improving data accessibility and collaborative operation capabilities, and promoting information sharing and application. Therefore, the present invention solves the problems of data integration, insufficient analysis and sharing in the creation of forestry information database through standardized data processing, spatial feature extraction, spatiotemporal data analysis, data change processing and cloud platform sharing, and improves the flexibility and scalability of forestry information database creation.
[0146] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.
[0147] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.
Claims
1. A method for creating a forestry information database, characterized in that: The following steps are involved: Step S1: obtaining a multi-source forestry information data set; performing data preprocessing on the multi-source forestry information data set to generate standard multi-source forestry information data; Optimize the data quality of standard multi-source forestry information data and generate standard multi-source forestry optimized data; Step S2: extracting spatial features from standard multi-source forestry optimization data to generate forestry spatial feature data; connecting forest units from forestry spatial feature data to generate topologically connected forest unit data; and performing attribute feature analysis on forestry spatial grid unit data to generate forestry spatial unit attribute feature data; Through topological connection of forest unit data, multi-attribute virtual unit feature fusion is performed on forestry space unit attribute feature data to generate forestry space virtual forest unit set; Step S3: extracting time dimension features from the forestry space virtual forest unit set to obtain forestry time series feature data; constructing a multidimensional feature space matrix from the forestry time series feature data and the forestry space virtual forest unit set to generate a multidimensional feature space matrix; Based on the multidimensional feature space matrix, the standard multi-source forestry optimization data is subjected to spatiotemporal collaborative analysis to generate dynamic spatiotemporal forest data; step S3 includes the following steps: Step S31: extracting time dimension features from the forestry space virtual forest unit set to obtain forestry time series feature data; Step S32: extracting environmental dimension features from forestry time series feature data using high-resolution remote sensing technology to obtain forestry environmental feature data; Step S33: constructing a multidimensional feature space matrix according to the forestry time series feature data, the forestry environment feature data and the forestry space virtual forest unit set to generate a multidimensional feature space matrix; Step S34: constructing a forest dynamic monitoring model based on a multidimensional feature space matrix to generate a forest dynamic monitoring model; using the forest dynamic monitoring model to perform spatiotemporal collaborative analysis on standard multi-source forestry optimization data to generate dynamic spatiotemporal forest data; Step S34 includes the following steps: Step S341: performing forest change analysis on the multi-dimensional feature space matrix to generate forest change analysis data; Step S342: Divide the forest change analysis data into data sets to generate a model training set and a model test set; perform model training on the model training set using a graph neural network algorithm to generate a forest dynamic monitoring pre-model; perform model testing on the forest dynamic monitoring pre-model using the model test set to generate a forest dynamic monitoring model; Step S343: importing the standard multi-source forestry optimization data into the forest dynamic monitoring model to predict forest dynamic changes and generate forest dynamic change prediction data; Step S344: performing spatiotemporal dynamic pattern analysis on standard multi-source forestry optimization data according to forest dynamic change prediction data to generate dynamic spatiotemporal forest data; Step S4: Perform data change intensity analysis on the dynamic spatiotemporal forest data to generate dynamic spatiotemporal forest data change intensity data; perform data frequency conversion slicing on the dynamic spatiotemporal forest data based on the dynamic spatiotemporal forest data change intensity data to generate dynamic forest change data slicing; perform data clustering on the dynamic forest change data slicing to generate a forest distributed data storage cluster; Step S4 includes the following steps: Step S41: performing data change intensity analysis on the dynamic spatiotemporal forest data to generate dynamic spatiotemporal forest data change intensity data; performing data frequency conversion slicing on the dynamic spatiotemporal forest data based on the dynamic spatiotemporal forest data change intensity data to generate dynamic forest change data slicing; Step S42: Distribute and store the dynamic forest change data slices to generate dynamic forest change storage data; perform data backup on the dynamic forest change storage data to generate dynamic forest change storage backup data; Step S43: performing load balancing access request configuration according to the dynamic forest change storage data and the dynamic forest change storage backup data to generate load balancing access request configuration data; performing distributed index establishment on the dynamic forest change storage backup data to generate distributed index configuration data; Step S44: clustering the dynamic forest change storage data using the load balancing access request configuration data and the distributed index configuration data to generate a forest distributed data storage cluster; Step S5: Perform data signature identification on the forest distributed data storage cluster to generate forest distributed storage cluster label data; upload the forest distributed storage cluster label data to the cloud platform to build a shared database, generate a forestry information sharing database, and perform cross-platform operations on forestry information data.
2. The method for creating a forestry information database according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: Acquire a multi-source forestry information dataset; Step S12: performing data denoising on the multi-source forestry information data set to generate multi-source forestry information denoised data; Step S13: performing data relationship analysis on the multi-source forestry information denoised data to generate multi-source forestry data association information; performing outlier correction on the multi-source forestry information denoised data based on the multi-source forestry data association information to generate multi-source forestry information data; Step S14: filling missing values in the multi-source forestry information data to generate multi-source forestry information filling data; standardizing the multi-source forestry information filling data to generate standard multi-source forestry information data; Step S15: Optimize the data quality of the standard multi-source forestry information data to generate standard multi-source forestry optimized data.
3. The method for creating a forestry information database according to claim 2, characterized in that: Step S15 includes the following steps: Step S151: Perform preliminary forestry feature screening on standard multi-source forestry information data to obtain preliminary forestry feature data; Step S152: recursively eliminating duplicate features from the preliminary forestry feature data to generate forestry feature simplified data; performing feature importance evaluation on the forestry feature simplified data to generate forestry feature importance evaluation data; Step S153: performing information gain feature selection on the forestry feature simplified data according to the forestry feature importance evaluation data to generate information gain feature data; Step S154: Based on the data entropy method and the signal-to-noise ratio method, the data quality of the standard multi-source forestry information data is optimized according to the information gain characteristic data to generate standard multi-source forestry optimized data.
4. The method for creating a forestry information database according to claim 1, characterized in that: Step S2 includes the following steps: Step S21: extracting spatial features from standard multi-source forestry optimization data to generate forestry spatial feature data; Step S22: dividing the forestry spatial feature data into spatial unit grids to generate forestry spatial grid unit data; Step S23: performing forestry space unit topological relationship analysis on the forestry space grid unit data to generate forestry space unit topological relationship data; performing forest unit connection on the forestry space grid unit data according to the forestry space unit topological relationship data to generate topological connection forest unit data; Step S24: Analyze the forestry space grid unit data for attribute characteristics to generate forestry space unit attribute characteristic data, wherein the attribute characteristic analysis includes ecological characteristic analysis, terrain characteristic analysis, and vegetation coverage characteristic analysis; perform multi-attribute virtual unit feature fusion on the forestry space unit attribute characteristic data by topologically linking the forest unit data to generate a forestry space virtual forest unit set.
5. The method for creating a forestry information database according to claim 1, characterized in that: Step S41 includes the following steps: Step S411: Obtain overall shard capacity data; perform data status screening on dynamic spatiotemporal forest data to obtain spatiotemporal forest dynamic change data and spatiotemporal forest static data; Step S412: performing a change data difference analysis on the spatiotemporal forest dynamic change data to generate a spatiotemporal forest dynamic change data difference; performing a change spectrum analysis on the spatiotemporal forest dynamic change data based on the spatiotemporal forest dynamic change data difference to generate spatiotemporal forest change spectrum data; Step S413: quantify the dynamic change intensity through the difference of the spatiotemporal forest dynamic change data and the spatiotemporal forest change spectrum data to generate the spatiotemporal forest dynamic change intensity data; perform dynamic data slicing on the spatiotemporal forest dynamic change data through the spatiotemporal forest dynamic change intensity data to generate the spatiotemporal forest dynamic data slicing; perform static data slicing on the spatiotemporal forest static data to generate the spatiotemporal forest static data slicing; Step S414: Optimize the shard space of the overall shard capacity data based on the spatiotemporal forest dynamic data shards and the spatiotemporal forest static data shards to generate dynamic forest change data shards.
6. The method for creating a forestry information database according to claim 1, characterized in that: Step S5 includes the following steps: Step S51: encrypting data in the forest distributed data storage cluster to generate forest distributed storage encrypted cluster data; Step S52: digitally sign and identify the forest distributed storage encrypted cluster data to generate forest distributed storage cluster label data; Step S53: upload the forest distributed storage cluster label data to the cloud platform to build a shared database, generate a forestry information shared database, and perform cross-platform operations on forestry information data.
7. A system for creating a forestry information database, characterized in that: Used to execute the method for creating a forestry information database as claimed in claim 1, the system for creating a forestry information database comprises: A data quality optimization module is used to obtain a multi-source forestry information data set; perform data preprocessing on the multi-source forestry information data set to generate standard multi-source forestry information data; perform data quality optimization on the standard multi-source forestry information data to generate standard multi-source forestry optimized data; The data space connection module is used to extract spatial features from standard multi-source forestry optimization data to generate forestry space feature data; to connect forestry space feature data to forest units to generate topological connection forest unit data; to analyze the attribute features of forestry space grid unit data to generate forestry space unit attribute feature data; to fuse multi-attribute virtual unit features of forestry space unit attribute feature data through topological connection forest unit data to generate forestry space virtual forest unit set; The data-time coordination module is used to extract the time dimension features of the forestry space virtual forest unit set to obtain forestry time series feature data; construct a multidimensional feature space matrix for the forestry time series feature data and the forestry space virtual forest unit set to generate a multidimensional feature space matrix; conduct spatiotemporal coordination analysis on the standard multi-source forestry optimization data based on the multidimensional feature space matrix to generate dynamic spatiotemporal forest data; The data storage sharding module is used to perform data change intensity analysis on dynamic spatiotemporal forest data and generate dynamic spatiotemporal forest data change intensity data; perform data frequency conversion sharding on dynamic spatiotemporal forest data based on the dynamic spatiotemporal forest data change intensity data and generate dynamic forest change data shards; perform data clustering on dynamic forest change data shards and generate forest distributed data storage clusters; The database construction module is used to sign and identify the forest distributed data storage cluster and generate forest distributed storage cluster label data; upload the forest distributed storage cluster label data to the cloud platform for shared database construction, generate a forestry information sharing database, and perform cross-platform operations on forestry information data.
Citation Information
Patent Citations
Micro-expression detection method based on fusion of time-space domain information and random forest
CN114429655A
Forest resource analysis method and system based on forestry ecological big data
CN118350554A
