Space-time cube based on dynamic combination weight and construction method thereof

Abstract

The invention relates to the field of geospatial data processing, in particular to a space-time cube based on dynamic combination weight and a construction method thereof. The method comprises the following steps: acquiring ecological index data of a human-ground system region, and constructing a three-dimensional space-time cube by the ecological index data according to a spatial dimension, a time dimension and an ecological index dimension; the three-dimensional space-time cube is composed of a plurality of cube elements, and each cube element is a geographic grid carrying a spatial position, a time sequence and an ecological index of a dynamic combination weight; and defining an operator subset of the three-dimensional space-time cube to obtain the space-time cube based on the dynamic combination weight. In this way, the problems that in a traditional method, data integration is insufficient, temporal-spatial resolution is not matched, and dynamic weight is lost can be solved, dynamic changes of an ecological system can be accurately captured, and a scientific basis is provided for ecological environment management and ecological restoration.

Description

Technical Field

[0001] The present invention generally relates to the field of geospatial data processing, and more specifically, to a spatio-temporal cube based on dynamic combined weights and a construction method thereof. Background Art

[0002] With the rapid development of globalization and urbanization, the human-earth system in the transitional period presents extremely comprehensive and complex characteristics and has entered a complex transitional period. The human-earth system is a complex system in which human activities interact with the natural environment, and its changes have characteristics such as long time series, spatial heterogeneity, and periodic changes. Driven by multiple factors such as rapid urbanization, industrialization, and climate change, its dynamic changes are more complex, and many limitations of traditional research methods have been exposed.

[0003] Traditional methods rely on single data sources or static models, and have problems such as insufficient data integration, mismatched spatio-temporal resolution, and lack of dynamic weights, making it difficult to comprehensively capture the change laws of the human-earth system. At the same time, problems such as the long time series, spatial heterogeneity, and periodic changes of the human-earth system in the prior art have not been effectively solved. Summary of the Invention

[0004] According to the present invention, there is provided a spatio-temporal cube based on dynamic combined weights and a construction scheme thereof. This scheme can effectively cope with the characteristics of the human-earth system, and its advantages lie in multi-source data integration, dynamic weight adjustment, and refined analysis. It can be widely applied to fields such as urbanization, ecological environment monitoring, and regional sustainable development, providing a scientific basis for policymakers and providing a new technical means for the refined research of the human-earth system.

[0005] In a first aspect of the present invention, there is provided a construction method of a spatio-temporal cube based on dynamic combined weights. The method includes:

[0006] Obtaining ecological index data of a human-earth system region, and constructing a three-dimensional spatio-temporal cube according to the spatial dimension, time dimension, and ecological index dimension; the three-dimensional spatio-temporal cube is composed of a plurality of cube elements, and each cube element is a geographical grid carrying ecological indices of spatial positions, time series, and dynamic combined weights.

[0007] Defining an operation operator set of the three-dimensional spatio-temporal cube to obtain a spatio-temporal cube based on dynamic combined weights.

[0008] In a second aspect of the present invention, there is provided a spatio-temporal cube based on dynamic combined weights. The spatio-temporal cube includes: The spatio-temporal cube is constructed by the construction method of the spatio-temporal cube based on dynamic combined weights described above.

[0009] In a third aspect of the present invention, an electronic device is provided. The electronic device includes at least one processor; and a memory communicatively connected to the at least one processor; the memory stores instructions executable by the at least one processor, and when the instructions are executed by the at least one processor, the at least one processor is enabled to execute the method of the first aspect of the present invention.

[0010] It should be understood that the content described in the summary of the invention is not intended to limit the key or important features of the embodiments of the present invention, nor to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. Brief Description of the Drawings

[0011] In combination with the drawings and with reference to the following detailed description, the above and other features, advantages and aspects of the embodiments of the present invention will become more apparent. In the drawings, the same or similar reference numerals denote the same or similar elements, where:

[0012] Figure 1 A flowchart showing a method for constructing a spatio-temporal cube based on dynamic combined weights according to an embodiment of the present invention is shown;

[0013] Figure 2 A schematic diagram showing the construction of an ecological index spatio-temporal cube according to an embodiment of the present invention is shown;

[0014] Figure 3 A schematic diagram showing the operation of an operation operator set according to an embodiment of the present invention is shown;

[0015] Figure 4 A block diagram showing an exemplary electronic device capable of implementing the embodiments of the present invention is shown;

[0016] Wherein, GM is a two-dimensional index grid module; index1 is index 1; index2 is index 2; SC is a spatio-temporal cube; IC is an index cube; p-index1 is index 1 at different times; p-index2 is index 2 at different times; pl-index1 is vegetation index 1; pl-index2 is vegetation index 2; w-index1 is water body index 1; w-index2 is water body index 2; x is the longitude spatial axis; y is the latitude spatial axis; SL is cutting; AG is aggregation; RO is rotation; PR is projection; 400 is an electronic device, 401 is a computing unit, 402 is a ROM, 403 is a RAM, 404 is a bus, 405 is an I / O interface, 406 is an input unit, 407 is an output unit, 408 is a storage unit, 409 is a communication unit. Detailed Description of the Embodiments

[0017] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0018] In addition, the term "and / or" in this article is merely a description of the association relationship of associated objects, indicating that three relationships can exist. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this article generally represents an "or" relationship between the associated objects before and after.

[0019] In the present invention, by constructing a spatio-temporal cube and defining an operation operator set, a spatio-temporal cube based on dynamic combined weights is obtained. This method solves problems such as insufficient data integration, mismatched spatio-temporal resolution, and lack of dynamic weights in traditional methods, and can accurately capture the dynamic changes of desert ecosystems, providing a scientific basis for desertification control and ecological restoration.

[0020] Figure 1 The flowchart of the construction method of the spatio-temporal cube based on dynamic combined weights according to an embodiment of the present invention is shown.

[0021] The method includes:

[0022] S101. Obtain the ecological index data of the human-earth system region, and construct a three-dimensional spatio-temporal cube with the ecological index data according to the spatial dimension, time dimension, and ecological index dimension; the three-dimensional spatio-temporal cube is composed of a number of cube elements, and each cube element is a geographical grid carrying the ecological index with spatial position, time series, and dynamic combined weights.

[0023] Specifically, as Figure 2 shown, divide the spatial dimension (X-axis longitude, Y-axis latitude) according to the two-dimensional grid index module (GM) containing ecological indexes by geographical grid, and then arrange them in chronological order to form a spatio-temporal cube (SC) and an index cube (LC); among them, the two-dimensional grid index module (GM) contains indexes (index1, index2...), the spatio-temporal cube (SC) contains indexes in different periods (p-index1, p-index2...), and the index cube (LC) contains different types of vegetation indexes (p1-index) and water body indexes w-index.

[0024] In this embodiment, the ecological index with dynamic combined weights includes:

[0025]

[0026] Among them, EC(p(x, y)) is the ecological index of the dynamic combined weight; p(x, y) is the pixel at the spatial position (x, y); i is the index value of the number of years; n is the number of years within the time range. is the weight after dynamic adjustment; z i is the i-th ecological index; t is the time point.

[0027] The weight after dynamic adjustment includes:

[0028]

[0029] Among them, is the original weight of the ecological index z i ; f(r ti , r si ) is the spatio-temporal resolution adjustment function; g(z i ) is the adjustment function related to the ecological index, which is used to further adjust the weight according to the characteristics of the ecological index.

[0030] Among them, the spatio-temporal resolution adjustment function f(r ti , r si ) includes:

[0031]

[0032] Among them, r ti is the time resolution; r si is the spatial resolution.

[0033] By constructing a three-dimensional spatio-temporal cube of the ecological index, multi-source data can be effectively integrated, so as to accurately capture ecological changes, providing comprehensive and accurate basis and support for ecological research, governance and sustainable development. This cube can not only reflect the complexity of the ecosystem, including long-term time series, spatial heterogeneity and periodic changes, but also timely detect changes in ecological elements through dynamic monitoring and accurate analysis, providing accurate data support for ecological environment governance and restoration. In addition, the three-dimensional spatio-temporal cube provides a new data model and analysis basis for ecological research, supports researchers to deeply study the relationships and evolution mechanisms of various elements of the ecosystem, and at the same time provides accurate information for policymakers to help formulate scientific and reasonable ecological protection and sustainable development policies.

[0034] As some alternative implementation manners of this embodiment, the construction steps of the three-dimensional spatio-temporal cube are as follows:

[0035] 1. Selection of the human-earth system area: Select a representative desert area - Kubuqi Desert, the seventh largest desert in China, located in the northern part of the Ordos Plateau, Inner Mongolia Autonomous Region, adjacent to the south bank of the "Ji" character bend of the Yellow River; covering an area of about 18,600 square kilometers, with complex terrain, mainly featuring crescent dunes and chain dunes. The dunes in the west are particularly tall and steep, while the eastern region is relatively flat, with saline-alkali land and remains of dried-up lake beds distributed. The vegetation in this area is sparse, mainly consisting of sand-loving plants such as Salix psammophila and Agriophyllum squarrosum. This area can fully reflect the complexity and diversity of the desert ecosystem. Since the implementation of ecological restoration projects such as "diverting the Yellow River to control sand" and "photovoltaic sand control", the spatial heterogeneity, periodic changes, and complexity of human intervention in the Kubuqi Desert have made it an ideal area for studying the desert ecosystem.

[0036] 2. Construction of the spatial dimension: Divide the human-earth system area according to a certain geographical coordinate grid to determine the spatial resolution. For example, according to the size of the human-earth system area and the requirements of data accuracy, the space can be divided into grid cells of 30m×30m and 10m×10m. Each grid cell serves as the basic unit of the spatio-temporal cube in the spatial dimension, storing the ecological index data at that location.

[0037] 3. Construction of the time dimension: Arrange the remote sensing data obtained in different periods and the calculated ecological indices in chronological order (from 2013 to 2024, from July to August each year). Taking one year as the time interval, store the ecological index data of multiple years on the time axis in sequence. Ensure the continuity and integrity of the time series for analyzing the temporal change trend of the ecological index.

[0038] 4. Ecological index dimension: Operate and set the spatio-temporal cube according to the set ecological index parameters, and each grid cell stores the calculated value of the ecological index.

[0039] Specifically, the steps to obtain the ecological index data of the human-earth system area from July to August each year from 2013 to 2024 are as follows:

[0040] 1. Obtain remote sensing image data: Collect images from multiple satellite sensors, including but not limited to the Landsat series, Sentinel series, etc. The above satellite data have different spatial resolutions, temporal resolutions, and spectral bands, which can provide complementary information. For example, Landsat 8 data provides medium-resolution (30m) remote sensing images with a long time series, suitable for analyzing the long-term trend of ecological changes; Sentinel-2 data has a high spatial resolution, providing high-spatial-resolution (10m) remote sensing images, which can capture more detailed ecological details and are suitable for calculating fine ecological indices.

[0041] 2. Data processing, including:

[0042] (1) Image correction: Radiometric and geometric corrections are performed on remote sensing images to ensure data accuracy and consistency. Radiometric correction is used to eliminate the influence of factors such as the sensor itself and the atmosphere on the radiance values of the images, while geometric correction makes the images match the actual geographical coordinates.

[0043] Specifically, radiometric and geometric corrections are performed on Sentinel-2 and Landsat 8 images to ensure data accuracy and consistency.

[0044] (2) Cloud removal: Cloud removal algorithms are used to remove cloud interference in the images. For example, for Landsat 8 data, a cloud removal method based on image spectral features and time series analysis is used; for Sentinel-2 data, the built-in Sen2Cor algorithm is used for cloud masking.

[0045] 3. Selection and calculation of ecological indices, including:

[0046] (1) RVI (Ratio Vegetation Index): RVI is usually used to monitor vegetation cover. In desertified areas, due to sparse vegetation, the reflectance difference between the NIR and RED bands is significant. Therefore, RVI can reflect changes in vegetation cover and play a role in early warning of desertification.

[0047]

[0048] Among them, NIR is the reflectance of the near-infrared band, indicating the health and growth status of vegetation, usually with a higher reflectance when vegetation is lush; RED is the reflectance of the red light band, indicating the ability of vegetation to absorb red light. Vegetation conducts photosynthesis by absorbing red light, so the red light reflectance is usually higher in areas with poor or sparse vegetation.

[0049] (2) NDVI (Normalized Difference Vegetation Index): NDVI is one of the most commonly used vegetation indices in the field of remote sensing, used to measure the health status and coverage of vegetation. NDVI is a commonly used indicator for monitoring vegetation health and coverage. In desertified areas, due to sparse or degraded vegetation, the NDVI value will decrease significantly. Therefore, NDVI is often used to identify desert expansion, land degradation, and the effects of ecological restoration.

[0050]

[0051] (3) SAVI (Soil Adjusted Vegetation Index): SAVI performs better in areas with high soil exposure (such as the edges of deserts or semi-arid regions). SAVI can more accurately reflect the vegetation cover by adjusting the influence of the soil background, and is especially suitable for desertified areas with large soil exposure.

[0052]

[0053] Among them, L is the correction factor, usually with a value of 0.5.

[0054] (4) EVI (Enhanced Vegetation Index): EVI is an improved vegetation index, specifically designed to overcome the saturation problem of NDVI in areas with high vegetation cover and provide more sensitive monitoring of vegetation changes. EVI corrects the influence of the atmosphere on the red light band by introducing the blue light band and reduces the influence of soil reflection by increasing the soil background correction factor. In desertified areas, EVI can be used for dynamic monitoring of vegetation changes and correction of atmospheric effects, and is especially suitable for detecting the effects of vegetation restoration projects.

[0055]

[0056] Among them, BLUE is the reflectance of the blue light band; G is the gain factor, usually 2.5; C1 is the first atmospheric correction coefficient, usually 6; C2 is the second atmospheric correction coefficient, usually 7.5; K is the soil correction coefficient, usually 1.

[0057] (5) GDVI (General Difference Vegetation Index): GDVI is one of the important parameters reflecting plant growth and is also one of the important parameters for classifying plant density. GDVI can better reflect the growth trend and coverage of vegetation. For semi-arid regions, the vegetation growth is poor and the vegetation coverage is low. The General Difference Vegetation Index (GDVI) has higher sensitivity and dynamic perception range in areas with low vegetation cover compared to the Normalized Difference Vegetation Index (NDVI). Therefore, in this embodiment, the General Difference Vegetation Index (GDVI) is used to represent the greenness index of the study area.

[0058]

[0059] Among them, ρ NIR is the reflectance of the near-infrared band; ρ R is the reflectance of the red light band.

[0060] (6) NDWI (Normalized Difference Water Index): NDWI is mainly used to monitor surface water bodies. In desert areas, water bodies are scarce and unevenly distributed. NDWI can effectively distinguish water bodies from other surface features, especially in applications such as desert oasis and water resource management, such as water source monitoring and identification of groundwater recharge areas.

[0061]

[0062] Among them, GREEN is the reflectance of the green light band.

[0063] (7) MNDWI (Modified Normalized Difference Water Index): MNDWI makes the detection of water bodies more accurate in urbanized and human activity-intensive areas by using the short-wave infrared band to replace the green light band in NDWI. In desertification monitoring, MNDWI can better suppress the interference of non-water features such as buildings and bare land on water body identification, and help monitor water resource changes in desert areas, such as the dynamic changes of the Yellow River and lakes.

[0064]

[0065] Among them, SWIR is the reflectance of the short-wave infrared band.

[0066] In this embodiment, the database used to store data can be:

[0067] 1. The relational database PostgreSQL, which is used to store spatio-temporal coordinates and structured ecological index data, including ecological index data such as time and spatial coordinate dimension information.

[0068] 2. The non-relational database MongoDB, which is used to store the ecological index calculation process and raw image metadata, and stores semi-structured or unstructured data in the BSON (Binary JSON) format, such as detailed calculation process records of ecological indexes and raw image metadata obtained by different sensors.

[0069] 3. The distributed storage architecture, which stores data in slices according to time or space, and adopts redundant backup and object storage technologies to store multiple copies of important data redundantly. Multiple copies of the same data are saved on different storage nodes; at the same time, combined with object storage technology, the ecological index data is stored in the distributed object storage system in the form of objects. Each object has a unique identifier, and data uploading, downloading and management are carried out through the object storage interface.

[0070] Specifically, index queries and indicator extensions can be established for the above three databases, including: establishing a composite index structure, using index algorithms such as B-trees and R-trees to support time-space joint index queries, and quickly locating ecological index data within a specified time range and spatial area; the database design adopts a flexible data architecture. When new ecological indicators are incorporated into the monitoring system, only new fields need to be added to the corresponding data tables and new index relationships need to be established to achieve dynamic expansion of the ecological indicator dimension.

[0071] S102. Define an operator set for the three-dimensional spatio-temporal cube to obtain a spatio-temporal cube based on dynamic combined weights.

[0072] In this embodiment, the operator set includes:

[0073] A slicing operator, used to extract a sub-dataset within a specified spatio-temporal range. Rotation RO: ROTATION Aggregation AG: AGGREGATION Projection PR: PROJECTION Slicing SL: SLICE

[0074] An aggregation operator, used to perform aggregation statistics on ecological indices in the time and space dimensions.

[0075] A projection operator, which reduces the three-dimensional data to two-dimensional data by dimension reduction according to the time-ecological index dimension or the space-ecological index dimension, generating a two-dimensional dataset.

[0076] A rotation operator, used to perform coordinate axis conversion through a rotation matrix.

[0077] Specifically, as Figure 3 shown, through the slicing (SL, SLICE) operation, the cross-section of the three-dimensional spatio-temporal cube is sliced; through the aggregation (AG, AGGREGATION) operation, the ecological indices of the three-dimensional spatio-temporal cube are aggregated in the time and space dimensions; through the rotation (RO, ROTATION) operation, the time and space axes of the three-dimensional spatio-temporal cube are exchanged to achieve the conversion of the observation perspective; through the projection (PR, PROJECTION) operation, the three-dimensional spatio-temporal cube is reduced to a two-dimensional dataset. Wherein, in the three-dimensional cube, x is the longitude spatial axis; y is the latitude spatial axis.

[0078] By improving the flexibility of data extraction and analysis, the cutting operator can accurately extract sub-datasets within a specified time and space range. Time slicing, space slicing and dicing operations support the analysis of ecological indexes in different years, local areas and specific time and space areas, respectively, making the research more targeted and flexible. Aggregation statistics and feature mining use aggregation operators to perform statistics on ecological indexes in time and space dimensions, revealing the trend of ecosystem changes and regional differences, and providing data support for rational resource allocation and ecological management. Data simplification and key information highlighting use projection operators to reduce the dimension of three-dimensional data to generate two-dimensional data sets, simplify the data structure and highlight the relationship between key ecological factors, and improve analysis efficiency. Multi-perspective analysis and dynamic relationship insight use rotation operators to transform coordinate axes, analyze the dynamic changes of ecological indexes in time and space from different angles, and explore the complex connections and potential laws within the ecosystem. The multi-dimensional information generated by the set of operation operators provides a comprehensive and accurate basis for ecological research, environmental governance and sustainable development decision-making, supports scientific researchers to conduct in-depth research on the evolution mechanism of ecosystems, helps policy makers evaluate the effects of ecological engineering, predict trends and formulate scientific and reasonable ecological protection and restoration policies, and achieves precise control and sustainable development of ecosystems.

[0079] In this embodiment, the cutting operator includes time slicing, space slicing and dicing. Specifically, the cutting operator only retains the ecological index calculation value within a specified time range, and extracts the pixel ecological index value of the corresponding time period from the time series.

[0080] As some optional implementations of this embodiment, in the remote sensing data analysis of the Kubuqi Desert, the cutting operator is used to extract ecological index data within a specified time and space range. NDVI, NDWI, SAVI and GDVI data from July to August 2013 to 2024 are selected, and the spatial range is [x1, x2, y1, y2] divided into 30m×30m grid cells.

[0081] The time slice is used to extract ecological index data within a specified time range; the time slice includes:

[0082]

[0083] As some optional implementations of this embodiment, the time slice is sliced ​​according to the time dimension, and the vegetation index and water index data of 2013 and 2024 are extracted to analyze the state of the desert ecosystem in that year. By comparing the slice data of different years, the changes of the ecosystem over time can be understood. Specifically, the NDVI data from July 2013 to August 2013 are extracted to generate a spatial distribution map of vegetation coverage in this time period. For example, the NDVI data from July 2013 to August 2013 are extracted, and the results are:

[0084] Table 1

[0085] Spatial position (x, y) Time (t) NDVI (x1, y1) 2013-07 0.45 (x1, y1) 2013-08 0.47 (x2, y2) 2013-07 0.50 (x2, y2) 2013-08 0.52

[0086] The spatial slice is used to extract ecological index data within a specified spatial range, where the spatial range is [s x1 ,s x2 ,s y1 ,s y2 ]; the spatial slice comprises:

[0087]

[0088] As some optional implementation methods of this embodiment, the spatial slicing is slicing in the spatial dimension, selecting ecological index data of a local area in the northwest of the Kubuqi Desert to study the ecological characteristics and changing patterns of the local area, which is helpful to formulate targeted ecological protection and restoration strategies.

[0089] The cutting is used to extract ecological index data of data blocks within a specified time and space range; the cutting includes:

[0090]

[0091] Among them, EC ts (p(x,y)) is the time slice; The spatial position is (x, y), the ecological index is z, and the starting time is t start The pixel value of The spatial position is (x, y), the ecological index is z, and the starting time is t end The pixel value of is the weight after dynamic adjustment; EC ss (p(x,y)) is a spatial slice; i is the index value of the year; n is the number of years in the time range; z i is the i-th ecological index; t is the time point; s x1 is the starting coordinate of the spatial range on the x-axis in the longitude direction; s x2 is the ending coordinate of the spatial range on the x-axis in the longitude direction; s y1 is the starting coordinate of the spatial range on the y-axis in the latitude direction; s y2 is the starting coordinate of the spatial range on the y-axis in the latitude direction; x is the longitude spatial axis; y is the latitude spatial axis; EC block (p(x,y)) is the cut block.

[0092] Specifically, for pixels within the selected spatial range, the ecological index value is calculated according to the above formula, while for pixels not within the range, the ecological index value is set to 0 (indicating that these pixels are not considered).

[0093] The slicing operator extracts sub-datasets within a specified spatio-temporal range, enabling precise acquisition of target data from a three-dimensional spatio-temporal cube to meet high-precision requirements. Time slices can be used to compare the ecosystem states in different periods, revealing evolution trends and periodic patterns; spatial slices focus on the ecological characteristics of local areas, providing a basis for formulating ecological strategies; the slicing operation combines spatio-temporal dimensions to meet complex spatio-temporal analysis, making the research more targeted and flexible. In addition, the slicing operator can also reduce the processing of redundant data, improve research efficiency, support multi-scale research, provide accurate ecological information for policymakers, enhance decision-making support capabilities, strongly promote ecological research, monitoring and governance work, and help to deeply understand the laws of ecosystem changes and formulate reasonable protection and restoration strategies.

[0094] In this embodiment, the aggregation operator includes time aggregation and spatial aggregation.

[0095] The time aggregation is used to calculate the statistics of the ecological index data in a specific area within a specified time range.

[0096] The spatial aggregation is used to calculate the statistics of the ecological index data at a specific time point within a specified spatial range.

[0097] In this embodiment, the statistics of the ecological index data in a specific area within a specified time range include the time aggregation maximum value, the time aggregation minimum value, and the time aggregation standard deviation.

[0098] As some alternative implementation manners of this embodiment, the time aggregation is performed along the time dimension to calculate the time aggregation maximum value, the time aggregation minimum value, and the time aggregation standard deviation of the ecological index of Kubuqi Desert from 2013 to 2024.

[0099] Specifically, the time range is expressed as [t start , t end , and the specific area is expressed as W (defined by the spatial range).

[0100] The time aggregation maximum value includes:

[0101]

[0102] The time aggregation minimum value includes:

[0103] EC min,t = min{EC(p(x, y)) (z,t) |t start ≤ t ≤ t end and (x, y) ∈ W}

[0104] The time aggregation standard deviation includes:

[0105]

[0106] Among them, EC max,t is the maximum value of time aggregation; p(x, y) is the pixel at the spatial position (x, y); i is the index value of the number of years; n is the number of years within the time range, n = t end - t start + 1; is the weight after dynamic adjustment; z i is the i-th ecological index; t is the time point; t start is the start time; t end is the end time; W is the rotation matrix, defined by the spatial range and containing multiple spatial positions (x, y); EC min,t is the minimum value of time aggregation; σ EC,t is the standard deviation of time aggregation; is; m is the number of pixels within the specific area W.

[0107] Time aggregation aggregates and statistically analyzes ecological indices in the time dimension. By calculating statistics such as the maximum value, minimum value, and standard deviation, it reveals the ecological change trend. For example, the change in the statistical value of the multi-year vegetation index can reflect the vegetation growth status and influencing factors, providing a quantitative basis for studying the long-term evolution of the ecosystem; the standard deviation of time aggregation can evaluate the stability of the ecosystem. Taking the water body index as an example, the change in the standard deviation can predict the situation of the ecosystem being affected by external disturbances, facilitating timely protection; combining historical data and models can assist in ecological environment prediction, predicting changes in vegetation coverage, and contributing to preventive management; it can also provide ecological information for policymakers, evaluate the implementation effect of ecological policies, and support the formulation of more scientific ecological protection policies; at the same time, the aggregated statistics in the unified time dimension facilitate cross-regional ecological comparison. By comparing data from different desert regions, analyzing the underlying factors, it promotes cross-regional ecological cooperation and coordinated development.

[0108] In this embodiment, the statistics of the ecological index data at a specific time point within the specified spatial range include the spatial aggregation mean, spatial aggregation maximum value, spatial aggregation minimum value, and spatial aggregation standard deviation.

[0109] The spatial aggregation mean includes:

[0110]

[0111] The spatial aggregation maximum value includes:

[0112]

[0113] The spatial aggregation minimum value includes:

[0114]

[0115] The spatial aggregation standard deviation includes:

[0116]

[0117] where is the spatial aggregation mean; p(x,y) is the pixel at the spatial position (x,y); i is the index value of the number of years; n is the number of years within the time range; is the weight after dynamic adjustment; z i is the i-th ecological index; t is the time point; W is the rotation matrix; σ EC,s is the spatial aggregation standard deviation, an indicator measuring the degree of dispersion of the ecological index at the first specific time point t0 within the specified spatial range W; m is the number of pixels within the region W; t0 is the first specific time point; EC max,s is the spatial aggregation maximum value; EC min,s is the spatial aggregation minimum value; t3 is the second specific time point, the moment used to calculate the spatial aggregation minimum value.

[0118] In this embodiment, the spatial aggregation is performed along the spatial dimension. For example, calculating the statistical values of the ecological index of Kubuqi Desert from 2013 to 2024 includes the spatial aggregation mean, spatial aggregation maximum value, spatial aggregation minimum value, spatial aggregation standard deviation, etc.

[0119] The spatial aggregation performs aggregation statistics on the ecological index in the spatial dimension. By calculating statistics such as the spatial aggregation maximum value, spatial aggregation minimum value, and spatial aggregation standard deviation, it reveals the ecological change trend. For example, the change in the statistical value of the multi-year vegetation index can reflect the vegetation growth status and influencing factors, providing a quantitative basis for studying the long-term evolution of the ecosystem; the spatial aggregation standard deviation can evaluate the stability of the ecosystem. Taking the water body index as an example, the change in the standard deviation can predict the situation of the ecosystem being disturbed by the outside world, facilitating timely protection; combining historical data and models can assist in ecological environment prediction, predicting the change in vegetation coverage, and helping with preventive management; it can provide ecological information for policymakers, evaluate the implementation effect of ecological policies, and support the formulation of more scientific ecological protection policies; at the same time, the aggregation statistics with a unified time dimension facilitate cross-regional ecological comparison. By comparing the data of different desert regions and analyzing the underlying factors, it promotes cross-regional ecological cooperation and coordinated development.

[0120] In this embodiment, the projection operator includes: assuming the ecological index data set is E, if projected onto the retained dimension D

[0121] As some alternative embodiments of this example, in the remote sensing data analysis of the Kubuqi Desert, the projection operator is used to reduce the dimensionality of three-dimensional data to generate a two-dimensional data set, simplify the data structure, and highlight the relationships of key ecological factors. Suppose we select the NDVI, NDWI, SAVI, and GDVI data from July to August from 2013 to 2024, with the spatial range of [x1, x2, y1, y2], divided into grid cells of 30m × 30m.

[0122] Specifically, for each element in the projected data set E p the calculation formula for the ecological index value is as follows:

[0123] (1) Retaining the spatial dimension and the ecological index dimension (e.g., NDVI values at different positions within a certain area), for the ecological index value EC p (x, y) at the point (x, y) is:

[0124]

[0125] where t′ is a comprehensive consideration value at a certain fixed time point or within a certain time range; w l is a weight factor associated with a specific ecological index; EC p (x, y) is the ecological index value, which is a two-dimensional data set.

[0126] As some alternative embodiments of this example, calculate the average NDVI value at each spatial position (x1, y1)(x2, y2), and the result is:

[0127] Table 2

[0128]

[0131] (2) Retaining the time dimension and the ecological index dimension (e.g., the change of a certain ecological index over time), for the ecological index value EC p (t) at the time point t is:

[0132]

[0133] As some alternative embodiments of this example, calculate the average NDVI value at each time point t, and the result is:

[0134] Table 3

[0135]

[0136] Retaining the spatial dimension and the time dimension, (e.g., the ecological index values at different positions within a certain area at different time points), the ecological index value EC p is:

[0137]

[0138] Among them, EC p (x, y, t) is the ecological index value at the spatial position (x, y) and time point t; w l is the weight factor associated with a specific ecological index z (assuming w l = 1 to simplify the calculation); p(x, y) (z,t) is the value of the ecological index z at the spatial position (x, y) and time point t; l is the index of the ecological index, used to distinguish different ecological indices.

[0139] As some alternative implementation manners of this embodiment, the NDVI values at each spatial position (x, y) at each time point t are extracted, and the result is:

[0140] Table 4

[0141]

[0142] In this embodiment, the projection operator, for example according to the research focus, mainly focuses on the relationship between vegetation and water bodies, only retains the dimensions related to the vegetation index and the water body index, projects a three-dimensional data subset, simplifies the data structure while highlighting the interaction relationship between key ecological factors, and then generates a two-dimensional data set.

[0143] Through these projection operations, the data structure can be simplified and the relationship between key ecological factors can be highlighted, the analysis efficiency can be improved, multi-dimensional research can be supported, and strong support can be provided for ecological research, monitoring and governance work.

[0144] In this embodiment, the rotation operator realizes the conversion of the observation perspective by exchanging the time and space axes. Specifically, the coordinate axis conversion is performed through a rotation matrix, where the rotation refers to rotating by an angle θ around the X axis, including:

[0145] The rotation matrix is:

[0146]

[0147] The coordinate axis conversion includes:

[0148]

[0149] Among them, R is the rotation matrix; x is the longitude spatial axis; y is the latitude spatial axis; z is the time axis; x' is the converted longitude spatial axis; y' is the converted latitude spatial axis; z' is the converted time axis.

[0150] Specifically, the rotation operation is to rearrange the coordinate axes of the spatio-temporal cube, changing the observation perspective of the data. The time axis and the space axis are exchanged to analyze the change of the ecological index in space over time from different perspectives. Then, different ecological index dimensions are used as the main observation axes to analyze their dynamic relationships with other dimensions.

[0151] As some alternative implementation manners of this embodiment, the data is arranged in chronological order, and different ecological index data is included at each time point. Suppose the Sentinel-2 data (from 2019 to 2021) is as follows:

[0152] (1) Data in 2019

[0153] The NDVI data is as follows:

[0154] Table 5

[0155] NDVI1 NDVI2 NDVI3 0.2 0.3 0.4

[0157] The NDWI data is as follows:

[0158] Table 6

[0159] NDWI1 NDWI2 NDWI3 0.1 0.2 0.3

[0161] (2) Data in 2020

[0162] The NDVI data is as follows:

[0163] Table 7

[0164] NDVI1 NDVI2 NDVI3 0.3 0.4 0.5

[0166] The NDWI data is as follows:

[0167] Table 8

[0168] NDWI1 NDWI2 NDWI3 0.2 0.3 0.4

[0170] (3) Data in 2021

[0171] The NDVI data is as follows:

[0172] Table 9

[0173] NDVI1 NDVI2 NDVI3 0.4 0.5 0.6

[0175] The NDWI data is as follows:

[0176] Table 10

[0177] NDWI1 NDWI2 NDWI3 0.3 0.4 0.5

[0179] Specifically, after rotation: The data is mainly in the sequence of ecological indices, and each ecological index contains data at different time points. According to the data before rotation (Table 4-10), the data after rotation is: (1) Data in 2019

[0180] The NDVI data is as follows:

[0181] Table 11

[0182] NDVI1 NDVI2 NDVI3 0.2 0.3 0.4

[0184] The NDWI data is as follows:

[0185] Table 12

[0186] NDWI1 NDWI2 NDWI3 0.1 0.2 0.3 (2) Data in 2020

[0188] The NDVI data is as follows:

[0189] Table 13

[0190] NDVI1 NDVI2 NDVI3 0.3 0.4 0.5

[0192] The NDWI data is as follows:

[0193] Table 14

[0194] NDWI1 NDWI2 NDWI3 0.2 0.3 0.4 (2) Data in 2021

[0196] The NDVI data is as follows:

[0197] Table 15

[0198] NDVI1 NDVI2 NDVI3 0.4 0.5 0.6

[0200] The NDWI data is as follows:

[0201] Table 16

[0202] NDWI1 NDWI2 NDWI3 0.3 0.4 0.5

[0204] The coordinates are transformed through a rotation matrix, and then the ecological index values are calculated to achieve the calculation and analysis of the ecological index after the rotation operation. In ecological research, monitoring, and decision-making, the rotation matrix is used to transform the coordinate axes and has obvious advantages in the calculation and analysis of ecological indices. The rotation operator breaks the limitation of the traditional single perspective, provides multiple analysis dimensions, and reveals the time series relationship of ecological elements; rotating with different ecological index dimensions as the observation axes can reveal the dynamic correlations between elements, helping to understand the complexity and integrity of the ecosystem; it provides a flexible data processing method for the construction of ecological models, improves the simulation accuracy and reliability of the models, and serves the prediction and early warning of the ecological environment; it can provide comprehensive information for policymakers, assist them in grasping the spatio-temporal changes of the ecosystem, scientifically planning ecological protection strategies, enhancing the scientificity and effectiveness of decision-making, and promoting the sustainable development of the ecosystem.

[0205] According to the embodiments of the present invention, compared with the prior art, the present invention has the following advantages:

[0206] 1. Greatly improved multi-source data integration ability: effectively solves the problem of insufficient data integration in traditional methods, and can comprehensively integrate various desert-related data from different sources and in different formats, such as socio-economic data, ecological environment data, remote sensing image data, etc., into the desert ecological cube through innovative technical means. This enables the analysis of the desert ecosystem not to be limited to a single type of data, thereby obtaining more comprehensive and accurate information, and providing a solid data foundation for subsequent analysis and decision-making.

[0207] 2. Precise response to spatio-temporal resolution differences: skillfully handles the huge differences in time and space resolutions of different data sources. In the time dimension, through the integrated analysis of multi-source remote sensing data in different time periods, it realizes the leap from "single-time-period ecological index analysis" to "multi-temporal ecological index comprehensive analysis", accurately capturing the dynamic changes of the desert ecosystem over time. In the space dimension, it can reasonably process and analyze the data according to the spatial characteristics of different regions, enabling the analysis results to accurately reflect the heterogeneity of the desert ecosystem at different spatial positions, greatly improving the analysis accuracy.

[0208] 3. Optimized analysis results by the dynamic weight adjustment mechanism: introduces an advanced dynamic combination weight mechanism, abandoning the disadvantages of fixed weights or simple weighting in traditional methods. Dynamically adjusts the weights according to the time resolution, space resolution of the data, and the importance of the indicators, ensuring that high-resolution data and key indicators can be fully reflected during the analysis process. This enables the analysis of the desert ecosystem to more accurately reflect its complex change process, providing more valuable reference results for scientific research and policy-making.

[0209] 4. Construct a comprehensive ecological data foundation: A spatiotemporal data cube model from "ecological index numerical information" to "ecological index correlation information" was successfully established. The ecological index numerical value is combined with the inherent correlation knowledge between the elements of the ecosystem to achieve the systematic integration and semantic interpretation of the knowledge contained in the ecological index. This not only enriches the application connotation of the ecological index, but also creates an ecological data foundation based on spatiotemporal information for precise control and scientific research of desert ecology, which helps to gain a deeper understanding of the overall picture of desert ecology and provide strong support for the protection and sustainable development of desert ecosystems.

[0210] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should know that the present invention is not limited by the described order of actions, because according to the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily required by the present invention.

[0211] The above is an introduction to the method embodiment. The following is a further description of the scheme of the present invention through a space-time cube embodiment having the same inventive concept as the method in the above embodiment:

[0212] The space-time cube is constructed by a space-time cube construction method based on dynamic combination weights.

[0213] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the described module can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0214] In the technical solution of the present invention, the acquisition, storage and application of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0215] According to an embodiment of the present invention, the present invention further provides an electronic device.

[0216] Figure 4FIG. 0 shows a schematic block diagram of an electronic device 400 that can be used to implement an embodiment of the present invention. The electronic device is intended to represent various forms of digital computers, such as, for example, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, for example, personal digital processors, cellular telephones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely exemplary and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0217] The electronic device 400 includes a computing unit 401 that can perform various appropriate actions and processes in accordance with a computer program stored in a read-only memory (ROM) 402 or a computer program loaded from a storage unit 408 into a random access memory (RAM) 403. In the RAM 403, various programs and data required for the operation of the electronic device 400 can also be stored. The computing unit 401, the ROM 402, and the RAM 403 are connected to each other via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0218] A plurality of components in the electronic device 400 are connected to the I / O interface 405, including: an input unit 406, such as a keyboard, a mouse, etc.; an output unit 407, such as various types of displays, speakers, etc.; a storage unit 408, such as a magnetic disk, an optical disk, etc.; and a communication unit 409, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 409 allows the electronic device 400 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0219] The computing unit 401 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 401 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 401 executes the various methods and processes described above, such as methods S101 - S102. For example, in some embodiments, methods S101 - S102 can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 408. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 400 via the ROM 402 and / or the communication unit 409. When the computer program is loaded into the RAM 403 and executed by the computing unit 401, one or more steps of the methods S101 - S102 described above can be executed. Alternatively, in other embodiments, the computing unit 401 can be configured to execute methods S101 - S102 by any other suitable means (e.g., by means of firmware).

[0220] Various embodiments of the systems and techniques described above in this document can be implemented in digital electronic circuitry, integrated circuit systems, field-programmable gate arrays (FPGA), application-specific integrated circuits (ASIC), application-specific standard products (ASSP), system-on-a-chip (SOC), complex programmable logic devices (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special-purpose or general-purpose programmable processor, that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0221] The program code for implementing the methods of the present invention can be written in any combination of one or more programming languages. These program codes can be provided to the processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the program code is executed by the processor or controller, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0222] It should be understood that the various forms of processes shown above can be used, with steps reordered, added or deleted. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and no limitation is made herein.

[0223] The above specific embodiments do not constitute a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub - combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for constructing a space-time cube based on dynamic combination weights, characterized in that: include: Obtaining ecological index data of a human-land system region, and constructing a three-dimensional space-time cube with the ecological index data according to spatial dimension, temporal dimension and ecological indicator dimension; The three-dimensional space-time cube is composed of a number of cube elements, each of which is a geographical grid carrying an ecological index of spatial position, time series and dynamic combination weight; An operation operator set of the three-dimensional space-time cube is defined to obtain a space-time cube based on dynamic combination weights.

2. The method according to claim 1, characterized in that The ecological index of the dynamic combined weight includes: Among them, EC(p(x,y)) is the ecological index of dynamic combined weights; p(x,y) is the pixel at the spatial position (x,y); i is the index value of the year; n is the number of years in the time range; is the weight after dynamic adjustment; z i is the ith ecological index; t is the time point.

3. The method according to claim 1, characterized in that: The operator set includes: Cutting operator, used to extract sub-datasets within a specified time and space range; Aggregation operator, used to aggregate statistics of ecological indexes in time and space dimensions; Projection operator, which reduces the dimensionality of three-dimensional data according to the time-ecological indicator dimension or space-ecological indicator dimension to generate a two-dimensional data set; Rotation operator, used to transform coordinate axes using a rotation matrix.

4. The method according to claim 3, characterized in that: The cutting operator includes time slicing, space slicing and dicing; The time slice is used to extract ecological index data within a specified time range; the time slice includes: The spatial slice is used to extract ecological index data within a specified spatial range; the spatial slice includes: The cutting is used to extract ecological index data of data blocks within a specified time and space range; the cutting includes: Among them, EC ts (p(x,y)) is the time slice; The spatial position is (x, y), the ecological index is z, and the starting time is t start The pixel value of The spatial position is (x, y), the ecological index is z, and the starting time is t end The pixel value of is the weight after dynamic adjustment; EC ss (p(x,y)) is a spatial slice; i is the index value of the year; n is the number of years in the time range; z i is the i-th ecological index; t is the time point; s x1 is the starting coordinate of the spatial range on the x-axis in the longitude direction; s x2 is the ending coordinate of the spatial range on the x-axis in the longitude direction; s y1 is the starting coordinate of the spatial range on the y-axis in the latitude direction; s y2 is the starting coordinate of the spatial range on the y-axis in the latitude direction; x is the longitude spatial axis; y is the latitude spatial axis; EC block (p(x,y)) is the cut block.

5. The method according to claim 3, characterized in that: The aggregation operator includes time aggregation and space aggregation; The time aggregation is used to calculate the statistics of ecological index data of a specific area within a specified time range; The spatial aggregation is used to calculate the statistics of ecological index data at a specific time point within a specified spatial range.

6. The method according to claim 5, characterized in that The statistics of the ecological index data of a specific area within the specified time range include the maximum value of time aggregation, the minimum value of time aggregation and the standard deviation of time aggregation; The maximum time aggregation value includes: The time aggregation minimum value includes: EC min,t = min{EC(p(x,y)) (z,t) | t start ≤ t ≤ t end and (x, y) ∈ W} The time aggregation standard deviation includes: Among them, EC max,t is the maximum value of temporal aggregation; p(x,y) is the pixel at the spatial position (x,y); i is the index value of the year; n is the number of years in the time range; is the weight after dynamic adjustment; z i is the i-th ecological index; t is the time point; t start is the starting time; t end is the end time; W is a specific area; EC min,t is the minimum value of time aggregation; σ EC,t is the time-aggregated standard deviation; is the average value of the ecological index in the region within the specified time range; m is the number of pixels in the specific area W.

7. The method according to claim 5, characterized in that The statistics of the ecological index data at a specific time point within the specified spatial range include the spatial aggregation mean, spatial aggregation maximum, spatial aggregation minimum and spatial aggregation standard deviation; The spatial aggregation mean includes: The maximum spatial aggregation value includes: The minimum spatial aggregation value includes: The spatial aggregation standard deviation includes: in, is the spatial aggregation mean; p(x,y) is the pixel at the spatial position (x,y); i is the index value of the year; n is the number of years in the time range; is the weight after dynamic adjustment; z i is the i-th ecological index; t is the time point; W is the specific area; σ EC,s is the spatial aggregation standard deviation, which is an indicator to measure the degree of dispersion of ecological index in a specified spatial range W at a specific time t0; m is the number of pixels in a specific area W; t0 is the first specific time point; EC max,s is the maximum value of spatial aggregation; EC min,s is the minimum value of spatial aggregation; t3 is the second specific time point.

8. The method according to claim 3, characterized in that The coordinate axis conversion by the rotation matrix includes: The rotation matrix is: The coordinate axis conversion includes: Among them, R is the rotation matrix; x is the longitude space axis; y is the latitude space axis; z is the time axis; x' is the converted longitude space axis; y' is the converted latitude space axis; z' is the converted time axis.

9. A space-time cube based on dynamic combination weights, characterized in that: include: The space-time cube is constructed by the space-time cube construction method based on dynamic combination weights described in any one of claims 1 to 8.

10. An electronic device comprising at least one processor; and a memory connected to the at least one processor in communication; characterized in that: The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 8.

Images