Land utilization monitoring method and system based on big data

Abstract

The invention relates to the technical field of land utilization, in particular to a land utilization monitoring method and system based on big data, and the method comprises the following steps: based on collected satellite image data, detecting the intensity change of pixels in an image, carrying out texture scanning and color analysis, and recognizing and determining the land boundary of the image; and matching the image data with geographic coordinates by using a GIS technology. According to the invention, by integrating and optimizing satellite image data and a GIS technology, accurate earth surface feature mapping is realized, the accuracy of land boundary identification is improved, the analysis and classification process of land features is optimized, the management of land resources becomes more accurate, and by dynamically calculating the pixel similarity in different image data, the accuracy of land boundary identification is improved. The method optimizes the refining degree of land classification, provides more detailed data support, combines real-time meteorological and surface temperature data, allows a monitoring system to analyze the influence of environmental changes on land utilization in real time, and improves the real-time performance of decision making.

Description

Technical Field

[0001] The present invention relates to the field of land use technology, and in particular to a land use monitoring method and system based on big data. Background Art

[0002] Land use technology covers multiple aspects such as planning, management, monitoring and evaluation of land resources, aiming to achieve optimal allocation and sustainable utilization of land resources. This technical field involves the use of remote sensing, GIS (geographic information system), big data analysis and other technologies to comprehensively monitor and analyze land use, distribution, and changing trends, providing accurate data support for land management, helping to understand the current status and changes in land use, and ensuring the rational allocation and effective utilization of land resources.

[0003] Big data land use monitoring methods, among others, utilize big data technology to monitor and analyze land resource usage. By integrating multi-source information such as satellite remote sensing data, geographic information system (GIS) data, and socioeconomic activity data, combined with machine learning and data mining techniques, they enable real-time monitoring and predictive analysis of land use changes. Their primary purpose is to provide a scientific basis for decision-making in areas such as optimizing land resource allocation, urban and rural planning, and environmental protection, while also supporting dynamic assessment, trend prediction, and problem warning for land use.

[0004] Although existing technologies utilize remote sensing, GIS, and big data methods for land monitoring, they have defects in data integration efficiency and processing speed. For example, the precise mapping of land characteristics and the lag in change monitoring often lead to untimely or inaccurate decision-making, the inability to fully realize the potential value of data, and a slow response to environmental change factors. The failure to update the data processing mechanism in a timely manner increases the risks of environmental protection and resource allocation, causes planning errors in resource management, and leads to irrational allocation and utilization of resources, making it impossible to achieve the optimal use of land resources and affecting the efficiency of land use monitoring. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the existing technology and propose a land use monitoring method and system based on big data.

[0006] In order to achieve the above purpose, the present invention adopts the following technical solution: a land use monitoring method based on big data, comprising the following steps:

[0007] S1: Based on the collected satellite image data, it detects pixel intensity changes in the image, performs texture scanning and color analysis, identifies and determines the land boundaries of the image, uses GIS technology to match image data with geographic coordinates, and optimizes the coordinate comparison process to obtain surface feature information;

[0008] S2: Based on the surface feature information, the similarity between adjacent pixel blocks is calculated, the consistency of color and texture is determined, various land cover types are analyzed, and the classification process is optimized in combination with existing land classification standards to obtain land attribute segmentation results;

[0009] S3: Based on the land attribute segmentation results, real-time meteorological data and surface temperature data are integrated, and remote sensing data is combined to perform multi-dimensional parameter analysis to evaluate the impact of environmental factors on land use and obtain environmental impact indicators;

[0010] S4: Based on the environmental impact indicators, analyze the satellite images at different time points to detect the areas of land cover change, compare them with the time series data, identify the dynamic change trend of land cover, and obtain land cover change records.

[0011] The present invention is improved in that the steps of identifying and determining the land boundary of the image are specifically as follows:

[0012] S111: Based on the collected satellite image data, performing denoising and contrast adjustment on the image data to optimize the distinction between land and non-land areas in the image, thereby obtaining improved image data;

[0013] S112: Applying an edge detection algorithm to extract land boundaries from the improved image data using the formula: ;

[0014] Calculate each pixel The gradient value , perform boundary point screening and obtain a set of boundary points, where and Represent the first-order derivatives of the image in the x and y directions respectively;

[0015] S113: Performing texture scanning and color analysis on the boundary point set to verify and correct the initially identified boundary to obtain an accurate land boundary.

[0016] The present invention is improved in that the step of obtaining the surface feature information is specifically as follows:

[0017] S121: using the GIS technology, matching the image data of the land boundary with the geographic coordinates, and obtaining image data matching the geographic coordinates through a coordinate conversion and boundary alignment process;

[0018] S122: Analyze the surface characteristics based on the image data matched with the geographic coordinates, using the formula:

[0019] ;

[0020] Get surface feature information score ,in, represents the land use type index, represents the vegetation coverage rate, Scoring other environmental indicators, and are weight coefficients, corresponding to the land use type index, vegetation coverage and other environmental indicator scores.

[0021] The present invention is improved in that the step of determining the consistency of color and texture is specifically as follows:

[0022] S211: extracting a color histogram and texture features of the image data based on the surface feature information, and generating a color and texture descriptor for each pixel block;

[0023] S212: Compare the color and texture descriptors of adjacent pixel blocks using the formula:

[0024] ;

[0025] Generate a consistency score for each pair of pixel patches ,in, is the color difference, calculated by the difference in color histograms, is the difference of texture, which is measured by the similarity of texture features. is the weight used to balance the influence of color and texture;

[0026] S213: Analyze the consistency score and determine whether the color and texture consistency is met by setting a threshold. If the score is lower than the threshold, the two pixel blocks are considered to be visually consistent, and a consistency determination result is obtained.

[0027] The present invention is improved in that the steps for obtaining the land attribute subdivision result are specifically as follows:

[0028] S221: Clustering the image data according to regions with high consistency to obtain image block classifications. Using existing land classification standards, performing land attribute analysis on the image block classifications, and combining color and texture data to determine the land cover type of each block, thereby obtaining land classification information.

[0029] S222: Optimize and refine the land classification information, compare existing land use data and GIS data, adjust and improve land attributes, and obtain land attribute segmentation results.

[0030] The present invention is improved in that the steps of obtaining the environmental impact index are specifically as follows:

[0031] S311: Based on the land attribute segmentation result, integrating real-time meteorological data and surface temperature data to generate a climate and temperature dataset;

[0032] S312: Based on the climate and temperature dataset, spatial mapping is performed in combination with remote sensing data using the formula:

[0033] ;

[0034] Calculate the mutual influence between multidimensional parameters and obtain the environmental assessment score ,in, represents the surface temperature, represents the climate parameters, Represents remote sensing data, is the weight of the surface temperature, is the weight of the climate parameter, is the weight of remote sensing data;

[0035] S313: Based on the environmental assessment score, assess the impact on land use, determine the extent to which each land block is affected by environmental variables, and obtain an environmental impact index.

[0036] The present invention is improved in that the step of detecting the land cover change area is specifically as follows:

[0037] S411: Based on the environmental impact indicators, integrating satellite images at different time points, collecting and processing image data, and obtaining a satellite image dataset for analysis;

[0038] S412: Analyze the satellite image dataset to detect differences between images using the formula:

[0039] ;

[0040] Calculate the average change intensity of each pixel , and obtain the change intensity map, where is the pixel value at the current time point, is the pixel value at the previous time point, is the total number of pixels;

[0041] S413: Analyze the change intensity map to identify key change areas, including vegetation degradation or urban expansion, and obtain regional change information.

[0042] The present invention is improved in that the steps of obtaining the land cover change record are specifically as follows:

[0043] S421: Compare with the time series data using the formula:

[0044] ;

[0045] Calculate the regional intensity of the changing trend , and obtain the results of the trend analysis, among which, Represents the image data at the current time point, Represents the image data at the previous time point, is the total regional area;

[0046] S422: According to the change trend analysis results, record the change type, time point and impact range of each change area to obtain a land cover change record.

[0047] A land use monitoring system based on big data, comprising:

[0048] The feature mapping module extracts pixel intensity information from the image based on the collected satellite image data, detects pixel changes, identifies land boundaries in the image, performs texture scanning and color analysis, and uses GIS technology to match image data with geographic coordinates, optimize the coordinate comparison process, and obtain surface feature information;

[0049] The image analysis module calculates the similarity of adjacent pixel blocks based on the surface feature information, determines the consistency of color and texture, analyzes various land cover types, and optimizes the classification process in combination with existing land classification standards to obtain land attribute segmentation results;

[0050] The classification assessment module integrates the real-time meteorological data and surface temperature data based on the land attribute segmentation results, and combines remote sensing data to perform multi-dimensional parameter analysis to evaluate the impact of environmental factors on land use and obtain environmental impact indicators;

[0051] The environmental monitoring module analyzes satellite images at different time points based on the environmental impact indicators, detects the spatial characteristics of land cover changes, and analyzes their impact on land use to obtain land change data;

[0052] The dynamic analysis module compares the land change data with the historical time series data to identify the dynamic change trend of land cover and obtain the land use trend analysis results.

[0053] Compared with the prior art, the advantages and positive effects of the present invention are:

[0054] In the present invention, by integrating and optimizing satellite image data and GIS technology, accurate surface feature mapping is achieved, the accuracy of land boundary identification is improved, the analysis and classification process of land features is optimized, and the management of land resources becomes more precise. By dynamically calculating the pixel similarity in different image data, the degree of refinement of land classification is optimized, thereby providing more detailed data support. Combined with real-time meteorological and surface temperature data, the monitoring system is allowed to analyze the impact of environmental changes on land use in real time, improving the real-time nature of decision-making. By comparing image data at different points in time and tracking land cover changes, the monitoring capability of land use dynamics is improved, and the timeliness of problem warning is strengthened. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 The present invention proposes a flow chart of a land use monitoring method based on big data;

[0056] Figure 2 A flow chart for identifying and determining land boundaries of an image in the present invention;

[0057] Figure 3 This is a flow chart for obtaining surface feature information in the present invention;

[0058] Figure 4 This is a flow chart for determining the consistency of color and texture in the present invention;

[0059] Figure 5 This is a flow chart for obtaining land attribute segmentation results in the present invention;

[0060] Figure 6 This is a flow chart for obtaining environmental impact indicators in the present invention;

[0061] Figure 7 This is a flow chart for detecting land cover change areas in the present invention;

[0062] Figure 8 This is a flow chart for obtaining land cover change records in the present invention. DETAILED DESCRIPTION

[0063] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0064] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings and are only for the convenience of describing the present invention and simplifying the description. They do not indicate or imply that the devices or elements referred to must have a specific direction, be constructed and operate in a specific direction, and therefore should not be understood as limiting the present invention. In addition, in the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0065] For examples, see Figure 1 The present invention provides a technical solution: a land use monitoring method based on big data, comprising the following steps:

[0066] S1: Based on the collected satellite image data, it detects pixel intensity changes in the image, performs texture scanning and color analysis, identifies and determines the land boundaries of the image, uses GIS technology to match image data with geographic coordinates, and optimizes the coordinate comparison process to obtain surface feature information;

[0067] S2: Based on surface feature information, calculate the similarity between adjacent pixel blocks, determine the consistency of color and texture, analyze various land cover types, and combine existing land classification standards to optimize the classification process and obtain land attribute segmentation results;

[0068] S3: Based on the land attribute segmentation results, real-time meteorological data and surface temperature data are integrated with remote sensing data to perform multi-dimensional parameter analysis to evaluate the impact of environmental factors on land use and obtain environmental impact indicators;

[0069] S4: Based on environmental impact indicators, analyze satellite images at different time points to detect areas of land cover change, compare them with time series data, identify dynamic change trends of land cover, and obtain land cover change records.

[0070] Surface feature information includes terrain height, terrain slope, and terrain orientation. Land attribute subdivision results include cultivated land type, grassland range, and building land area. Environmental impact indicators include temperature fluctuations, precipitation frequency, and sunshine duration. Land cover change records specifically refer to changes in land area, land use conversion information, and vegetation cover changes.

[0071] See also Figure 2 , the steps for identifying and determining the land boundary of the image are as follows:

[0072] S111: Based on the collected satellite image data, performing denoising and contrast adjustment on the image data to optimize the distinction between land and non-land areas in the image, thereby obtaining improved image data;

[0073] Using a multi-scale median filtering method, the median value of the neighborhood is calculated pixel by pixel and replaced with the original pixel value. The image contrast is then adjusted in combination with the Laplace enhancement algorithm. The second-order derivative of each pixel is calculated and superimposed on the original image value to improve the visual segmentation effect of land and non-land areas. Histogram equalization is performed on the denoised image to balance the brightness and contrast of the image by adjusting the grayscale value distribution, and ultimately highlight the differences between regions. After all processing steps are applied sequentially to the original image, the improved image data is finally obtained.

[0074] S112: Apply edge detection algorithm to extract land boundaries from the improved image data using the formula: ;

[0075] Calculate each pixel The gradient value , perform boundary point screening and obtain a set of boundary points, where and Represent the first-order derivatives of the image in the x and y directions respectively;

[0076] A specific image pixel position is (150, 150), and its pixel intensity gradients in the x and y directions are and , according to the formula, calculate the gradient amplitude of the point:

[0077] ;

[0078] The result shows that the gradient amplitude of this pixel is large, indicating that there is a significant image edge here and it is part of the land boundary.

[0079] S113: Perform texture scanning and color analysis on the boundary point set to verify and correct the initially identified boundary to obtain an accurate land boundary;

[0080] First, a fixed-size sliding window is set around each boundary point, and the gray-level co-occurrence matrix within the window is calculated to obtain the corresponding texture feature parameters (such as contrast, entropy, etc.). Then, color analysis is performed to extract the RGB value of each pixel in the window and calculate the average value. By comparing the distribution patterns of texture parameters and color averages, abnormal boundary points are identified and eliminated. The corrected boundary point set is then processed by curve fitting to generate a continuous boundary curve to obtain an accurate land boundary.

[0081] See also Figure 3 , the specific steps for obtaining surface feature information are:

[0082] S121: using GIS technology, matching the image data of the land boundary with the geographic coordinates, and obtaining image data matching the geographic coordinates through a coordinate conversion and boundary alignment process;

[0083] The image data needs to be partitioned first, and the boundary information is divided into several image blocks. Each image block is matched with geographic coordinates separately. The coordinate matching process is completed by calling the correspondence table between known geographic coordinate points and image pixel points. The bilinear interpolation method is used to adjust the pixel position of the image boundary so that it is accurately aligned with the geographic coordinate grid points. In this process, the geographic projection difference needs to be considered. The image data is converted from the original projection to a unified geographic projection to ensure projection consistency. The aligned boundary line is then verified and the coordinate point deviation value is checked pixel by pixel. If the deviation exceeds the set threshold, the coordinate matching algorithm parameters are readjusted until the deviation values of all points are below the threshold range, and image data matching the geographic coordinates is obtained.

[0084] S122: Analyze the surface characteristics based on the image data matching the geographic coordinates using the formula:

[0085] ;

[0086] Get surface feature information score ,in, It represents the land use type index, which reflects the extent to which land is used for different purposes (such as agriculture, commerce, residential, etc.). Represents vegetation coverage, which indicates the coverage and density of vegetation in a specific area. Scoring of other environmental indicators, including a collection of factors such as water cover, soil moisture, and topographic changes, and are weight coefficients, corresponding to the land use type index, vegetation coverage and other environmental index scores;

[0087] The actual parameters are , reflecting a higher land utilization rate, , indicating a medium degree of vegetation cover, , showing a better environmental quality, the weight factor is set to 、 and , the calculation process is:

[0088] ;

[0089] The results show that the score It is pointed out that in the analyzed area, the land use efficiency and environmental quality are good, but the vegetation coverage is slightly low, indicating that the area needs certain improvements in vegetation restoration and environmental protection.

[0090] See also Figure 4 ,The steps to determine the consistency of color and texture are as follows:

[0091] S211: extracting the color histogram and texture features (such as gray-level co-occurrence matrix) of the image data based on the surface feature information, and generating color and texture descriptors for each pixel block;

[0092] Each pixel block in the image data is partitioned, and the color information of each pixel block is divided into three channels and calculated separately to generate histogram distribution data of red, green and blue colors. At the same time, by calculating the grayscale co-occurrence matrix, the grayscale value arrangement pattern and directionality between pixel blocks are analyzed, and statistical features such as texture contrast, entropy and consistency are extracted. The color and texture information are normalized to ensure that the numerical range of the descriptor is consistent, and color and texture descriptors are generated. The descriptors are the basic data for subsequent analysis and accurately reflect the color distribution and texture characteristics of each pixel block.

[0093] S212: Compare the color and texture descriptors of adjacent pixel blocks using the formula:

[0094] ;

[0095] Generate a consistency score for each pair of pixel patches ,in, is the color difference, calculated by the difference in color histograms, is the difference of texture, which is measured by the similarity of texture features. is the weight used to balance the influence of color and texture;

[0096] For a group of adjacent pixel blocks, the color difference and texture difference , weight , substitute into the formula to calculate:

[0097] ;

[0098] ;

[0099] ;

[0100] ;

[0101] The result shows that when the weights of color and texture are evenly distributed, the resulting consistency score is 0.538, which reflects the degree of visual consistency between the two pixel blocks. A lower score indicates that the two are more visually consistent.

[0102] S213: Analyze the consistency score and determine whether the color and texture consistency is met by setting a threshold. If the score is lower than the threshold, the two pixel blocks are considered to be visually consistent, and a consistency judgment result is obtained;

[0103] The threshold is set according to the application context. Assume that the threshold is 0.6. When the consistency score is lower than this threshold, such as 0.538 in the above example, since it is lower than the threshold, the two pixel blocks are considered to be visually consistent, making the consistency judgment more intuitive and easy to operate. The obtained consistency judgment result can be directly applied to image segmentation, object recognition, etc.

[0104] See also Figure 5 , the specific steps for obtaining land attribute segmentation results are:

[0105] S221: Clustering the image data according to regions with high consistency to obtain image block classifications. Using existing land classification standards, performing land attribute analysis on the image block classifications, and combining color and texture data to determine the land cover type of each block, thereby obtaining land classification information.

[0106] Cluster analysis is performed on the generated color and texture consistency judgment results. The pixels are divided into different area groups by similarity threshold. The pixels in each group have high consistency in color and texture. Then, the clustering results are attribute analyzed using the land classification standard. The characteristic mean of color and texture is calculated for each group of areas to evaluate its land cover type. By comparing the characteristic values of each area with the predefined land classification characteristic range, the corresponding land cover category is determined. After completion, the land cover type information of each block is generated.

[0107] S222: Optimize and refine land classification information, compare existing land use data and GIS data, adjust and improve land attributes, and obtain land attribute subdivision results;

[0108] The land classification results of each area are compared with the existing land use data to identify inconsistencies or anomalies in the classification. Then, the boundaries of each area are adjusted based on the geographic coordinate information in the GIS data, and the classification results of the boundary pixels are re-evaluated to ensure that the regional boundaries are consistent with the actual land use patterns. Finally, combined with the distribution trend of the surface feature data, the classification results of each area are locally optimized to ensure the accuracy and completeness of the information and obtain the land attribute subdivision results.

[0109] See also Figure 6 , the specific steps for obtaining environmental impact indicators are as follows:

[0110] S311: Based on the land attribute segmentation results, real-time meteorological data and surface temperature data are integrated to generate climate and temperature datasets;

[0111] First, the land parcel boundary data is extracted from the land attribute subdivision results of the partition to accurately locate the spatial distribution range of the data. Then, real-time meteorological data, including environmental parameters such as temperature, humidity, and wind speed, are collected through meteorological monitoring equipment. At the same time, surface temperature values are obtained from surface temperature sensors or satellite monitoring. All data are unified to the same temporal and spatial resolutions. Finally, through parameter alignment, time series smoothing and spatial interpolation methods, the above multi-source data are integrated to generate a climate and temperature dataset that contains real-time climate conditions and surface temperature.

[0112] S312: Based on climate and temperature datasets, combined with remote sensing data for spatial mapping, remote sensing images obtained by satellites and drones, using the formula:

[0113] ;

[0114] Calculate the mutual influence between multidimensional parameters and obtain the environmental assessment score ,in, represents the surface temperature, represents the climate parameters, Represents remote sensing data, is the weight of the surface temperature, is the weight of the climate parameter, is the weight of remote sensing data;

[0115] The following data were collected: surface temperature Degrees, climate parameters (such as humidity 80%), remote sensing data (a remote sensing index value), the weights are , , , substitute into the formula to calculate:

[0116] ;

[0117] ;

[0118] The results show that the comprehensive environmental score is 21.24, which reflects the comprehensive environmental conditions of the current observation point. A higher score indicates that the land is greatly affected by environmental factors and corresponding land management and protection measures need to be taken.

[0119] S313: Based on the environmental assessment score, assess the impact on land use, determine the extent to which each land block is affected by environmental variables, and obtain an environmental impact index;

[0120] First, the score of each land block is extracted from the environmental assessment score and matched with the corresponding land attribute segmentation results. Then, combined with the historical environmental data in the region, the sensitivity of the regional environmental impact is identified by comparing the current assessment score with the historical score change trend. Finally, according to the set environmental impact threshold, different sensitivity areas are graded and specific indicators are generated to represent the degree of environmental impact. The indicators will be used to further analyze the suitability and potential risks of land use.

[0121] See also Figure 7 ,The detection steps of the land cover change area are as follows:

[0122] S411: Based on the environmental impact indicators, integrate satellite images at different time points, collect and process image data, and obtain a satellite image dataset for analysis;

[0123] By analyzing the spatiotemporal characteristics of environmental impact indicators, the acquisition time and geographical range of satellite images are determined. Then, the collected original satellite image data are geometrically corrected to ensure the spatial alignment of images at different time points and to filter out noise in the images. Multispectral information is then extracted from the processed images, including parameters such as vegetation index, surface temperature, and humidity. All processed and extracted image data are combined in a unified data structure to generate a satellite image dataset.

[0124] S412: Analyze the satellite image dataset and detect differences between images using the formula:

[0125] ;

[0126] Calculate the average change intensity of each pixel , and obtain the change intensity map, where is the pixel value at the current time point, is the pixel value at the previous time point, is the total number of pixels;

[0127] In a specific area, The pixel value is 120, The pixel value is 100, the total number of pixels considered is 1000, It is the sum of the change intensities of all pixels, not the difference between individual pixels. Substituting it into the formula, we get:

[0128] ;

[0129] The results show that the average change intensity per pixel in the considered area is 20, which indicates that there are significant image differences in the area, which are caused by vegetation changes or construction activities.

[0130] S413: Analyze the change intensity map to identify key change areas, including vegetation degradation or urban expansion, and obtain regional change information;

[0131] Pixel points with significant changes are extracted from the change intensity map, and the spatial clustering characteristics of the points are calculated. The clustered areas are then compared with land cover classification data to identify the type of change, such as vegetation reduction or urban land increase. Finally, the dynamic trend of the change area is further analyzed by combining time series data, recording the location, type and impact range of the changes within the area to obtain regional change information.

[0132] See also Figure 8 , the specific steps for obtaining land cover change records are:

[0133] S421: Compare with time series data using the formula:

[0134] ;

[0135] Calculate the regional intensity of the changing trend , and obtain the results of the trend analysis, among which, Represents the image data at the current time point, Represents the image data at the previous time point, is the total regional area;

[0136] Image data at the current time point The value is 120, the image data at the previous time point The value is 100, the total area If is 50, the calculation process is:

[0137] ;

[0138] The results show that the obtained trend analysis results show that the average change intensity of each pixel is 0.4, indicating that the land cover has changed significantly between the two time points.

[0139] S422: Based on the change trend analysis results, record the change type, time point, and impact range of each change area to obtain a land cover change record;

[0140] The data in the change trend analysis results are grouped and organized according to geographic location and time to ensure that the data in each change area is complete and non-repetitive. Then, the change intensity, change type and other information of each area are compared with the actual observed change type to confirm the specific category of change, such as vegetation degradation or building expansion. The specific time point of each change event is then recorded to determine the specific time range of the change. In combination with the regional impact range, the spatial scale of coverage and the ripple effect are recorded. The detailed information is summarized to form a land cover change record.

[0141] A land use monitoring system based on big data, the system comprising:

[0142] The feature mapping module extracts pixel intensity information from the image based on the collected satellite image data, detects pixel changes, identifies land boundaries in the image, performs texture scanning and color analysis, and uses GIS technology to match image data with geographic coordinates, optimize the coordinate comparison process, and obtain surface feature information;

[0143] The image analysis module calculates the similarity of adjacent pixel blocks based on surface feature information, determines the consistency of color and texture, analyzes various land cover types, and optimizes the classification process in combination with existing land classification standards to obtain land attribute segmentation results;

[0144] The classification assessment module integrates real-time meteorological data and surface temperature data based on land attribute segmentation results, and combines remote sensing data to perform multi-dimensional parameter analysis to evaluate the impact of environmental factors on land use and obtain environmental impact indicators;

[0145] The environmental monitoring module analyzes satellite images at different time points based on environmental impact indicators, detects the spatial characteristics of land cover changes, and analyzes their impact on land use to obtain land change data;

[0146] The dynamic analysis module compares land change data with historical time series data to identify the dynamic change trend of land cover and obtain land use trend analysis results.

[0147] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. A land use monitoring method based on big data, characterized in that: The following steps are involved: Based on the collected satellite image data, we detect pixel intensity changes in the image, perform texture scanning and color analysis, identify and determine the land boundaries of the image, and use GIS technology to match image data with geographic coordinates to obtain surface feature information; Based on the surface feature information, the similarity between adjacent pixel blocks is calculated, the consistency of color and texture is determined, various land cover types are analyzed, and the land attribute segmentation results are obtained by combining the existing land classification standards; Based on the land attribute segmentation results, real-time meteorological data and surface temperature data are integrated, and multi-dimensional parameter analysis is performed in combination with remote sensing data to evaluate the impact of environmental factors on land use and obtain environmental impact indicators; Based on the environmental impact indicators, satellite images at different time points are analyzed to detect land cover change areas, and compared with time series data to identify the dynamic change trend of land cover and obtain land cover change records.

2. The land use monitoring method based on big data according to claim 1, characterized in that: The steps of identifying and determining the land boundary of the image are specifically as follows: Based on the collected satellite image data, the image data is denoised and contrast adjusted to optimize the distinction between land and non-land areas in the image to obtain improved image data; Apply edge detection algorithm to extract land boundary from the improved image data using the formula: ； Calculate each pixel The gradient value , perform boundary point screening and obtain a set of boundary points, where and Represent the first-order derivatives of the image in the x and y directions respectively; Texture scanning and color analysis are performed on the boundary point set to verify and correct the initially identified boundary to obtain an accurate land boundary.

3. The land use monitoring method based on big data according to claim 1, characterized in that: The steps for obtaining the surface feature information are specifically as follows: Using the GIS technology, the image data of the land boundary is matched with the geographic coordinates, and through the coordinate conversion and boundary alignment process, the image data matching the geographic coordinates is obtained; Based on the image data matched with the geographic coordinates, surface feature analysis is performed using the formula: ； Get surface feature information score ,in, represents the land use type index, represents the vegetation coverage rate, Scoring other environmental indicators, and are weight coefficients, corresponding to the land use type index, vegetation coverage and other environmental indicator scores.

4. The land use monitoring method based on big data according to claim 1, characterized in that: The steps of determining the consistency of color and texture are specifically as follows: Extracting a color histogram and texture features of the image data based on the surface feature information, and generating a color and texture descriptor for each pixel block; The color and texture descriptors of adjacent pixel blocks are compared using the formula: ； Generate a consistency score for each pair of pixel patches ,in, is the color difference, calculated by the difference in color histograms, is the difference of texture, which is measured by the similarity of texture features. is the weight used to balance the influence of color and texture; The consistency score is analyzed, and a threshold is set to determine whether the color and texture consistency is met. If the score is lower than the threshold, the two pixel blocks are considered to be visually consistent, and a consistency judgment result is obtained.

5. The land use monitoring method based on big data according to claim 1, characterized in that: The steps for obtaining the land attribute segmentation results are as follows: Cluster the image data according to the areas with high consistency to obtain the image block classification. Use the existing land classification standards to analyze the land attributes of the image block classification, and combine the color and texture data to determine the land cover type of each block to obtain land classification information. The land classification information is optimized and refined, and the existing land use data and GIS data are compared to adjust and improve the land attributes to obtain the land attribute segmentation results.

6. The land use monitoring method based on big data according to claim 1, characterized in that: The specific steps for obtaining the environmental impact indicators are as follows: Based on the land attribute segmentation results, real-time meteorological data and surface temperature data are integrated to generate a climate and temperature dataset; Based on the climate and temperature dataset, combined with remote sensing data, spatial mapping is performed using the formula: ； Calculate the mutual influence between multidimensional parameters and obtain the environmental assessment score ,in, represents the surface temperature, represents the climate parameters, Represents remote sensing data, is the weight of the surface temperature, is the weight of the climate parameter, is the weight of remote sensing data; Based on the environmental assessment score, the degree of impact on land use is assessed, the degree to which each land block is affected by environmental variables is determined, and an environmental impact index is obtained.

7. The land use monitoring method based on big data according to claim 1, characterized in that: The steps for detecting the land cover change area are specifically as follows: Based on the environmental impact indicators, integrating satellite images at different time points, collecting and processing image data, and obtaining a satellite image dataset for analysis; The satellite image dataset is analyzed to detect differences between images using the formula: ； Calculate the average change intensity of each pixel , and obtain the change intensity map, where is the pixel value at the current time point, is the pixel value at the previous time point, is the total number of pixels; The change intensity map is analyzed to identify key change areas, including vegetation degradation or urban expansion, and obtain regional change information.

8. The land use monitoring method based on big data according to claim 1, characterized in that: The steps for obtaining the land cover change records are specifically as follows: Comparing with the time series data, the formula is used: ； Calculate the regional intensity of the changing trend , and obtain the change trend analysis results, among which, Represents the image data at the current time point, Represents the image data at the previous time point, is the total regional area; According to the change trend analysis results, the change type, time point and impact range of each change area are recorded to obtain a land cover change record.

9. A land use monitoring system based on big data, characterized in that: The method for land use monitoring based on big data according to any one of claims 1 to 8 is implemented, wherein the system comprises: The feature mapping module extracts pixel intensity information from the image based on the collected satellite image data, detects pixel changes, identifies land boundaries in the image, performs texture scanning and color analysis, and uses GIS technology to match image data with geographic coordinates, optimize the coordinate comparison process, and obtain surface feature information; The image analysis module calculates the similarity of adjacent pixel blocks based on the surface feature information, determines the consistency of color and texture, analyzes various land cover types, and optimizes the classification process in combination with existing land classification standards to obtain land attribute segmentation results; The classification assessment module integrates the real-time meteorological data and surface temperature data based on the land attribute segmentation results, and combines remote sensing data to perform multi-dimensional parameter analysis to evaluate the impact of environmental factors on land use and obtain environmental impact indicators; The environmental monitoring module analyzes satellite images at different time points based on the environmental impact indicators, detects the spatial characteristics of land cover changes, and analyzes their impact on land use to obtain land change data; The dynamic analysis module compares the land change data with the historical time series data to identify the dynamic change trend of land cover and obtain the land use trend analysis results.

Images