A water conservation function assessment method based on multi-source data

Abstract

The present invention discloses a water conservation function evaluation method based on multi-source data, which belongs to the field of data analysis technology and includes the following steps: S1, obtaining a water sample set of a region, and cleaning the water sample set to obtain a standard water sample set; S2, constructing a characteristic response function for each pixel point in the real-time image data of the region, and calculating the real-time vegetation coverage rate of the region; S3, based on the standard water sample set and the real-time vegetation coverage rate of the region, generating the water conservation capacity of the region, and determining the water conservation function evaluation result. The calculation result of the water conservation capacity intuitively reflects the reserve capacity and conservation effect of regional water resources, which helps decision makers to quickly grasp the regional water resources situation and formulate scientific and reasonable protection measures, and is applied to multiple fields such as water resources management, ecological protection and disaster prevention.

Description

Technical Field

[0001] The present invention belongs to the technical field of data analysis, and in particular relates to a water conservation function evaluation method based on multi-source data. Background Art

[0002] With the intensification of global climate change and human activities, water resource management and protection have become important global issues. As a key component of ecosystem services, water conservation is of great significance for maintaining ecological balance, ensuring water supply and promoting sustainable development. Water conservation mainly refers to the ecosystem's interception, absorption, storage and slow release of precipitation through mechanisms such as vegetation, soil and hydrological processes, thereby maintaining a dynamic balance between surface water and groundwater.

[0003] Traditional water conservation function assessment methods mostly rely on field observations and statistical data analysis. Although these methods can provide certain scientific basis, they often have limitations such as difficulty in data acquisition, long assessment cycle and high cost. In addition, due to the complexity and spatial heterogeneity of ecosystems, a single data source is often difficult to fully reflect the true status of water conservation functions.

[0004] In recent years, with the rapid development of remote sensing technology, geographic information systems and big data analysis technology, multi-source data fusion has become an important way to solve the above problems. However, although multi-source data provides new opportunities for water conservation function assessment, how to effectively integrate these data, build a scientific assessment model and obtain reliable assessment results is still a challenging issue. Summary of the invention

[0005] In order to solve the above problems, the present invention proposes a water conservation function evaluation method based on multi-source data.

[0006] The technical solution of the present invention is: a water conservation function evaluation method based on multi-source data comprises the following steps:

[0007] S1. Obtain a water sample set of the area, and clean the water sample set to obtain a standard water sample set;

[0008] S2, constructing a feature response function for each pixel in the real-time image data of the region, and calculating the real-time vegetation coverage of the region;

[0009] S3. Based on the regional standard water sample set and real-time vegetation coverage, generate the regional water conservation capacity and determine the water conservation function assessment results.

[0010] Furthermore, in S1, the water sample set of the region includes the flow velocity and soil saturated hydraulic conductivity of the region.

[0011] Furthermore, S2 includes the following sub-steps:

[0012] S21, collecting real-time image data of the area, and constructing a relative color response set for each pixel in the real-time image data;

[0013] S22, generating a characteristic response function for the pixel according to the relative color response set of the pixel;

[0014] S23. Calculate the vegetation coverage of the area based on the characteristic response function of the pixel points.

[0015] The beneficial effect of the above further scheme is: in the present invention, by collecting real-time image data of the area, dynamic monitoring of the regional ecological environment can be achieved, and a relative color response set is constructed for each pixel point, so that the image can be processed in a refined manner. The characteristic response function can extract characteristic information closely related to vegetation coverage, which is used to distinguish vegetation from non-vegetation areas, thereby improving the calculation accuracy of vegetation coverage. By considering the adjacent pixel values ​​in four directions above, to the right, below and to the left of the pixel point, the color relationship between the pixel point and its neighborhood can be fully captured, and the type of land object to which the pixel point belongs can be more accurately identified, especially in the distinction between vegetation and non-vegetation.

[0016] Further, in S21, the relative color response set of the pixel points includes a first relative color matrix, a second relative color matrix, a third relative color matrix and a fourth relative color matrix;

[0017] First relative color matrix The expression is:

[0018] ;

[0019] In the formula, Represents the pixel value of a pixel. Indicates the pixel value of the pixel above the pixel;

[0020] Second relative color matrix The expression is:

[0021] ;

[0022] In the formula, Indicates the pixel value of the pixel to the right of the pixel;

[0023] The third relative color matrix The expression is:

[0024] ;

[0025] In the formula, Indicates the pixel value of the pixel below the pixel;

[0026] Fourth relative color matrix The expression is:

[0027] ;

[0028] In the formula, Indicates the pixel value of the pixel to the left of the pixel.

[0029] Furthermore, in S22, the characteristic response function of the pixel point is The expression is:

[0030] ;

[0031] In the formula, represents the maximum singular value operation, represents the first relative color matrix, represents the second relative color matrix, represents the third relative color matrix, represents the fourth relative color matrix, Represents the pixel value of a pixel. Indicates the pixel value of the pixel above the pixel. Indicates the pixel value of the pixel below the pixel. Indicates taking the maximum value, It means taking a random number in the range of 0 to 1.

[0032] The beneficial effect of the above further scheme is that in the present invention, by combining the maximum singular value of the relative color matrix from different directions (up, right, down, left), the multi-dimensional feature fusion of the pixel neighborhood information is realized, and the complex relationship between the pixel and its surrounding environment is captured more comprehensively, which helps to improve the accuracy and robustness of the feature response function. And a term of taking a random number in the range of 0 to 1 is added to adjust the dynamic range of the feature response function so that it can adapt to image data with different brightness and contrast. At the same time, the introduction of random numbers increases the diversity of the response function, which helps to improve the generalization ability in classification or recognition tasks.

[0033] Further, S23 includes the following sub-steps:

[0034] S231, arranging the pixel values ​​of each row of pixels in the real-time image data according to the positions of the pixels to generate a plurality of row fragment matrices;

[0035] S232, arranging the pixel values ​​of each column of pixels in the real-time image data according to the pixel positions to generate a plurality of column fragment matrices;

[0036] S233, generating a vegetation coverage adaptation value according to a plurality of row fragment matrices and column fragment matrices of the real-time image data;

[0037] S234, will satisfy The pixel point is regarded as the vegetation existence point. represents the vegetation cover adaptation value, Represents the characteristic response function of the pixel, represents the vegetation threshold;

[0038] S235. The ratio between the number of vegetation points and the number of pixels in the real-time image data is taken as the vegetation coverage rate.

[0039] The beneficial effects of the above further scheme are as follows: In the present invention, by dividing the image data into a fragment matrix according to rows and columns, the structured processing of the image data is realized, and the local features in the image are extracted. The vegetation coverage adaptation value is generated by combining the row fragment matrix and the column fragment matrix, and the comprehensive extraction of vegetation features in the image is realized. The vegetation threshold that can be set manually or empirically is introduced to distinguish between vegetation points and non-vegetation points.

[0040] Furthermore, in S233, the vegetation coverage adaptation value The calculation formula is:

[0041] ;

[0042] In the formula, Indicates the real-time image data A column fragment matrix, Indicates the real-time image data row fragment matrices, represents the index, Indicates the number of columns of real-time image data, Indicates the number of rows of real-time image data. Represents matrix transpose.

[0043] Furthermore, S3 includes the following sub-steps:

[0044] S31, generating a water area factor of the region using an inversion algorithm according to the real-time vegetation coverage of the region;

[0045] S32. Calculate the water conservation capacity using the area’s water area factors and determine the water conservation function assessment results.

[0046] The beneficial effect of the above further solution is that in the present invention, the relationship between vegetation coverage and terrain factors is usually nonlinear, and a machine learning model (such as random forest and neural network, etc.) can be used for inverse prediction. The inversion algorithm can consider multiple influencing factors, such as vegetation type, soil conditions and climate conditions, etc.

[0047] Furthermore, in S32, water conservation capacity The calculation formula is:

[0048] ;

[0049] In the formula, represents the flow velocity in the area, represents a constant, represents the saturated hydraulic conductivity of the soil in the region, Indicates the water production of the area, represents the water area factor, Indicates taking the minimum value.

[0050] The beneficial effects of the present invention are as follows: the present invention constructs a feature response function for each pixel point in the real-time image data of a region, which can capture key feature information in the image and help to accurately calculate the vegetation coverage rate; the water conservation capacity of the region is generated based on the standard water area sample set and the real-time vegetation coverage rate, which comprehensively considers the water area characteristics and vegetation coverage, and can more comprehensively evaluate the water conservation function of the region; the calculation result of the water conservation capacity intuitively reflects the reserve capacity and water conservation capacity of regional water resources, which helps decision makers to quickly grasp the regional water resources situation and formulate scientific and reasonable protection measures, which are applied to water resources management, ecological protection, disaster prevention and other fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 This is a flow chart of the water conservation function assessment method based on multi-source data. DETAILED DESCRIPTION

[0052] The embodiments of the present invention will be further described below in conjunction with the accompanying drawings.

[0053] like Figure 1 As shown, the present invention provides a water conservation function evaluation method based on multi-source data, comprising the following steps:

[0054] S1. Obtain a water sample set of the area, and clean the water sample set to obtain a standard water sample set;

[0055] S2, constructing a feature response function for each pixel in the real-time image data of the region, and calculating the real-time vegetation coverage of the region;

[0056] S3. Based on the regional standard water sample set and real-time vegetation coverage, generate the regional water conservation capacity and determine the water conservation function assessment results.

[0057] In the embodiment of the present invention, in S1, the water area sample set of the region includes the flow velocity and soil saturated hydraulic conductivity of the region.

[0058] In this embodiment of the present invention, S2 includes the following sub-steps:

[0059] S21, collecting real-time image data of the area, and constructing a relative color response set for each pixel in the real-time image data;

[0060] S22, generating a characteristic response function for the pixel according to the relative color response set of the pixel;

[0061] S23. Calculate the vegetation coverage of the area based on the characteristic response function of the pixel points.

[0062] In the present invention, by collecting real-time image data of the area, dynamic monitoring of the regional ecological environment can be achieved, and a relative color response set is constructed for each pixel point, so that the image can be processed in a refined manner. The characteristic response function can extract characteristic information closely related to vegetation coverage, which is used to distinguish vegetation from non-vegetation areas, thereby improving the calculation accuracy of vegetation coverage. By considering the adjacent pixel values ​​in four directions above, to the right, below and to the left of the pixel point, the color relationship between the pixel point and its neighborhood can be fully captured, and the type of land object to which the pixel point belongs can be more accurately identified, especially in the distinction between vegetation and non-vegetation.

[0063] In the embodiment of the present invention, in S21, the relative color response set of the pixel points includes a first relative color matrix, a second relative color matrix, a third relative color matrix and a fourth relative color matrix;

[0064] First relative color matrix The expression is:

[0065] ;

[0066] In the formula, Represents the pixel value of a pixel. Indicates the pixel value of the pixel above the pixel;

[0067] Second relative color matrix The expression is:

[0068] ;

[0069] In the formula, Indicates the pixel value of the pixel to the right of the pixel;

[0070] The third relative color matrix The expression is:

[0071] ;

[0072] In the formula, Indicates the pixel value of the pixel below the pixel;

[0073] Fourth relative color matrix The expression is:

[0074] ;

[0075] In the formula, Indicates the pixel value of the pixel to the left of the pixel.

[0076] In the embodiment of the present invention, in S22, the characteristic response function of the pixel point The expression is:

[0077] ;

[0078] In the formula, represents the maximum singular value operation, represents the first relative color matrix, represents the second relative color matrix, represents the third relative color matrix, represents the fourth relative color matrix, Represents the pixel value of a pixel. Indicates the pixel value of the pixel above the pixel. Indicates the pixel value of the pixel below the pixel. Indicates taking the maximum value, It means taking a random number in the range of 0 to 1.

[0079] In the present invention, by combining the maximum singular value of the relative color matrix from different directions (up, right, down, left), multi-dimensional feature fusion of pixel neighborhood information is achieved, which more comprehensively captures the complex relationship between the pixel and its surrounding environment, and helps to improve the accuracy and robustness of the feature response function. And a term that takes a random number in the range of 0 to 1 is added to adjust the dynamic range of the feature response function so that it can adapt to image data with different brightness and contrast. At the same time, the introduction of random numbers increases the diversity of the response function, which helps to improve the generalization ability in classification or recognition tasks.

[0080] In this embodiment of the present invention, S23 includes the following sub-steps:

[0081] S231, arranging the pixel values ​​of each row of pixels in the real-time image data according to the positions of the pixels to generate a plurality of row fragment matrices;

[0082] S232, arranging the pixel values ​​of each column of pixels in the real-time image data according to the pixel positions to generate a plurality of column fragment matrices;

[0083] S233, generating a vegetation coverage adaptation value according to a plurality of row fragment matrices and column fragment matrices of the real-time image data;

[0084] S234, will satisfy The pixel point is regarded as the vegetation existence point. represents the vegetation cover adaptation value, Represents the characteristic response function of the pixel, represents the vegetation threshold;

[0085] S235. The ratio between the number of vegetation points and the number of pixels in the real-time image data is taken as the vegetation coverage rate.

[0086] In the present invention, by dividing the image data into fragment matrices according to rows and columns, the structured processing of the image data is realized, and the local features in the image are extracted. The vegetation coverage adaptation value is generated by combining the row fragment matrix and the column fragment matrix, and the comprehensive extraction of vegetation features in the image is realized. The vegetation threshold that can be set manually or empirically is introduced to distinguish between vegetation points and non-vegetation points.

[0087] In the embodiment of the present invention, in S233, the vegetation coverage adaptation value The calculation formula is:

[0088] ;

[0089] In the formula, Indicates the real-time image data A column fragment matrix, Indicates the real-time image data row fragment matrices, represents the index, Indicates the number of columns of real-time image data, Indicates the number of rows of real-time image data. Represents matrix transpose.

[0090] In this embodiment of the present invention, S3 includes the following sub-steps:

[0091] S31, generating a water area factor of the region using an inversion algorithm according to the real-time vegetation coverage of the region;

[0092] S32. Calculate the water conservation capacity using the area’s water area factors and determine the water conservation function assessment results.

[0093] In the present invention, the relationship between vegetation coverage and terrain factors is usually nonlinear, and machine learning models (such as random forests and neural networks, etc.) can be used for inverse prediction. The inversion algorithm can consider multiple influencing factors, such as vegetation type, soil conditions, and climate conditions.

[0094] In S32, the evaluation result can be determined according to the set water conservation threshold, for example, if the water conservation amount is greater than or equal to the water conservation threshold, the evaluation result is qualified, otherwise it is unqualified. Alternatively, whether the water conservation amount is qualified can also be determined according to actual conditions.

[0095] In the embodiment of the present invention, in S32, the water conservation capacity The calculation formula is:

[0096] ;

[0097] In the formula, represents the flow velocity in the area, represents a constant, represents the saturated hydraulic conductivity of the soil in the region, Indicates the water production of the area, represents the water area factor, Indicates taking the minimum value.

[0098] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific variations and combinations that do not deviate from the essence of the present invention based on the technical revelations disclosed by the present invention, and these variations and combinations are still within the protection scope of the present invention.

Claims

1. A water conservation function assessment method based on multi-source data, characterized in that: The following steps are involved: S1. Obtain a water sample set of the area, and clean the water sample set to obtain a standard water sample set; S2, constructing a feature response function for each pixel in the real-time image data of the region, and calculating the real-time vegetation coverage of the region; S3. Based on the regional standard water sample set and real-time vegetation coverage, generate the regional water conservation capacity and determine the water conservation function assessment results; The S2 comprises the following sub-steps: S21, collecting real-time image data of the area, and constructing a relative color response set for each pixel in the real-time image data; S22, generating a characteristic response function for the pixel according to the relative color response set of the pixel; S23, calculating the vegetation coverage rate of the area according to the characteristic response function of the pixel point; In S21, the relative color response set of the pixel points includes a first relative color matrix, a second relative color matrix, a third relative color matrix and a fourth relative color matrix; The expression of the first relative color matrix X1 is: In the formula, C * Represents the pixel value of the pixel, C up Indicates the pixel value of the pixel above the pixel; The expression of the second relative color matrix X2 is: X2=[C * C right ]; In the formula, C right Indicates the pixel value of the pixel to the right of the pixel; The expression of the third relative color matrix X3 is: In the formula, C down Indicates the pixel value of the pixel below the pixel; The expression of the fourth relative color matrix X4 is: X4=[C left C * ]; In the formula, C left Indicates the pixel value of the pixel to the left of the pixel; In S22, the expression of the characteristic response function F of the pixel point is: In the formula, SVD max (·) represents the maximum singular value operation, X1 represents the first relative color matrix, X2 represents the second relative color matrix, X3 represents the third relative color matrix, X4 represents the fourth relative color matrix, C * Represents the pixel value of the pixel, C up Indicates the pixel value of the pixel above the pixel, C down Indicates the pixel value of the pixel below the pixel, max(·) means taking the maximum value, and rand(0,1) means taking a random number between 0 and 1; The S23 comprises the following sub-steps: S231, arranging the pixel values ​​of each row of pixels in the real-time image data according to the positions of the pixels to generate a plurality of row fragment matrices; S232, arranging the pixel values ​​of each column of pixels in the real-time image data according to the pixel positions to generate a plurality of column fragment matrices; S233, generating a vegetation coverage adaptation value according to a plurality of row fragment matrices and column fragment matrices of the real-time image data; S234, taking the pixel points satisfying SF≥THRESHOLD as vegetation existence points, where S represents the vegetation coverage adaptation value, F represents the characteristic response function of the pixel point, and THRESHOLD represents the vegetation threshold; S235, taking the ratio between the number of vegetation points and the number of pixels in the real-time image data as the vegetation coverage rate; In S233, the calculation formula of the vegetation coverage adaptation value S is: In the formula, x i represents the i-th row fragment matrix of real-time image data, y j represents the j-th column fragment matrix of the real-time image data, e represents the index, I represents the number of rows of the real-time image data, J represents the number of columns of the real-time image data, and T represents the matrix transpose.

2. The water conservation function evaluation method based on multi-source data according to claim 1 is characterized in that: In S1, the water area sample set of the region includes the flow velocity and soil saturated hydraulic conductivity of the region.

3. The water conservation function evaluation method based on multi-source data according to claim 1 is characterized in that: The S3 comprises the following sub-steps: S31, generating a water area factor of the region using an inversion algorithm according to the real-time vegetation coverage of the region; S32. Calculate the water conservation capacity using the area’s water area factors and determine the water conservation function assessment results.

4. The water conservation function evaluation method based on multi-source data according to claim 3 is characterized in that: In S32, the calculation formula of the water conservation capacity W is: Where V represents the flow velocity of the region, c represents a constant, K represents the saturated hydraulic conductivity of the soil in the region, Y represents the water yield in the region, ρ represents the water area factor, and min(·) represents the minimum value.

Images