A dynamic construction method of three-dimensional hydrogeological model

Abstract

The present invention relates to the field of image data processing technology, and in particular to a method for dynamically constructing a three-dimensional hydrogeological model, including: obtaining the possibility that each pixel in the hyperspectral image of the current period is a junction point of different regions at each wavelength according to the grayscale difference of the pixel points along the eight neighborhood directions at the same wavelength from each pixel point in the hyperspectral image of the current period and the degree to which each pixel point is affected by external factors such as weather at each wavelength; obtaining the possibility that each pixel in the hyperspectral image of the current period is a key region point according to the possibility that each pixel in the hyperspectral image of the current period is a junction point of different regions at all different wavelengths; screening out key regions according to the possibility that the pixel points in the suspected key region cluster are key region points; and constructing a three-dimensional hydrogeological model corresponding to the key region. The present invention constructs the accuracy and efficiency of the three-dimensional hydrogeological model.

Description

Technical Field

[0001] The present application relates to the technical field of image data processing, and in particular to a method for dynamically constructing a three-dimensional hydrogeological model. Background Art

[0002] The dynamic construction of three-dimensional hydrogeological models is dynamically affected by factors such as seasonal changes and human activities. In order to make the constructed three-dimensional hydrogeological model more accurately describe and simulate the complexity and dynamics of the groundwater system, it is necessary to conduct real-time monitoring and prediction of the groundwater system so that the model can better reflect the groundwater flow, water level distribution and water quality changes. This is of great significance for emergencies and the formulation of emergency measures. For the entire study area, according to the sensitivity of groundwater level changes and the complexity of hydrogeological conditions, different regions have different degrees of significant changes in a short period of time, so there are different degrees of critical areas in the entire study area. However, when constructing a three-dimensional hydrogeological model, real-time relevant data is generally obtained for the entire study area, which will result in too much data information, requiring more time and resources to process and analyze, resulting in the model being unable to be updated in time, thus affecting the real-time and accuracy of the model. Because the traditional K-means clustering algorithm cannot comprehensively consider the changes in data characteristics when measuring the distance between different sample points (such as the changes in data characteristics caused by weather, environment, etc.), the distance measurement between different sample points is inaccurate, which in turn affects the accuracy of the construction of the three-dimensional hydrogeological model. Summary of the invention

[0003] The present invention provides a method for dynamically constructing a three-dimensional hydrogeological model to solve the problem that the existing traditional K-means clustering algorithm cannot comprehensively consider the changes in data characteristics when measuring the distances between different samples, resulting in inaccurate distance measurement between different sample points, thereby affecting the accuracy of constructing a three-dimensional hydrogeological model.

[0004] A dynamic construction method of a three-dimensional hydrogeological model of the present invention adopts the following technical scheme:

[0005] An embodiment of the present invention provides a method for dynamically constructing a three-dimensional hydrogeological model, the method comprising the following steps:

[0006] Obtain hyperspectral images of the same study area at different times; each pixel in the hyperspectral image corresponds to grayscale values ​​at different wavelengths, and different times include the current period and a reference historical period;

[0007] According to the difference between the grayscale values ​​of each pixel in the hyperspectral image of the current period and the reference historical period at the same wavelength, the degree to which each pixel in the hyperspectral image of the current period is affected by external factors such as weather at each wavelength is obtained;

[0008] According to the grayscale difference of each pixel point along the eight neighborhood directions at the same wavelength in the hyperspectral image of the current period and the degree to which each pixel point is affected by external factors such as weather at each wavelength, the possibility of each pixel point in the hyperspectral image of the current period being the boundary point of different regions at each wavelength is obtained;

[0009] According to the possibility that each pixel point in the hyperspectral image of the current period is the boundary point of different regions at all different wavelengths, the possibility that each pixel point in the hyperspectral image of the current period is the key region point is obtained;

[0010] The difference in the probability that any two pixels in the hyperspectral image of the current period are key area points is taken as the clustering distance of any two pixels, and the hyperspectral image of the current period is clustered to obtain several suspected key area clusters;

[0011] According to the possibility that the pixel points in the suspected key area cluster are key area points, the key areas are screened out; and a three-dimensional hydrogeological model corresponding to the key areas is constructed.

[0012] Preferably, the step of obtaining the degree to which each pixel in the hyperspectral image of the current period is affected by external factors such as weather at each wavelength based on the difference between the grayscale values ​​of each pixel in the hyperspectral image of the current period and the reference historical period at the same wavelength comprises the following specific steps:

[0013] According to the gray value of each pixel point at the same wavelength in the hyperspectral image of the current period and the reference historical period, the gray difference of each pixel point at each wavelength in the hyperspectral image of the current period is obtained;

[0014] According to the difference between the grayscale differences of all pixels in the hyperspectral image of the current period at each wavelength, the degree to which each pixel in the hyperspectral image of the current period is affected by external factors such as weather at each wavelength is obtained.

[0015] Preferably, the grayscale difference of each pixel in the hyperspectral image of the current period at each wavelength is obtained according to the grayscale value of each pixel in the hyperspectral image of the current period and the reference historical period at the same wavelength, and the specific steps include the following:

[0016] The absolute value of the difference in grayscale value of the mth pixel at the i-th wavelength in the hyperspectral image of the current period and the reference historical period is recorded as the grayscale difference of the mth pixel at the i-th wavelength in the hyperspectral image of the current period.

[0017] Preferably, the degree to which each pixel in the hyperspectral image of the current period is affected by external factors such as weather at each wavelength is obtained according to the difference between the grayscale differences of all pixels in the hyperspectral image of the current period at each wavelength, including the specific formula as follows:

[0018]

[0019] In the formula, Q i,m Indicates the degree to which the mth pixel in the hyperspectral image of the current period is affected by external factors such as weather at the i-th wavelength; Δq i,m Represents the grayscale difference of the mth pixel at the i-th wavelength in the hyperspectral image of the current period; It represents the mean grayscale difference of all pixels in the hyperspectral image of the current period at the i-th wavelength; norm() is the normalization function; M represents the number of pixels in the hyperspectral image of the current period; || is the absolute value function.

[0020] Preferably, the method of obtaining the possibility that each pixel in the hyperspectral image of the current period is a boundary point of different regions at each wavelength based on the grayscale difference of the pixel at the same wavelength along the eight neighborhood directions starting from each pixel in the hyperspectral image of the current period and the degree to which each pixel is affected by external factors such as weather at each wavelength comprises the following specific steps:

[0021] In the hyperspectral image of the current period, the local neighborhood of each pixel is constructed;

[0022] According to the grayscale difference of pixels at the same wavelength in the local neighborhood of each pixel in the hyperspectral image of the current period, along the eight neighborhood directions starting from each pixel, and the degree to which each pixel is affected by external factors such as weather at each wavelength, the possibility of each pixel in the hyperspectral image of the current period being the boundary point of different regions at each wavelength is obtained.

[0023] Preferably, the local neighborhood of each pixel point in the hyperspectral image of the current period is constructed, including the following specific steps:

[0024] In the hyperspectral image of the current period, a circle with each pixel as the center and a radius of r is recorded as the local neighborhood of each pixel; r is the preset radius.

[0025] Preferably, the possibility that each pixel in the hyperspectral image of the current period is the boundary point of different regions at each wavelength is obtained based on the grayscale difference of the pixel at the same wavelength along the eight neighborhood directions starting from each pixel in the local neighborhood of each pixel in the hyperspectral image of the current period and the degree to which each pixel is affected by external factors such as weather at each wavelength, including the specific formula as follows:

[0026]

[0027] Where W i,j It indicates the possibility that the jth pixel in the hyperspectral image of the current period is the boundary point of different regions at the i-th wavelength; Q i,m Indicates the degree to which the mth pixel in the hyperspectral image of the current period is affected by external factors such as weather at the i-th wavelength; Δq i,j,l,b Indicates the grayscale difference of the bth pixel at the i-th wavelength along the l-th direction in the eight-neighborhood direction starting from the j-th pixel in the hyperspectral image of the current period; Δq i,j,l,max Indicates the maximum value of the grayscale difference of all pixels at the i-th wavelength in the local neighborhood of the j-th pixel in the hyperspectral image of the current period, starting from the j-th pixel along the l-th direction of the eight neighborhood directions; B i,j,l It represents the number of all pixels at the i-th wavelength in the local neighborhood of the j-th pixel in the hyperspectral image of the current period, starting from the j-th pixel along the l-th direction of the eight neighborhood directions; b belongs to the interval [1,B i,j,l ] within all Δq i,j,l,max -Δq i,j,l,b The standard deviation of When l is in the interval [1,8], The standard deviation of .

[0028] Preferably, the possibility that each pixel point in the hyperspectral image of the current period is a point in the key area is obtained according to the possibility that each pixel point in the hyperspectral image of the current period is a boundary point of different areas at all different wavelengths, and the specific formula included is as follows:

[0029]

[0030] In the formula, E j represents the possibility that the j-th pixel point in the hyperspectral image of the current period is a key area point; I represents the number of wavelengths corresponding to the j-th pixel point in the hyperspectral image of the current period; W i,j It indicates the possibility that the j-th pixel in the hyperspectral image of the current period is the boundary point of different regions at the i-th wavelength.

[0031] Preferably, the specific steps of obtaining the cluster distance between any two pixels are as follows:

[0032] The absolute value of the difference between the possibilities of any two pixels in the hyperspectral image of the current period being key area points is recorded as the clustering distance of the any two pixels.

[0033] Preferably, the method of screening out the key areas according to the possibility that the pixel points in the suspected key area cluster are key area points comprises the following specific steps:

[0034] The mean of the likelihood that all pixels in each suspected critical region cluster are critical region points is calculated, and the suspected critical region cluster corresponding to the largest mean of the likelihood that all pixels are critical region points is recorded as the critical region.

[0035] The beneficial effects of the technical solution of the present invention are as follows: according to the grayscale difference of the pixel points at the same wavelength along the eight neighborhood directions from each pixel point in the hyperspectral image of the current period and the degree to which each pixel point is affected by external factors such as weather at each wavelength, the possibility of each pixel point in the hyperspectral image of the current period being the junction point of different regions at each wavelength is obtained; the degree to which each pixel point in the hyperspectral image of the current period is affected by external factors such as weather at each wavelength is comprehensively analyzed, and the data feature changes caused by the influence of weather, environment, etc. are considered, so as to improve the accuracy of key area screening, thereby improving the accuracy of constructing a three-dimensional hydrogeological model. According to the possibility that each pixel point in the hyperspectral image of the current period is the junction point of different regions at all different wavelengths, the possibility that each pixel point in the hyperspectral image of the current period is a key area point is obtained; the difference in the possibility that any two pixel points in the hyperspectral image of the current period are key area points is taken as the clustering distance of any two pixel points, and the hyperspectral image of the current period is clustered to obtain a number of suspected key area clusters; according to the possibility that the pixel points in the suspected key area clusters are key area points, the key areas are screened out; and a three-dimensional hydrogeological model corresponding to the key areas is constructed. By screening out key areas and collecting key information in key areas, we can reduce the acquisition of information in non-key areas, thereby reducing the amount of data for dynamically constructing three-dimensional hydrogeological models, making the dynamic construction of three-dimensional hydrogeological models more efficient and adaptable. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0037] Figure 1 The present invention is a flowchart of the steps of a method for dynamically constructing a three-dimensional hydrogeological model. DETAILED DESCRIPTION

[0038] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the dynamic construction method of a three-dimensional hydrogeological model proposed by the present invention, its specific implementation method, structure, characteristics and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0039] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0040] The specific scheme of the dynamic construction method of a three-dimensional hydrogeological model provided by the present invention is described in detail below with reference to the accompanying drawings.

[0041] See also Figure 1 , which shows a flowchart of a method for dynamically constructing a three-dimensional hydrogeological model provided by an embodiment of the present invention, the method comprising the following steps:

[0042] Step S001: Acquire hyperspectral images of the same study area at several different periods; wherein each pixel in the hyperspectral image corresponds to grayscale values ​​at several different wavelengths, and the different periods include a current period and a reference historical period.

[0043] Acquire hyperspectral images of the same study area at several different times; wherein each pixel point in the hyperspectral image corresponds to a reflection intensity at several different wavelengths. In this embodiment, the reflection intensity is called grayscale value, that is, each pixel point corresponds to a grayscale value at several different wavelengths.

[0044] It should be noted that the reference historical period selected in this embodiment is one week ago, and this is used as an example for description.

[0045] The study area refers to the area where a three-dimensional hydrogeological model needs to be constructed; hyperspectral images of the same study area at different times are obtained through satellite remote sensing.

[0046] It should be noted that each pixel in a hyperspectral image can reflect reflection intensity or reflectivity. Each pixel in a hyperspectral image actually corresponds to a small area on the surface of the earth. Each band in the area has a corresponding radiation intensity value; these values ​​reflect the spectral characteristics of the surface in different bands. These reflection intensity values ​​are usually expressed in digital form, which can be floating point numbers between 0 and 1, or integers (grayscale values) between 0 and 255. They are collectively referred to as grayscale values ​​hereafter, and the reflection intensity is reflected in this form.

[0047] It should be noted that when analyzing the hydrogeological conditions of the current study area, hyperspectral images can identify different types of land objects such as water bodies, soil, vegetation, etc. due to their high spatial resolution and multi-spectral information, providing a richer data basis for hydrogeological analysis, better understanding the surface environment, and discovering hidden geological and hydrological features. Therefore, this embodiment obtains a hyperspectral image of the current study area, thereby providing a basis for subsequent analysis of key areas in the current study area.

[0048] Step S002: According to the difference between the grayscale values ​​of each pixel in the hyperspectral image of the current period and the reference historical period at the same wavelength, the degree to which each pixel in the hyperspectral image of the current period is affected by external factors such as weather at each wavelength is obtained.

[0049] It should be noted that in the current study area, the hydrogeological characteristics are reflected by hyperspectral images, and there are differences in the geological stability, soil hydrological conditions and other conditions in different areas of the entire study area. In this way, under the influence of rainfall or human activities, the degree of change of hydrogeological characteristics in different areas is different, which will lead to different degrees of differences in the hyperspectral images collected in the current study area at different times. For example, the geological structure in a part of the current study area is relatively stable, the underground rock layer is solid and has low permeability, so that the groundwater level changes slowly; or its soil type and moisture content are suitable, which is conducive to water infiltration and storage. Therefore, in the above situation, even if there is the influence of rainfall or human activities, the change of its hydrogeological characteristics is relatively slow compared with the area that does not have the above characteristics (the above characteristics are just examples, and there are other factors that help maintain hydrogeological characteristics). For the hyperspectral images collected at different times, based on the analysis of the hydrological characteristics of each pixel at different wavelengths, the possibility of each pixel being a key area is determined, and then the distance measurement method between different pixels as sample points is determined when clustering.

[0050] What needs to be explained is that for the hyperspectral images of the study area at different periods, the hyperspectral images of different areas in the study area will change due to factors such as rainfall or human activities. The magnitude of the change is affected by factors such as the geological stability of the region itself and soil hydrological conditions. Therefore, by comparing the change amplitude of the hyperspectral images at the same wavelength at different periods, it can be reflected to what extent the study area is affected by external factors such as weather at different wavelengths, so as to determine the representation degree of key areas in the study area at different wavelengths.

[0051] First, the definition of key areas in the study area is explained. Due to the large scope of the study area, the hydrogeological characteristics of different areas at different times are different. For areas with rapid changes in hydrogeological characteristics, they need to be updated and measured in a timely manner, so they are called key areas.

[0052] The absolute value of the difference in grayscale value of the mth pixel point at the i-th wavelength in the hyperspectral image of the current period and the reference historical period is recorded as the grayscale difference of the mth pixel point at the i-th wavelength in the hyperspectral image of the current period.

[0053] The calculation method for the degree to which the mth pixel in the hyperspectral image of the current period is affected by external factors such as weather at the i-th wavelength is as follows:

[0054]

[0055] In the formula, Q i,m Indicates the degree to which the mth pixel in the hyperspectral image of the current period is affected by external factors such as weather at the i-th wavelength; Δq i,m Represents the grayscale difference of the mth pixel at the i-th wavelength in the hyperspectral image of the current period; It represents the mean grayscale difference of all pixels in the hyperspectral image of the current period at the i-th wavelength; norm() is the normalization function; M represents the number of pixels in the hyperspectral image of the current period; || is the absolute value function.

[0056] It should be noted that Δq i,m It represents the grayscale difference of the mth pixel in the hyperspectral image of the current period at the i-th wavelength. The larger the value, the more obvious the difference in the hydrogeological characteristics of the pixel in the hyperspectral image of the current period at the current wavelength, that is, the greater the degree of influence by external factors such as weather; however If only the mean is used, the grayscale difference corresponding to some pixels may be too large or too small. It may also be larger, but this can only mean that some areas in the current study area are too greatly affected by factors such as weather. When mapped to the entire study area, it appears as if the current study area is greatly affected by external factors such as weather. Therefore, in this embodiment, The smaller the value, the closer the grayscale difference corresponding to each pixel in the current study area is to the average value, that is, the influence of weather and other factors on each area in the current study area is more balanced (indicating that the overall study area is greatly affected, so the corresponding overall study area is more real and accurate in terms of the influence of weather factors). Use this as the benchmark value The smaller the adjustment coefficient, the greater the degree of adjustment to the benchmark value.

[0057] At this point, the degree to which each pixel in the hyperspectral image of the current period is affected by external factors such as weather at each wavelength is obtained.

[0058] Step S003: Based on the grayscale differences of each pixel point along the eight neighborhood directions at the same wavelength in the hyperspectral image of the current period and the degree to which each pixel point is affected by external factors such as weather at each wavelength, the possibility of each pixel point in the hyperspectral image of the current period being the boundary point of different areas at each wavelength is obtained.

[0059] It should be noted that the above analysis shows the degree of influence of external factors such as weather at different periods at each wavelength. This is determined based on the overall research area corresponding to each wavelength. Here, the degree of influence of each pixel point at each wavelength can be analyzed and judged to determine the possibility that each pixel point at each wavelength is the boundary point of different regions (regions with different geological stability and soil hydrological conditions).

[0060] It should be further explained that in the entire study area, the geological stability and soil hydrological conditions of different regions are different. For pixels in the same area (areas with the same or similar geological stability and soil hydrological conditions), the grayscale difference corresponding to the same wavelength at different times is similar to that of other pixels in the neighborhood, while the grayscale difference corresponding to pixels at the junction of different regions and other pixels in the neighborhood may be significantly different. These locations may be key nodes of the hydrogeological process and have important geological, hydrological and ecological significance. Therefore, by analyzing this characteristic, we can better discover the criticality of the hydrogeological characteristic information presented by each pixel in the entire study area based on the current wavelength.

[0061] In the hyperspectral image of the current period, a circle with each pixel as the center and a radius of r is recorded as the local neighborhood of each pixel; r is the preset radius, r=13, and this is used as an example for description.

[0062] The calculation method for the possibility that the jth pixel in the hyperspectral image of the current period is the boundary point of different regions at the i-th wavelength is as follows:

[0063]

[0064] Where W i,j It indicates the possibility that the jth pixel in the hyperspectral image of the current period is the boundary point of different regions at the i-th wavelength; Q i,m Indicates the degree to which the mth pixel in the hyperspectral image of the current period is affected by external factors such as weather at the i-th wavelength; Δq i,j,l,b Indicates the grayscale difference of the bth pixel at the i-th wavelength along the l-th direction in the eight-neighborhood direction starting from the j-th pixel in the hyperspectral image of the current period; Δq i,j,l,max Indicates the maximum value of the grayscale difference of all pixels at the i-th wavelength in the local neighborhood of the j-th pixel in the hyperspectral image of the current period, starting from the j-th pixel along the l-th direction of the eight neighborhood directions; B i,j,l It represents the number of all pixels at the i-th wavelength in the local neighborhood of the j-th pixel in the hyperspectral image of the current period, starting from the j-th pixel along the l-th direction of the eight neighborhood directions; b belongs to the interval [1,B i,j,l ] within all Δq i,j,l,max -Δq i,j,l,b The standard deviation of When l is in the interval [1,8], all The standard deviation of .

[0065] When there is no pixel in a direction of the eight neighborhood directions of a pixel, this direction is not analyzed in the above formula.

[0066] The eight-neighborhood direction is a well-known technology, and the specific method will not be introduced here.

[0067] It should be noted that different regions represent areas of different geological stability and soil and hydrological conditions.

[0068] What needs to be explained is that Q i As the weight, Q i The larger the value, the greater the degree of influence of external factors such as weather in the study area corresponding to the current pixel at this wavelength, which lays the overall tone of the wavelength at which the current pixel is located. Therefore, Q i The larger the value is, the greater the impact on the current pixel being the boundary point of different areas.

[0069] It should be noted that, with Δq i,j,l,maxIt represents the maximum value of the grayscale difference of all pixels at the i-th wavelength in the local neighborhood of the j-th pixel in the hyperspectral image of the current period, starting from the j-th pixel along the l-th direction of the eight neighborhood directions; By comparing the proximity of the grayscale difference of other pixels in the same direction to the grayscale difference of the maximum pixel, To determine the local neighborhood of the pixel point, the consistency of the changes in hydrogeological characteristics in the eight neighborhood directions of the pixel point. The larger the value, the greater the difference in the changes in hydrogeological characteristics in each direction. The greater the possibility that the j-th pixel point in the hyperspectral image of the current period is the boundary point of different regions at the i-th wavelength.

[0070] At this point, the possibility that each pixel in the hyperspectral image of the current period is the boundary point of different regions at each wavelength is obtained.

[0071] Step S004: According to the possibility that each pixel point in the hyperspectral image of the current period is a boundary point of different regions at all different wavelengths, the possibility that each pixel point in the hyperspectral image of the current period is a key region point is obtained.

[0072] It should be noted that the above process analysis determines the possibility of intersection points of different areas (areas with different geological stability and soil hydrological conditions) of each pixel point at different wavelengths, and summarizes them to comprehensively determine the possibility of each pixel point being a key area point based on different wavelengths.

[0073] The calculation method for the probability that the jth pixel point in the hyperspectral image of the current period is a key area point is as follows:

[0074]

[0075] In the formula, E j represents the possibility that the j-th pixel point in the hyperspectral image of the current period is a key area point; I represents the number of wavelengths corresponding to the j-th pixel point in the hyperspectral image of the current period; W i,j It indicates the possibility that the j-th pixel in the hyperspectral image of the current period is the boundary point of different regions at the i-th wavelength.

[0076] What needs to be explained is that the possibility that the j-th pixel point in the hyperspectral image of the current period is a key area point is comprehensively determined by accumulating the possibility that the j-th pixel point in the hyperspectral image of the current period is a boundary point of different areas at different wavelengths.

[0077] At this point, the possibility that each pixel point in the hyperspectral image of the current period is a key area point is obtained.

[0078] Step S005: Taking the difference in the possibility that any two pixels in the hyperspectral image of the current period are key area points as the clustering distance of any two pixels, clustering operation is performed on the hyperspectral image of the current period to obtain several suspected key area clusters; based on the possibility that the pixels in the suspected key area clusters are key area points, the key areas are screened out; and a three-dimensional hydrogeological model corresponding to the key areas is constructed.

[0079] Each pixel point in the hyperspectral image of the current period is used as a sample point for clustering, and the distance measurement method in the K-means clustering algorithm is used; the distance measurement method between different sample points is the absolute value of the difference between the possibilities of any two pixel points in the hyperspectral image of the current period being key area points.

[0080] According to the distance measurement method between different sample points mentioned above, the preset number of clusters is 8, which is used as an example for description; therefore, the absolute value of the difference between the possibility of any two pixel points in the hyperspectral image of the current period being key area points is taken as the clustering distance of any two pixel points, and the K-means clustering algorithm is used to cluster the hyperspectral image of the current period to obtain several suspected key area clusters.

[0081] The mean of the likelihood that all pixels in each suspected critical region cluster are critical region points is calculated, and the suspected critical region cluster corresponding to the largest mean of the likelihood that all pixels are critical region points is recorded as the critical region.

[0082] Based on the data of groundwater level, hydrogeological conditions and terrain undulation in key areas, a three-dimensional hydrogeological model is dynamically constructed using numerical simulation methods.

[0083] What needs to be explained is that, for key information collection areas in key areas, the information update of key areas can be set to once every two days, and for other areas it can be set to once a week. By using multi-source data such as geology, hydrology, and topography, a three-dimensional hydrogeological model is dynamically constructed using numerical simulation methods. By simulating groundwater flow, hydrological processes, etc., future water resource changes and groundwater dynamics can be predicted.

[0084] Among them, the K-means clustering algorithm and the numerical simulation method are well-known technologies, and the specific methods are not introduced here.

[0085] At this point, this embodiment is completed.

[0086] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for dynamically constructing a three-dimensional hydrogeological model, characterized in that: The method comprises the following steps: Obtain hyperspectral images of the same study area at different times; each pixel in the hyperspectral image corresponds to grayscale values ​​at different wavelengths, and different times include the current period and a reference historical period; According to the difference between the grayscale values ​​of each pixel in the hyperspectral images of the current period and the reference historical period at the same wavelength, the degree to which each pixel in the hyperspectral images of the current period is affected by external factors such as weather at each wavelength is obtained; According to the grayscale difference of each pixel point along the eight neighborhood directions at the same wavelength in the hyperspectral image of the current period and the degree to which each pixel point is affected by external factors such as weather at each wavelength, the possibility of each pixel point in the hyperspectral image of the current period being the boundary point of different regions at each wavelength is obtained; According to the possibility that each pixel point in the hyperspectral image of the current period is the boundary point of different regions at all different wavelengths, the possibility that each pixel point in the hyperspectral image of the current period is the key region point is obtained; The difference in the possibility that any two pixels in the hyperspectral image of the current period are key area points is taken as the clustering distance of any two pixels, and the hyperspectral image of the current period is clustered to obtain several suspected key area clusters; According to the possibility that the pixel points in the suspected key area cluster are key area points, the key areas are screened out; and a three-dimensional hydrogeological model corresponding to the key areas is constructed; The method of obtaining the degree to which each pixel in the hyperspectral image of the current period is affected by external factors such as weather at each wavelength based on the difference between the grayscale values ​​of each pixel in the hyperspectral image of the current period and the reference historical period at the same wavelength includes the following specific steps: According to the gray value of each pixel point at the same wavelength in the hyperspectral image of the current period and the reference historical period, the gray difference of each pixel point at each wavelength in the hyperspectral image of the current period is obtained; According to the difference between the grayscale differences of all pixels in the hyperspectral image of the current period at each wavelength, the degree to which each pixel in the hyperspectral image of the current period is affected by external factors such as weather at each wavelength is obtained, including the specific formula as follows: ； In the formula, Indicates the number of hyperspectral images in the current period. The pixel at The degree to which the wavelength is affected by external factors such as weather; Indicates the number of hyperspectral images in the current period. The pixel at Grayscale difference under wavelength; Indicates that all pixels in the hyperspectral image of the current period are The mean value of grayscale difference under wavelengths; is the normalization function; Indicates the number of pixels in the hyperspectral image of the current period; is the absolute value function; The method of obtaining the possibility that each pixel in the hyperspectral image of the current period is a boundary point of different regions at each wavelength based on the grayscale difference of each pixel along the eight neighborhood directions at the same wavelength and the degree to which each pixel is affected by external factors such as weather at each wavelength in the hyperspectral image of the current period, includes the following specific steps: In the hyperspectral image of the current period, the local neighborhood of each pixel is constructed; According to the grayscale difference of each pixel point in the local neighborhood of each pixel point in the hyperspectral image of the current period along the eight neighborhood directions at the same wavelength and the degree to which each pixel point is affected by external factors such as weather at each wavelength, the possibility of each pixel point in the hyperspectral image of the current period being the boundary point of different regions at each wavelength is obtained, including the specific formula as follows: ； In the formula, Indicates the number of hyperspectral images in the current period. The pixel at The possibility of being the boundary point of different regions at different wavelengths; Indicates the hyperspectral image of the current period, from The pixel starts from the first pixel in the eight-neighborhood direction. Direction The pixel at Grayscale difference under wavelength; Indicates the number of hyperspectral images in the current period. In the local neighborhood of pixels, from the The pixel starts from the first pixel in the eight-neighborhood direction. All pixels in the direction The maximum value of the grayscale difference under wavelengths; Indicates the first In the local neighborhood of pixels, from the The pixel starts from the first pixel in the eight-neighborhood direction. All pixels in the direction The number of wavelengths; for Belong to the interval All internal time The standard deviation of for Belong to the interval All internal time The standard deviation of The method of screening out key areas according to the possibility that the pixel points in the suspected key area cluster are key area points includes the following specific steps: The mean of the likelihood that all pixels in each suspected critical region cluster are critical region points is calculated, and the suspected critical region cluster corresponding to the largest mean of the likelihood that all pixels are critical region points is recorded as the critical region.

2. A method for dynamically constructing a three-dimensional hydrogeological model according to claim 1, characterized in that: The step of obtaining the grayscale difference of each pixel in the hyperspectral image of the current period at each wavelength according to the grayscale value of each pixel in the hyperspectral image of the current period and the reference historical period at the same wavelength includes the following specific steps: The hyperspectral images of the current period and the reference historical period are The pixel at The absolute value of the difference in grayscale values ​​at wavelengths is recorded as the The pixel at The grayscale difference at different wavelengths.

3. A method for dynamically constructing a three-dimensional hydrogeological model according to claim 1, characterized in that: The specific steps of constructing the local neighborhood of each pixel in the hyperspectral image of the current period are as follows: In the hyperspectral image of the current period, a circle with each pixel as the center and a radius of r is recorded as the local neighborhood of each pixel; r is the preset radius.

4. A method for dynamically constructing a three-dimensional hydrogeological model according to claim 1, characterized in that: The possibility that each pixel point in the hyperspectral image of the current period is a boundary point of different regions at all different wavelengths is obtained according to the possibility that each pixel point in the hyperspectral image of the current period is a key region point, including the specific formula as follows: ； In the formula, Indicates the number of hyperspectral images in the current period. The possibility that a pixel point is a key area point; Indicates the number of hyperspectral images in the current period. The number of wavelengths corresponding to each pixel; Indicates the number of hyperspectral images in the current period. The pixel at The possibility of the intersection point of different regions at different wavelengths.

5. A method for dynamically constructing a three-dimensional hydrogeological model according to claim 1, characterized in that: The specific steps for obtaining the cluster distance between any two pixels are as follows: The absolute value of the difference between the possibilities of any two pixels in the hyperspectral image of the current period being key area points is recorded as the clustering distance of the any two pixels.

Images