CN118113788B - GIS data processing method and system based on multi-source data fusion - Google Patents

Abstract

The present invention relates to the field of geographic information system data processing technology, and in particular to a GIS data processing method and system based on multi-source data fusion. The method obtains the similarity of spatial attribute information between unsampled positions and collection areas, and then determines the spatial correlation factor. The isolation factor of each collection area is analyzed, and the bias factor is obtained according to the relative distance between each collection area and the unsampled position. The interpolation weight of each collection area relative to the unsampled position is obtained by combining the spatial correlation factor, the isolation factor and the bias factor. The unsampled position is interpolated using the interpolation weight with strong reference, and the corresponding elevation data is obtained, and then the multi-source data is fused according to the interpolated GIS data. The present invention can obtain accurate interpolation results of the unsampled position by obtaining the interpolation weight with strong reference, and then realize the fusion display of multi-source data according to the complete GIS data.

Description

GIS data processing method and system based on multi-source data fusion

Technical Field

The present invention relates to the technical field of geographic information system data processing, and in particular to a GIS data processing method and system based on multi-source data fusion.

Background Art

Geographic Information System (GIS) is a technology used to capture, store, retrieve and display geographic information data. In the process of building a geographic information system for urban environments, the collected GIS data contains spatial attribute information such as elevation data and digital line data. When performing multi-dimensional data fusion display, the data in each dimension will be visualized, and then the visualized data of each dimension will be fused and displayed on a display platform. However, in the process of collecting elevation data in the city, due to equipment problems and urban policy regulations, there is no elevation data in some locations. These unsampled locations need to be interpolated, and after the data is obtained, it is visualized to achieve multi-source data fusion display.

In the prior art, when interpolating elevation data in GIS data, the distance between the unsampled location and the collection area where data already exists is considered to construct interpolation weights for interpolation. However, in the actual urban environment, due to the distribution of buildings, the urban terrain has complex characteristics and there are different types of buildings. If the distance is directly used to construct the interpolation weight, the final interpolation result will have poor reference value and cannot form an accurate visual fusion display result.

Summary of the invention

In order to solve the technical problem that the reasonable interpolation weight cannot be obtained in the prior art, resulting in poor reference of the interpolation result, which in turn affects the multi-source data fusion display of GIS data, the purpose of the present invention is to provide a GIS data processing method and system based on multi-source data fusion. The technical scheme adopted is as follows:

The present invention proposes a GIS data processing method based on multi-source data fusion, the method comprising:

Acquire GIS data within the city; the GIS data includes spatial attribute information of the entire city and elevation data within a preset collection area; the city also includes unsampled locations;

According to the similarity of spatial attribute information between the preset neighborhood range of the unsampled position and the collection area, a spatial correlation factor between each collection area and the unsampled position is obtained; according to the position distribution between the collection areas and the similarity of spatial attribute information, an isolation factor of each collection area is obtained; the relative distance between the unsampled position and each of the collection areas is obtained, and according to the difference in relative distance between each collection area and other collection areas, and the position distribution between each collection area and the unsampled position, a bias factor between each collection area and the unsampled position is obtained;

Obtaining an interpolation weight of each acquisition area relative to the unsampled position according to the isolation factor, the deflection factor and the spatial correlation factor; interpolating the data of the unsampled position according to the elevation data in the acquisition area based on the interpolation weight to obtain the elevation data at the unsampled position;

Multi-source data fusion display is performed based on the GIS data after interpolation processing.

Furthermore, the size of the smallest acquisition area is selected as the size of the neighborhood range, and the neighborhood range is constructed with the unsampled position as the center.

Furthermore, the spatial attribute information includes digital line drawing data and building facility type text vectors.

Furthermore, the method for obtaining the spatial correlation factor includes:

Obtain a first cosine similarity of the building facility type text vector between the preset neighborhood range of the unsampled position and the collection area, and a second cosine similarity of the digital line stroke data; and use the sum of the first cosine similarity and the second cosine similarity as the spatial correlation factor.

Furthermore, the formula for obtaining the isolation factor includes:

Where g _n is the isolation factor of the nth collection area, N is the number of collection areas, j is the serial number of all collection areas except the nth collection area, e is a natural constant, l _nj is the distance between the nth collection area and the jth other collection area, and a _nj is the spatial correlation factor between the nth collection area and the jth other collection area.

Furthermore, the formula for obtaining the bias factor includes:

Among them, _Gn is the bias factor of the nth acquisition area; softmax() is the normalized exponential function; _sn is the area of the nth acquisition area; N is the number of acquisition areas; j is the serial number of all acquisition areas except the nth acquisition area; _ln is the relative distance between the nth acquisition area and the unsampled position; _lj is the relative distance between the jth other acquisition area and the unsampled position; with the unsampled position as the vertex, rays are drawn to the center points of the nth acquisition area and the jth other acquisition area, and the angle of the ray is _θnj ; sin() is the sine function; _sj is the area of the jth other acquisition area.

Furthermore, the method for obtaining the interpolation weight of each acquisition area relative to the unsampled position according to the isolation factor, the bias factor and the spatial correlation factor includes:

The isolation factor, the bias factor and the spatial correlation factor are multiplied and then normalized to obtain the interpolation weight of each acquisition area relative to the unsampled position.

Furthermore, after obtaining the interpolation weight, the following steps are also included:

According to the relative distance between each unsampled position and each of the acquisition areas, the distance weight of each acquisition area relative to the unsampled position is obtained, and the interpolation weight is updated by taking the average value of the distance weight and the interpolation weight as a new interpolation weight.

Furthermore, the data of the unsampled position is interpolated using an inverse distance weighted interpolation method to obtain elevation data at the unsampled position.

The present invention also proposes a GIS data processing system based on multi-source data fusion, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any one of the steps of the GIS data processing method based on multi-source data fusion.

The present invention has the following beneficial effects:

The embodiment of the present invention is based on the characteristics of GIS data. Considering that there is rich spatial attribute information in GIS data, and the spatial attribute information is not easy to be missing or collected, the spatial attribute information similarity is used to analyze when considering the correlation between the unsampled position and the collection area to obtain the spatial correlation factor. Further considering that when analyzing the correlation between the collection area and the unsampled position, it is not only necessary to consider the distance, but also to consider the location distribution in the urban area. If the unsampled position and a certain collection area are both isolated areas in the urban area, and the unsampled position is more inclined to the collection area, it means that the data of the collection area is of great reference to the unsampled position. Therefore, the isolation factor of the collection area and the bias factor between the collection area and the unsampled position are analyzed, and then the interpolation weight with strong reference can be obtained by combining the spatial correlation factor. Then, the data of the unsampled position can be interpolated to obtain accurate elevation data, and the unsampled or missing data can be filled by interpolation. Visual processing is performed in the subsequent process, and effective multi-source data fusion display can be completed by combining data of other dimensions.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are 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.

FIG1 is a flow chart of a GIS data processing method based on multi-source data fusion provided by an embodiment of the present invention.

DETAILED DESCRIPTION

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 GIS data processing method and system based on multi-source data fusion proposed by the present invention, its specific implementation method, structure, features and effects, in conjunction 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.

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.

The following is a detailed description of a specific solution of a GIS data processing method and system based on multi-source data fusion provided by the present invention in conjunction with the accompanying drawings.

GIS data processing method embodiment based on multi-source data fusion:

Please refer to FIG. 1 , which shows a flow chart of a GIS data processing method based on multi-source data fusion provided by an embodiment of the present invention. The method includes:

Step S1: Obtain GIS data within the city.

Based on the geographic information system, GIS data within the city can be obtained, where GIS data contains spatial attribute information of the entire city and elevation data within the preset collection area. It should be noted that the reason for the preset collection area is that when collecting elevation areas within the city, some important areas such as government agencies cannot be collected. Therefore, it is necessary to pre-set the collection area for elevation data collection. In each collection area, there is clear longitude and latitude data, and each longitude and latitude data corresponds to a sampling position, that is, a collection area is composed of multiple sampling positions, and then each sampling position is processed using an elevation data collection device such as a drone to obtain the corresponding elevation data, that is, the elevation data in a collection area is a data set composed of the elevation data of the corresponding multiple sampling positions.

In one embodiment of the present invention, the interval between sampling locations within the collection area may be set to 1 meter.

When collecting elevation data for transmission, data at a specific sampling location may be lost due to equipment problems. In the final result, it can be considered that no sampling has been performed, that is, the location becomes an unsampled location. Also considered as unsampled locations are locations where no sampling operations have been performed. Such locations need to be interpolated to fill in the data, so that each location can be guaranteed to have spatial attribute information and elevation data in the subsequent multi-source fusion display.

Preferably, in one embodiment of the present invention, the spatial attribute information includes digital line drawing data and building facility type text vectors. It should be noted that the digital line drawing data is the data in the digital line drawing map in the geographic information system, which is a vector data set that stores the basic geographic elements in layers on the existing topographic map. The digital line drawing data includes both spatial information and attribute information, such as vector information such as rivers and roads, and also includes attribute information such as road grade, speed limit, width, etc. The building type text can be obtained based on a known city map, such as text descriptions of residential buildings, office buildings, etc., and then the corresponding text is output as a vector based on the language model. It should be noted that the methods for obtaining digital line drawing data and building facility type text vectors are technical means well known to those skilled in the art, and will not be elaborated here.

Step S2: According to the similarity of spatial attribute information between the preset neighborhood range of the unsampled position and the collection area, obtain the spatial correlation factor between each collection area and the unsampled position; according to the position distribution between the collection areas and the similarity of spatial attribute information, obtain the isolation factor of each collection area; obtain the relative distance between the unsampled position and each collection area, and according to the difference in relative distance between each collection area and other collection areas, and the position distribution between each collection area and the unsampled position, obtain the bias factor between each collection area and the unsampled position.

Considering that the prior art only considers the distance between the acquisition area and the unsampled position to construct the interpolation weight, which is not very referenceable, the final interpolation result has poor accuracy. Therefore, the embodiment of the present invention first analyzes according to the spatial attribute information. Because the unsampled position is a specific position point, the spatial attribute information it contains is less than the spatial attribute information in the acquisition area. In order to accurately characterize the correlation between the unsampled position and the acquisition area, the preset neighborhood range is constructed with the unsampled position as the center, that is, the spatial attribute information similarity between the neighborhood range and the acquisition area is analyzed to obtain the spatial correlation factor between each acquisition area and the corresponding unsampled position. The larger the spatial correlation factor, the closer the spatial attribute information between the acquisition area and the unsampled position is. In the subsequent interpolation, the interpolation weight of the corresponding acquisition area needs to be increased so that an accurate interpolation result can be obtained.

Preferably, in one embodiment of the present invention, it is considered that the neighborhood range of the unsampled position should not be set too large. If it is set too large, the information in the range will be less relevant to the unsampled position, which will reduce the reference value; if it is set too small, it will also lead to less information and reduce the reference value. Therefore, taking the acquisition area as a reference, the size of the smallest acquisition area is selected as the size of the neighborhood range, and the neighborhood range is constructed with the unsampled position as the center.

It should be noted that if the collection area is a regular shape, such as a circle or a square, the neighborhood range can be constructed according to the specific size; if the collection area is an irregular shape, the size of the minimum circumscribed circle or the minimum circumscribed rectangle of the minimum collection area can be selected as the size of the neighborhood range to construct a circle or rectangle, and the details will not be repeated.

Preferably, in one embodiment of the present invention, considering that the spatial attribute information has information in two dimensions, namely, it contains building facility text vectors and digital line stroke data, it is necessary to calculate the similarity separately, so the first cosine similarity of the building facility type text vector between the preset neighborhood range of the unsampled position and the collection area, and the second cosine similarity of the digital line stroke data are obtained; the sum of the first cosine similarity and the second cosine similarity is used as the spatial correlation factor. In one embodiment of the present invention, considering that the value range of the cosine similarity is [-1,1], in order to facilitate subsequent calculations, the calculated cosine similarity is normalized, and considering that there are multiple digital line stroke data, it is necessary to calculate the cosine similarity between each digital line stroke data, and then average it to obtain the second cosine similarity. The specific spatial correlation factor is expressed by the formula:

Among them, a _n is the spatial correlation factor between the nth acquisition area and the unsampled position, softmax () is the normalized exponential function, cosθ _n is the first cosine similarity between the nth acquisition area and the unsampled position; M is the number of cosine similarities between the digital line data between the nth acquisition area and the neighborhood range of the unsampled position, that is, the cosine similarity between each digital line data in the neighborhood range of the unsampled position and each digital line data in the nth acquisition area is counted to obtain the corresponding number; h _nm is the cosine similarity between the mth digital line data between the nth acquisition area and the neighborhood range of the unsampled position, that is, is the second cosine similarity.

In the city, due to urban planning issues, different areas are distributed differently in the city. Some areas are more concentrated, while others are more isolated. If you need to obtain accurate interpolation results for unsampled locations, you need to make the interpolation weights of the collection areas with the same location distribution larger. Therefore, only considering the spatial correlation factors obtained from the spatial attribute information to obtain the interpolation weights cannot obtain effective results. It is also necessary to further consider the distribution characteristics of the collection areas and unsampled locations within the city. Further considering that the collection area is a pre-set collection area, it is possible that for a complete area, the complete area is divided into two collection areas due to some reasons. If the isolation is evaluated only based on the distance between the collection areas, it will cause misjudgment. Therefore, on the basis of considering the location distribution, it is also necessary to combine the similarity of spatial attribute information to obtain the isolation factor of each collection area.

Preferably, in one embodiment of the present invention, the formula for obtaining the isolation factor includes:

Where g _n is the isolation factor of the nth collection area, N is the number of collection areas, j is the serial number of all collection areas except the nth collection area, e is a natural constant, l _nj is the distance between the nth collection area and the jth other collection area, and a _nj is the spatial correlation factor between the nth collection area and the jth other collection area.

In the isolation factor formula, The function is to normalize l _nj , that is, first use the exponential function with natural constant as the base to perform negative correlation mapping and normalization, and use 1 as the minuend to obtain the normalized result. That is, the greater the distance between the collection areas, the more remote and isolated the nth collection area is relative to the overall collection area, and the elevation information in the collection area may have certain specificity, and the greater the degree of isolation; further, the spatial correlation factor is negatively correlated, and the larger the spatial correlation factor is, the two collection areas have similar spatial attribute information and may be the same type of area, that is, the smaller the isolation factor is.

It should be noted that in the embodiment of the present invention, the designed distance can be the distance between objects in the real world, and the unit can be set to kilometers. It can also be the distance between two locations on a map, and the unit can be set to other units such as pixels. Because there is an extra normalization step in the embodiment of the present invention, there is no dimension effect, so the content here is not limited or repeated.

After obtaining the isolation factor of each collection area, the isolation of each collection area in the city is analyzed. If the unsampled position is more biased towards the collection area, it means that the data in the collection area is more reference-oriented to the unsampled position than other collection areas. Therefore, it is necessary to further analyze the bias factor between the unsampled position and the collection area. First, the relative distance between the unsampled position and each collection area needs to be obtained. If the relative distance difference between a collection area and other collection areas is large and the relative distance between the collection area and the unsampled position is small, it means that in the entire city, the unsampled position is more biased towards the collection area. Therefore, it is necessary to obtain the bias factor between each collection area and the unsampled position based on the difference in relative distance between each collection area and other collection areas, and the position distribution between each collection area and the unsampled position.

Preferably, in one embodiment of the present invention, the formula for obtaining the bias factor includes:

Among them, _Gn is the bias factor of the nth acquisition area; softmax() is the normalized exponential function; _sn is the area of the nth acquisition area; N is the number of acquisition areas; j is the serial number of all acquisition areas except the nth acquisition area; _ln is the relative distance between the nth acquisition area and the unsampled position; _lj is the relative distance between the jth other acquisition area and the unsampled position; with the unsampled position as the vertex, rays are drawn to the center points of the nth acquisition area and the jth other acquisition area, and the angle of the ray is _θnj ; sin() is the sine function; _sj is the area of the jth other acquisition area.

In the formula for obtaining the bias factor, the area information of the collection area is introduced. For the nth collection area, the larger its area is, the richer the information it contains and the more complete the measured data is. Therefore, the area is used as a weight in the formula. Similarly, the area of the jth other collection area is used as the weight for analyzing the relative situation between two collection areas. The smaller the relative distance between the nth acquisition area and the unsampled position, the more the unsampled position is biased towards the acquisition area, and the larger G _n is. By constructing rays and taking the angle θ _nj as the positional relationship between the nth acquisition area, the jth other acquisition area and the unsampled position, the larger the angle is, the greater the distribution of the nth acquisition area and the jth other acquisition area with the unsampled position as the center is in opposition, and the relative distance difference weight between the two acquisition areas should also be greater, that is, is one of the weights of the relative distance difference between the two acquisition areas. Since the angle between the two rays is less than or equal to 180°, in order to make the angle positively correlated with the weight, θ _nj is divided by 2 to obtain the sine function value and the corresponding weight. It should be noted that and softmax(s _j ) are the weights of (l _n -l _j ). The larger the (l _n -l _j ) is, the greater the relative distance difference between the nth collection area and the jth other collection areas is, and the more isolated the nth collection area is. Combined with the relationship between various parameters, the larger the G _n is, the more the unsampled locations are biased towards the nth collection area in the entire city, and the data in the nth collection area is more reference-oriented than the unsampled locations.

It should be noted that the relative distance can be obtained by calculating the Euclidean distance between the unsampled position in the map and the center point of the collection area through the Euclidean distance algorithm. The mathematical means based on the method for obtaining the center point of the collection area will not be described in detail here.

Step S3: obtaining the interpolation weight of each acquisition area relative to the unsampled position according to the isolation factor, the bias factor and the spatial correlation factor; based on the interpolation weight, interpolating the data of the unsampled position according to the elevation data in the acquisition area to obtain the elevation data at the unsampled position.

Combining the isolation factor, bias factor and spatial correlation factor, the interpolation weight of each collection area relative to the unsampled position can be obtained. That is, the closer the spatial attribute information of a collection area and the unsampled position is, the more isolated the collection area is and the more biased the unsampled position is towards the collection area, the greater the interpolation weight of the collection area relative to the unsampled position. Based on the interpolation weight, the data of the unsampled position is interpolated according to the elevation data in the collection area to obtain the elevation data at the unsampled position. At this point, the data of the elevation data dimension in the GIS data has been supplemented by the interpolation algorithm.

Preferably, in one embodiment of the present invention, the isolation factor, the bias factor and the spatial correlation factor are multiplied and then normalized to obtain the interpolation weight of each acquisition area relative to the unsampled position. It should be noted that the normalization algorithm used in one embodiment of the present invention is to use the softmax function for mapping, and other normalization methods such as maximum and minimum value normalization can also be used in other embodiments of the present invention for processing, which is not limited or elaborated here.

Preferably, in one embodiment of the present invention, in order to make the interpolation weight more accurate, after obtaining the interpolation weight, the following steps are further included:

According to the relative distance between each unsampled position and each acquisition area, the distance weight of each acquisition area relative to the unsampled position is obtained, and the average value of the distance weight and the interpolation weight is used as the new interpolation weight to update the interpolation weight. It should be noted that the distance weight can be regarded as the weight obtained according to the distance between objects in the traditional interpolation algorithm, and its acquisition method is a technical means well known to those skilled in the art. In one embodiment of the present invention, the relative distance can be directly negatively mapped and normalized, and it is ensured that the distance weights of all acquisition areas add up to 1. Because the distance weight is the normalized result, the original interpolation weight is also the normalized result, so the updated interpolation weight obtained after the average value calculation is also the normalized result, which is convenient for subsequent interpolation calculations.

In one embodiment of the present invention, the interpolation algorithm uses an inverse distance weighted interpolation model. The specific algorithm steps are technical means well known to those skilled in the art and will not be described in detail here.

Step S4: Perform multi-source data fusion display based on the interpolated GIS data.

The interpolated GIS data contains complete elevation data. By visualizing the data in each dimension and displaying the visualization results on the same display platform, multi-distance data fusion display can be completed. Users can simultaneously obtain multi-source data at each location in the city on the display platform.

In summary, the embodiment of the present invention obtains the similarity of spatial attribute information between the unsampled position and the collection area, and then determines the spatial correlation factor. The isolation factor of each collection area is further analyzed, and the bias factor is obtained according to the relative distance between each collection area and the unsampled position. The interpolation weight of each collection area relative to the unsampled position is obtained by combining the spatial correlation factor, the isolation factor and the bias factor. The unsampled position is interpolated using the interpolation weight with strong reference to obtain the corresponding elevation data, and then multi-source data fusion is performed based on the interpolated GIS data. The present invention can obtain accurate interpolation results for the unsampled position by obtaining the interpolation weight with strong reference, and then realize multi-source data fusion display based on the complete GIS data.

GIS data processing system implementation based on multi-source data fusion:

The present invention also proposes a GIS data processing system based on multi-source data fusion, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any step of a GIS data processing method based on multi-source data fusion.

A GIS data processing method embodiment for elevation data interpolation:

In the prior art, when interpolating elevation data in GIS data, the distance between the unsampled location and the collection area where data already exists is considered to construct the interpolation weight for interpolation. However, in the actual urban environment, due to the distribution of buildings, the urban terrain has complex characteristics and there are different types of buildings. If the distance is directly used to construct the interpolation weight, the final interpolation result is of poor reference. In order to solve this problem, the present invention proposes a GIS data processing method for elevation data interpolation:

Step S1: Obtain GIS data within the city.

Step S2: According to the similarity of spatial attribute information between the preset neighborhood range of the unsampled position and the collection area, obtain the spatial correlation factor between each collection area and the unsampled position; according to the position distribution between the collection areas and the similarity of spatial attribute information, obtain the isolation factor of each collection area; obtain the relative distance between the unsampled position and each collection area, and according to the difference in relative distance between each collection area and other collection areas, and the position distribution between each collection area and the unsampled position, obtain the bias factor between each collection area and the unsampled position.

Step S3: obtaining the interpolation weight of each acquisition area relative to the unsampled position according to the isolation factor, the bias factor and the spatial correlation factor; based on the interpolation weight, interpolating the data of the unsampled position according to the elevation data in the acquisition area to obtain the elevation data at the unsampled position.

Since the specific implementation of steps S1-S3 has been described in detail in the above-mentioned embodiment of the GIS data processing method based on multi-source data fusion, it will not be repeated here.

The embodiment of the present invention is based on the characteristics of GIS data. Considering that there is rich spatial attribute information in GIS data, and the spatial attribute information is not easy to be missing or collected, the spatial attribute information similarity is used to analyze when considering the correlation between the unsampled position and the collection area to obtain the spatial correlation factor. Further considering that when analyzing the correlation between the collection area and the unsampled position, it is not only necessary to consider the distance, but also the location distribution in the urban area. If the unsampled position and a certain collection area are both isolated areas in the urban area, and the unsampled position is more biased towards the collection area, it means that the data of the collection area is of great reference value for the unsampled position. Therefore, the isolation factor of the collection area and the bias factor between the collection area and the unsampled position are analyzed, and then the interpolation weight with strong reference value can be obtained by combining the spatial correlation factor. Then, the data of the unsampled position can be interpolated to obtain accurate elevation data.

It should be noted that the sequence of the above embodiments of the present invention is for description only and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims (6)

1. A GIS data processing method based on multi-source data fusion, the method comprising:

Acquiring GIS data in a city range; the GIS data comprises spatial attribute information of the whole city range and elevation data in a preset acquisition area; the city range also comprises an un-sampled position;

Obtaining a spatial correlation factor between each acquisition region and the non-sampling position according to the similarity of spatial attribute information between the preset neighborhood range of the non-sampling position and the acquisition region; obtaining an isolated factor of each acquisition region according to the position distribution among the acquisition regions and the similarity condition of the spatial attribute information; obtaining the relative distance between the non-sampling position and each acquisition region, and obtaining a deflection factor between each acquisition region and the non-sampling position according to the difference of the relative distance between each acquisition region and other acquisition regions and the position distribution between each acquisition region and the non-sampling position;

Obtaining interpolation weights of each acquisition region relative to the non-sampling positions according to the isolation factors, the bias factors and the spatial correlation factors; interpolating the data of the non-sampling position according to the elevation data in the acquisition area based on the interpolation weight to obtain the elevation data of the non-sampling position;

Carrying out multi-source data fusion display according to the GIS data after interpolation processing;

The space attribute information comprises digital line drawing data and a construction facility type text vector;

the method for acquiring the spatial correlation factor comprises the following steps:

Acquiring a first cosine similarity of a text vector of a building facility type between a preset neighborhood range of the non-sampling position and the acquisition region and a second cosine similarity of digital line drawing data; taking the sum of the first cosine similarity and the second cosine similarity as the spatial correlation factor;

The obtaining formula of the isolation factor comprises the following steps:

; wherein the method comprises the steps of N is the number of acquisition areas, j is the serial number of other acquisition areas except the nth acquisition area in all the acquisition areas, e is a natural constant,For the distance between the nth acquisition region and the jth other acquisition region,A spatial correlation factor between the nth acquisition region and the jth other acquisition region;

the acquisition formula of the deflection factor comprises the following steps:

; wherein, A bias factor for the nth acquisition region; Is a normalized exponential function; Is the area of the nth acquisition region; n is the number of acquisition areas; j is the serial number of the other acquisition areas except the nth acquisition area in all the acquisition areas; is the relative distance between the nth acquisition region and the non-sampled position; A relative distance between the j-th other acquisition region and the non-sampled position; taking the non-sampling position as a vertex to respectively take rays from the center point of the nth acquisition region and the center point of the jth other acquisition region, wherein the included angle of the rays is ；Is a sine function; Is the area of the jth other acquisition region.

2. The method for processing GIS data based on multi-source data fusion according to claim 1, wherein a size of an acquisition region with a minimum area is selected as a size of the neighborhood range, and the neighborhood range is constructed with the non-sampled position as a center.

3. The GIS data processing method based on multi-source data fusion according to claim 1, wherein the obtaining method for obtaining the interpolation weight of each acquisition region relative to the non-sampling position according to the isolation factor, the bias factor and the spatial correlation factor includes:

And multiplying the isolation factor, the deflection factor and the space correlation factor, and then carrying out normalization processing to obtain the interpolation weight of each acquisition region relative to the non-sampling position.

4. The method for processing GIS data based on multi-source data fusion according to claim 1, wherein obtaining interpolation weights further comprises:

And obtaining the distance weight of each acquisition region relative to the non-sampling position according to the relative distance between each non-sampling position and each acquisition region, and updating the interpolation weight by taking the average value of the distance weight and the interpolation weight as a new interpolation weight.

5. The method for processing GIS data based on multi-source data fusion according to claim 1, wherein the data of the non-sampling position is interpolated by an inverse distance weighted interpolation method to obtain elevation data of the non-sampling position.

6. A GIS data processing system based on multi-source data fusion, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of a GIS data processing method based on multi-source data fusion as claimed in any one of claims 1 to 5 when executing the computer program.