A method and system for fusion processing of geographic information data based on cloud computing

Abstract

The present invention provides a geographic information data fusion processing method and system based on cloud computing, which relates to the field of data processing technology. The method comprises: on a cloud computing platform, encoding parameters in the reflection data fusion process into genes, randomly generating an initial population, and each individual represents a parameter combination; defining a fitness function for evaluating the quality of each individual; calculating a fitness value according to the fitness function, and selecting a corresponding individual according to the fitness value to perform crossover and mutation operations to generate a new population, repeating the crossover and mutation operations until a preset number of iterations is reached to obtain the genes of the final individual; decoding the genes of the final individual into parameters of the reflection data fusion process; and calculating dynamic factors according to the characteristics of the current reflection data and the historical fusion effect. The present invention can improve the accuracy of data fusion.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a geographic information data fusion processing method and system based on cloud computing. Background Art

[0002] In recent years, cloud computing technology has been widely used in the field of geographic information data processing. Cloud computing platforms have the advantages of powerful computing power and elastic expansion, and can efficiently process massive amounts of geographic information data. However, how to achieve effective integration of geographic information data on cloud computing platforms remains an urgent problem to be solved.

[0003] Most of the existing geographic information data fusion methods are based on static weight allocation, without considering the dynamic changes and characteristic differences of data. As a result, the fusion results may not accurately reflect the real situation of geographic information data, reducing the accuracy and reliability of data fusion. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a geographic information data fusion processing method and system based on cloud computing, which can improve the accuracy of data fusion.

[0005] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0006] In a first aspect, a geographic information data fusion processing method based on cloud computing, the method comprising:

[0007] Step 11, use a radio wave measurement device to scan the target area, and obtain the original reflection data of the surface and underground structures by transmitting radio waves and receiving the reflected signals;

[0008] Step 12, preprocessing the original reflection data to obtain preprocessed reflection data;

[0009] Step 13, uploading the preprocessed reflection data to the cloud computing platform;

[0010] Step 14: On the cloud computing platform, encode the parameters in the reflection data fusion process into genes, randomly generate an initial population, and each individual represents a parameter combination; define a fitness function for evaluating the quality of each individual;

[0011] Step 15, according to the fitness function, calculate the fitness value, and select the corresponding individual according to the fitness value to perform crossover and mutation operations to generate a new population, repeat the crossover and mutation operations until the preset number of iterations is reached to obtain the genes of the final individual; decode the genes of the final individual into parameters of the reflection data fusion process;

[0012] Step 16, calculating the dynamic factor according to the characteristics of the current reflection data and the historical fusion effect;

[0013] Step 17, dynamically adjust the fusion weights of data from different sources according to the dynamic factors to obtain the final geographic information data.

[0014] Further, the pre-processed reflection data is uploaded to the cloud computing platform, including:

[0015] Step 131, evaluating the preprocessed reflection data to obtain an evaluation result, and preliminarily determining a data block size range based on the evaluation result;

[0016] Step 132, initializing the boundary position of the data block as the starting point, calculating the loss function value according to the current data block boundary, and updating the boundary position of the data block until a preset number of iterations is reached to obtain a determined data block boundary;

[0017] Step 133, performing final segmentation on the preprocessed reflection data according to the determined data block boundaries to obtain segmented data blocks;

[0018] Step 134, uniquely mark each segmented data block, create an index file, and record the mark, size and location information of each data block in the original data;

[0019] Step 135 , uploading the marked data blocks one by one to the cloud computing platform according to the size of the data blocks and the network conditions.

[0020] Further, the preprocessed reflection data is evaluated to obtain an evaluation result, and based on the evaluation result, a data block size range is preliminarily determined, including:

[0021] The preprocessed reflection data is processed by Calculate the evaluation value ,in, Indicates the amount of data transferred each time; Indicates the size of the preprocessed reflection data; Indicates the bandwidth of a network, which refers to the amount of data that the network can transmit per second; Indicates the theoretical upper limit of network bandwidth; Indicates the actual transmission capacity of the current network; represents a fixed delay in the data transmission process; Indicates the physical distance that data travels on a network; is the category in the preprocessed reflection data The probability of occurrence, is the number of different categories in the preprocessed reflection data; , and represents the weight coefficient;

[0022] Analyze the evaluation values ​​under different data block sizes , according to the evaluation value Based on the changing trend of the data, the size range of the data block can be preliminarily determined.

[0023] Furthermore, the corresponding individuals are selected according to the fitness value to perform crossover and mutation operations to generate a new population, and the crossover and mutation operations are repeated until the preset number of iterations is reached to obtain the genes of the final individuals, including:

[0024] According to the fitness value, a part of individuals are selected to enter the next generation using the tournament selection algorithm;

[0025] For the selected individuals, random pairing is performed and crossover operation is performed through multi-point crossover according to the preset crossover rate to generate new individuals;

[0026] For the new individuals, mutation operation is performed according to the preset mutation rate to obtain the mutated new individuals;

[0027] The new individuals and the mutated new individuals are combined into a new population, and a check is made to see whether the preset number of iterations has been reached. If reached, the iteration is stopped and the genes of the final individuals are output.

[0028] Furthermore, according to the characteristics of the current reflection data and the historical fusion effect, the dynamic factors are calculated, including:

[0029] According to the characteristics of the current reflection data and the historical fusion effect, Calculating dynamic factors ;

[0030] in, , and is the weight coefficient; Indicates the bandwidth of the current network; Indicates the maximum value of network bandwidth under ideal conditions; represents the adjustment coefficient; represents a constant factor; Indicates the distance from the transmitting source to the receiver; Indicates the reference distance; represents the decay index;

[0031] ;

[0032] in, is the effective power of the signal, indicating the strength of the target signal itself; is the received signal strength, which indicates the power of the target signal received at the receiving end; , and is the adjustment coefficient; is the interference strength, which indicates the power of non-target signals in the signal; Indicates the rate of errors that occur during data transmission; represents the adjustment coefficient; where:

[0033] ;

[0034] in, represents the adjustment coefficient; Indicates The error of the data source; Indicates the number of data sources; Indicates the index of the error.

[0035] Furthermore, the fusion weights of data from different sources are dynamically adjusted according to dynamic factors to obtain the final geographic information data, including:

[0036] For each data source, set an initial fusion weight ;

[0037] Using dynamic factors pass Calculate the new fusion weights for each data source, where It is The adjusted weights of the data sources; is the adjustment factor corresponding to each data source;

[0038] According to the new fusion weight of each data source, the geographic information data from different sources are fused to obtain the final geographic information data.

[0039] Furthermore, according to the new fusion weight of each data source, the geographic information data from different sources are fused to obtain the final geographic information data, including:

[0040] According to the new fusion weight of each data source, Merge geographic information data from different sources to obtain the final geographic information data ;

[0041] in, Indicates Geographic information data from various data sources; represents the total number of adjusted weights; The index representing the adjusted weight.

[0042] In a second aspect, a geographic information data fusion processing system based on cloud computing includes:

[0043] An acquisition module is used to scan the target area using a radio wave measurement device, and acquire the original reflection data of the surface and underground structures by transmitting radio waves and receiving the reflected signals;

[0044] A preprocessing module, used for preprocessing the original reflection data to obtain preprocessed reflection data;

[0045] An upload module, used to upload the preprocessed reflection data to the cloud computing platform;

[0046] The fusion module is used to encode the parameters in the reflection data fusion process into genes on the cloud computing platform, randomly generate the initial population, and each individual represents a parameter combination; define the fitness function used to evaluate the quality of each individual;

[0047] The iteration module is used to calculate the fitness value according to the fitness function, and select the corresponding individuals according to the fitness value to perform crossover and mutation operations to generate a new population, and repeat the crossover and mutation operations until the preset number of iterations is reached to obtain the genes of the final individual; the genes of the final individual are decoded into parameters of the reflection data fusion process;

[0048] A calculation module is used to calculate the dynamic factor according to the characteristics of the current reflection data and the historical fusion effect;

[0049] The adjustment module is used to dynamically adjust the fusion weights of data from different sources according to dynamic factors to obtain the final geographic information data.

[0050] According to a third aspect, a computing device includes:

[0051] one or more processors;

[0052] The storage device is used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method described.

[0053] In a fourth aspect, a computer-readable storage medium stores a program, and when the program is executed by a processor, the method described is implemented.

[0054] The above scheme of the present invention includes at least the following beneficial effects.

[0055] By using genetic algorithms to optimize the parameters of the reflection data fusion process, the present invention can find a more suitable parameter combination, thereby improving the accuracy of data fusion. At the same time, dynamically adjusting the fusion weights of data from different sources can further ensure that the fusion results more accurately reflect the actual situation.

[0056] The present invention utilizes the powerful computing power of the cloud computing platform to efficiently process massive amounts of reflection data. Through the parallel processing mechanism of cloud computing, data processing time can be greatly shortened and overall processing efficiency can be improved.

[0057] Traditional data fusion methods are often based on static weights, but the present invention can adjust the fusion weights in real time according to data characteristics and historical fusion effects by calculating dynamic factors, making data fusion more flexible and dynamic.

[0058] The application of genetic algorithms enables the system to adaptively find the optimal solution when facing a complex and changing data environment, enhances the robustness and stability of the system, and maintains a high fusion accuracy even when there is noise or anomalies in the data. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 It is a flowchart of a geographic information data fusion processing method based on cloud computing provided by an embodiment of the present invention.

[0060] Figure 2 It is a schematic diagram of a geographic information data fusion processing system based on cloud computing provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0061] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0062] like Figure 1 As shown, an embodiment of the present invention proposes a geographic information data fusion processing method based on cloud computing, and the method includes the following steps:

[0063] Step 11, use a radio wave measurement device to scan the target area, and obtain the original reflection data of the surface and underground structures by transmitting radio waves and receiving the reflected signals;

[0064] Step 12, preprocessing the original reflection data to obtain preprocessed reflection data;

[0065] Step 13, uploading the preprocessed reflection data to the cloud computing platform;

[0066] Step 14: On the cloud computing platform, encode the parameters in the reflection data fusion process into genes, randomly generate an initial population, and each individual represents a parameter combination; define a fitness function for evaluating the quality of each individual;

[0067] Step 15, according to the fitness function, calculate the fitness value, and select the corresponding individual according to the fitness value to perform crossover and mutation operations to generate a new population, repeat the crossover and mutation operations until the preset number of iterations is reached to obtain the genes of the final individual; decode the genes of the final individual into parameters of the reflection data fusion process;

[0068] Step 16, calculating the dynamic factor according to the characteristics of the current reflection data and the historical fusion effect;

[0069] Step 17, dynamically adjust the fusion weights of data from different sources according to the dynamic factors to obtain the final geographic information data.

[0070] In an embodiment of the present invention, by using a genetic algorithm to optimize the parameters of the reflection data fusion process, the present invention can find a more suitable combination of parameters, thereby improving the accuracy of data fusion. At the same time, dynamically adjusting the fusion weights of data from different sources can further ensure that the fusion results more accurately reflect the actual situation. The present invention utilizes the powerful computing power of the cloud computing platform to efficiently process massive amounts of reflection data. Through the parallel processing mechanism of cloud computing, the data processing time can be greatly shortened and the overall processing efficiency can be improved. Traditional data fusion methods are often based on static weights, while the present invention can adjust the fusion weights in real time according to data characteristics and historical fusion effects by calculating dynamic factors, making data fusion more flexible and dynamic. The application of genetic algorithms enables the system to adaptively find the optimal solution when facing a complex and changeable data environment, enhances the robustness and stability of the system, and maintains a high fusion accuracy even when there is noise or anomalies in the data.

[0071] In a preferred embodiment of the present invention, the above step 11, using a radio wave measurement device to scan the target area, by transmitting radio waves and receiving the reflected signals, to obtain the original reflection data of the surface and underground structures, may include:

[0072] Select radio wave measurement equipment, which can transmit radio waves and receive their reflected signals; scan the target area, the equipment will transmit radio waves and receive the signals reflected from the surface and underground structures.

[0073] In a preferred embodiment of the present invention, the above step 12, preprocessing the original reflection data to obtain preprocessed reflection data, may include:

[0074] The raw reflection data obtained by scanning are sorted to ensure the integrity and accuracy of the data; since various interferences may occur during the data collection process, generating noise data, denoising processing is required to improve data quality.

[0075] For example, the engineering team selected suitable radio wave measurement equipment and made detailed planning and marking of the target area. The target area was comprehensively scanned using the radio wave measurement equipment. The equipment emitted radio waves and received reflected signals, successfully obtaining the original reflection data of the underground pipelines. After the scan was completed, the engineer imported the data into professional software for preprocessing. Through operations such as denoising and signal enhancement, clearer and more accurate underground pipeline data was obtained. After data fusion processing, the distribution of underground pipelines in the target area was successfully obtained, providing an important reference for subsequent construction. This not only improved construction efficiency, but also greatly enhanced construction safety.

[0076] In a preferred embodiment of the present invention, uploading the preprocessed reflection data to a cloud computing platform includes:

[0077] Step 131, evaluating the preprocessed reflection data to obtain an evaluation result, and preliminarily determining a data block size range based on the evaluation result;

[0078] Step 132, initializing the boundary position of the data block as the starting point, calculating the loss function value according to the current data block boundary, and updating the boundary position of the data block until a preset number of iterations is reached to obtain a determined data block boundary;

[0079] Step 133, performing final segmentation on the preprocessed reflection data according to the determined data block boundaries to obtain segmented data blocks;

[0080] Step 134, uniquely mark each segmented data block, create an index file, and record the mark, size and location information of each data block in the original data;

[0081] Step 135 , uploading the marked data blocks one by one to the cloud computing platform according to the size of the data blocks and the network conditions.

[0082] In the embodiment of the present invention, by evaluating the data and preliminarily determining the data block size range in step 131, and by iteratively optimizing the data block boundaries in step 132, it is possible to ensure that the size of the data blocks after segmentation is moderate, and neither the size of the data blocks will cause difficulty in uploading due to being too large, nor the number of uploads and the cost will be increased due to being too small. This optimization significantly improves the overall efficiency of data uploading. In step 133, the data is finally segmented according to the determined data block boundaries, so that different parts of the data can be processed independently. This segmentation method facilitates parallel processing on the cloud computing platform and improves the flexibility and speed of data processing. In step 134, each segmented data block is uniquely marked and an index file is created to record detailed information. This management method greatly simplifies the subsequent retrieval, update and maintenance of data, and enhances the convenience and accuracy of data management. By uploading data blocks one by one according to the data block size and network conditions in step 135, network fluctuations and interruptions can be effectively dealt with. Even if the upload of a data block fails, it will not affect the transmission of other data blocks, thereby ensuring the reliability and stability of the overall data transmission. By reasonably segmenting and uploading the pre-processed reflection data to the cloud computing platform, the distributed storage and computing power of cloud computing can be better utilized. This not only helps improve the efficiency of data processing and analysis, but also promotes the effective use of cloud computing resources and cost savings.

[0083] In a preferred embodiment of the present invention, the above step 131 evaluates the preprocessed reflection data to obtain an evaluation result, and preliminarily determines the data block size range according to the evaluation result, including:

[0084] The preprocessed reflection data is processed by Calculate the evaluation value ,in, Indicates the amount of data transferred each time; Indicates the size of the preprocessed reflection data; Indicates the bandwidth of a network, which refers to the amount of data that the network can transmit per second; Indicates the theoretical upper limit of network bandwidth; Indicates the actual transmission capacity of the current network; represents a fixed delay in the data transmission process; Indicates the physical distance that data travels on a network; is the category in the preprocessed reflection data The probability of occurrence, is the number of different categories in the preprocessed reflection data; , and represents the weight coefficient;

[0085] Analyze the evaluation values ​​under different data block sizes , according to the evaluation value The change trend of the data block size is used to preliminarily determine the size range of the data block, which specifically includes: selecting different data block sizes as candidate values. These data block sizes can be set according to actual conditions, such as starting from a smaller value and gradually increasing until reaching a larger value. For each candidate data block size, the corresponding evaluation value L is calculated using the above evaluation formula. This evaluation formula comprehensively considers multiple factors, including data volume, network bandwidth, transmission delay, data category distribution, etc., to quantify the impact of different data block sizes on overall performance. The evaluation values ​​L under different data block sizes are compared and analyzed. The evaluation value L will show a certain trend as the data block size changes. By observing the change trend of the evaluation value L, it can be found that when the data block size increases to a certain range, the evaluation value L will reach a relatively low level, which indicates that the data block size within this range has a better performance. According to the change trend of the evaluation value L, a range of data block sizes can be preliminarily determined. This range is an area where the evaluation value L is relatively low and stable. This range can be used as a reference for subsequent data transmission and processing to ensure that the size of the data block is neither too large to cause low transmission and processing efficiency, nor too small to increase additional overhead.

[0086] In the embodiment of the present invention, the evaluation formula comprehensively considers multiple factors such as data volume, network bandwidth, transmission delay, etc., and can more accurately find the data block size suitable for the current network environment and data characteristics. This helps to reduce congestion and delay in the data transmission process, thereby improving data transmission efficiency. Reasonable data block size division can make more effective use of the resources of the cloud computing platform. Too large data blocks may lead to insufficient processing resources, while too small data blocks may increase additional overhead. The data block size range determined by evaluation helps to achieve balanced utilization of resources. The evaluation formula takes into account the category distribution of data (reflected by the calculation of entropy), which means that the division of data blocks will pay more attention to the intrinsic characteristics of the data. By dynamically determining the data block size range based on the evaluation results, the method can adapt to different network environments and data characteristics. Reasonable data block size division can reduce unnecessary data transmission and processing, thereby reducing the cost of cloud computing services.

[0087] In a preferred embodiment of the present invention, the above step 132, initializing the boundary position of the data block as the starting point, calculating the loss function value according to the current data block boundary, and updating the boundary position of the data block until a preset number of iterations is reached to obtain a determined data block boundary, may include:

[0088] Based on the data block size range initially determined in step 131, a starting point is selected as the initial boundary position of the data block, and this starting point can be the midpoint of the range. The role of the loss function here is the same as the evaluation formula in step 131, which is used to quantify the quality of the data block division. Therefore, the loss function in step 132 can be calculated using the same formula as step 131.

[0089] Specifically, according to the current data block boundary, an evaluation formula (i.e., loss function) is used to calculate the loss value. This loss value reflects the impact of the current boundary division on data uploading and processing efficiency. According to the calculated loss function value, the boundary position of the data block is adjusted. If the loss value is large, it means that the current boundary division may not be optimal, and the loss needs to be reduced by moving the boundary. The update of the boundary position can be based on the gradient descent method, and the goal is to find the boundary position that minimizes the loss function value. Repeat the above calculation and update process until the preset number of iterations is reached. Each iteration will try to find a better data block boundary division to reduce the loss function value. Through multiple iterations, the optimal data block boundary division can be gradually approached.

[0090] In a preferred embodiment of the present invention, the above step 133 may include:

[0091] The determined optimal data block boundary information is read from the output result of step 132. This information is saved in a temporary file or a specific data structure, and the start and end positions of each data block are recorded in detail.

[0092] Open the preprocessed reflection data file or data stream, prepare for the reading operation, load the optimal data block boundary information, and ensure the accuracy of the boundary position. According to the starting position in the boundary information, locate the starting point of the first data block in the file, and prepare a buffer for temporary storage of the read data. Starting from the current positioning position, read the data into the buffer according to the length specified in the boundary information, ensure that the amount of data read strictly conforms to the size of each data block defined in the boundary information, and process or store the data in the buffer as an independent data block. Move the file pointer or data stream position to the starting point of the next data block, and repeat the steps until all defined data blocks have been read and processed. Store each segmented data block as needed, which can be written to different files or saved in a data structure in memory, ensuring that each data block has a unique identifier for subsequent management and access. Close the file stream or data input source, release related resources, and output the information on the completion of the segmentation, including the number and size of the segmented data blocks and other statistical information.

[0093] Apply the selected data segmentation algorithm and segment the preprocessed reflection data strictly according to the optimal boundary information. The segmentation method can include byte-by-byte, line-by-line processing, or based on a specific format of data blocks. The specific method should be determined according to the actual organization and storage method of the data. After the segmentation is completed, each data block is verified to ensure that its content is complete and the boundary division is accurate. The verification method can include comparing the total amount of data before and after the segmentation, checking whether the boundaries of each data block are consistent with the expectations, or performing content verification on the data. The verified data blocks are saved in the specified storage location, which can be a local file system or a temporary cache area. The stored data blocks will be used for subsequent tagging and uploading operations. During or after the data segmentation, detailed information about the segmentation is recorded, such as the timestamp of the segmentation operation, the number of data blocks generated, and their respective sizes.

[0094] In a preferred embodiment of the present invention, the above step 134 may include:

[0095] Each segmented data block is uniquely labeled to ensure that each data block can be accurately identified in subsequent processing. At the same time, an index file is created to record the label, size and location information of each data block in the original data. This index file will serve as a reference for data upload and retrieval.

[0096] In a preferred embodiment of the present invention, the above step 135 may include:

[0097] According to the size of the data block and the network condition, the marked data blocks are uploaded to the cloud computing platform one by one. During the upload process, the index file created in step 134 can be used to ensure the integrity and consistency of the data. At the same time, the upload strategy is dynamically adjusted according to the network condition to maximize the upload efficiency and stability.

[0098] In a preferred embodiment of the present invention, in step 14, on a cloud computing platform, the parameters in the reflection data fusion process are encoded as genes, and an initial population is randomly generated, where each individual represents a parameter combination; and a fitness function for evaluating the quality of each individual is defined, which may include:

[0099] Step 14 describes a key step in the process of performing reflection data fusion on the cloud computing platform, which involves the application of genetic algorithm (GA). Genetic algorithm is an optimization algorithm that simulates natural selection and genetics mechanism and is used to find the optimal solution in a complex search space.

[0100] In this step, the key parameters in the reflection data fusion process are encoded as "genes". Each parameter may correspond to one or more genes, and the combination of these genes will determine the effect of data fusion. For example, if reflection data fusion involves three key parameters: fusion weight, filter coefficient, and time domain window size, then these three parameters can be encoded as three groups of genes respectively.

[0101] The initial population is composed of multiple individuals, each of which represents a parameter combination, that is, a set of specific genetic codes. These individuals are randomly generated to ensure the diversity of the search space. The size of the population (that is, the number of individuals) is pre-set and can be adjusted according to the complexity of the problem and computing resources; the fitness function is used to evaluate the pros and cons of each individual, that is, the performance of each parameter combination in the reflection data fusion process. For example, if the goal is to maximize the signal-to-noise ratio (SNR) of the fused data, then the fitness function can be designed as a calculation formula for SNR.

[0102] ;

[0103] in, Is a combination of parameters The fitness value of In the parameter combination The signal-to-noise ratio calculated below; In the parameter combination The power of the next signal; In the parameter combination The power of the noise.

[0104] In a preferred embodiment of the present invention, the corresponding individuals are selected according to the fitness value to perform crossover and mutation operations to generate a new population, and the crossover and mutation operations are repeated until a preset number of iterations is reached to obtain the genes of the final individuals, including:

[0105] According to the fitness value, a part of individuals are selected to enter the next generation using the tournament selection algorithm;

[0106] For the selected individuals, random pairing is performed and crossover operation is performed through multi-point crossover according to the preset crossover rate to generate new individuals;

[0107] For the new individuals, mutation operation is performed according to the preset mutation rate to obtain the mutated new individuals;

[0108] The new individuals and the mutated new individuals are combined into a new population, and a check is made to see whether the preset number of iterations has been reached. If reached, the iteration is stopped and the genes of the final individuals are output.

[0109] In an embodiment of the present invention, according to the fitness value, a tournament selection algorithm is used to select a part of individuals with excellent performance from the current population to enter the next generation. This method selects better individuals by comparing the fitness values ​​between individuals, thereby ensuring the evolutionary direction of the population. For the selected individuals, random pairing is performed and crossover operation is performed through multi-point crossover according to a preset crossover rate (such as 0.7 or 0.8). This step simulates the gene recombination of organisms in the genetic process, which helps to generate new gene combinations that may have better performance. For the new individuals generated after crossover, mutation operation is performed according to a preset mutation rate (usually low, such as 0.01~0.1). The mutation operation increases the diversity of the population by randomly changing some gene values ​​in the individual, which helps the algorithm to jump out of the local optimal solution and explore a wider search space. The new individuals generated after crossover and mutation form a new population. Then check whether the algorithm has reached the preset number of iterations. If it has reached, stop the iteration and output the genes of the final individual; otherwise, continue to repeat the above selection, crossover and mutation operations. By simulating natural selection and genetic mechanisms, the genetic algorithm can find the global optimal solution or approximate optimal solution in a complex search space. This global optimization capability gives genetic algorithms a significant advantage in solving many practical problems. Since genetic algorithms use populations for search and continuously evolve populations through selection, crossover, and mutation operations, they are not very sensitive to the selection of the initial population and the setting of parameters. This makes genetic algorithms highly robust and able to function stably in different application scenarios. The various steps of genetic algorithms (such as selection, crossover, and mutation) can be easily implemented in parallel, thereby making full use of the multi-core or multi-threaded processing capabilities of modern computing devices. This parallel processing capability can significantly improve the operating efficiency of the algorithm and shorten the solution time.

[0110] For example, in order to obtain a suitable fusion effect, it is necessary to optimize several key parameters in the fusion process, such as fusion weight, filter coefficient, etc. It is decided to use a genetic algorithm to automatically find a suitable combination of these parameters. The fusion weight (ranging from 0 to 1, accurate to two decimal places) and the filter coefficient (integer, ranging from 1 to 10) are taken as the parameters to be optimized. The fusion weight can be encoded as a binary representation of two decimal places, while the filter coefficient can be directly represented by its integer value. An initial population of 100 individuals is randomly generated, and each individual represents a parameter combination. Define the fitness function; randomly select 5 individuals from the population as a group, compare their fitness values, and select the 2 individuals with the highest scores to enter the next generation. Repeat this process until 50 individuals are selected to enter the next generation; for the 50 selected individuals, randomly pair them into 25 pairs, and each pair of individuals performs a multi-point crossover operation with a crossover rate of 0.8. For example, two points can be selected and the gene fragments between the two points can be exchanged to generate new individuals. For the new individuals generated after crossover, mutation operation is performed at a mutation rate of 0.05. The mutation can be a bit flip of the binary code of the fusion weight, or a random increase or decrease of the filter coefficient. The new individuals generated after crossover and mutation form a new population, and check whether the algorithm has reached the preset number of iterations (for example, 100 times). If reached, the iteration is stopped; otherwise, the above repetition is continued. When the algorithm stops iterating, the gene of the individual with the highest fitness value is output, that is, the optimal combination of fusion weights and filter coefficients.

[0111] In a preferred embodiment of the present invention, the final individual's genes are decoded into parameters of the reflection data fusion process, including:

[0112] At the beginning of the genetic algorithm, it has been determined that the parameters of the reflection data fusion process will be encoded as genes, which involves discretizing continuous parameters or using binary, integer, etc. encoding methods. For example, the fusion weights may be encoded in binary, while the filter coefficients may be directly encoded in integers. After the genetic algorithm runs, one or more individuals with the highest fitness will be obtained, and their genes represent the optimized parameter combination. For binary-encoded parameters: If some parameters (such as fusion weights) are encoded in binary, these binary strings need to be converted back to the original values. This usually involves converting the binary string to decimal numbers, and may need to be adjusted accordingly according to the scaling factor when encoding. Finally, the decoded parameters are applied to the reflection data fusion process.

[0113] In a preferred embodiment of the present invention, the dynamic factor is calculated according to the characteristics of the current reflection data and the historical fusion effect, including:

[0114] According to the characteristics of the current reflection data and the historical fusion effect, Calculating dynamic factors ;

[0115] in, , and is the weight coefficient; Indicates the bandwidth of the current network; Indicates the maximum value of network bandwidth under ideal conditions; represents the adjustment coefficient; represents a constant factor; Indicates the distance from the transmitting source to the receiver; Indicates the reference distance; represents the decay index;

[0116] ;

[0117] in, is the effective power of the signal, indicating the strength of the target signal itself; is the received signal strength, which indicates the power of the target signal received at the receiving end; , and is the adjustment coefficient; is the interference strength, which indicates the power of non-target signals in the signal; Indicates the rate of errors that occur during data transmission; represents the adjustment coefficient; where:

[0118] ;

[0119] in, represents the adjustment coefficient; Indicates The error of the data source; Indicates the number of data sources; Indicates the index of the error.

[0120] In the embodiment of the present invention, the calculation method of the dynamic factor D integrates multiple influencing factors, including network bandwidth, signal strength, interference strength, error rate, and data source error, so that the method can adaptively adjust the fusion strategy according to different environments and data characteristics, thereby improving the adaptability and flexibility of the algorithm. , and The setting of can flexibly adjust the influence of various factors on the dynamic factor D, so as to optimize the fusion effect according to the actual situation. While considering the current network bandwidth A, this method also introduces multiple parameters such as signal effective power P, received signal strength R, interference strength G, and data transmission error rate E to enhance the robustness of the algorithm to changes in signal quality and network status. Even when the signal is interfered or the network conditions are poor, a relatively stable fusion effect can be maintained. By introducing the H term, this method can take into account the errors of different data sources. , and adjust the fusion strategy according to the combined impact of these errors. This helps to correct the bias or uncertainty in the data source itself, thereby improving the accuracy and reliability of the fusion results.

[0121] In a preferred embodiment of the present invention, the fusion weights of data from different sources are dynamically adjusted according to dynamic factors to obtain final geographic information data, including:

[0122] For each data source, set an initial fusion weight ;

[0123] Using dynamic factors pass Calculate the new fusion weights for each data source, where It is The adjusted weights of the data sources; is the adjustment factor corresponding to each data source;

[0124] According to the new fusion weight of each data source, the geographic information data from different sources are fused to obtain the final geographic information data.

[0125] In the embodiment of the present invention, by introducing the dynamic factor D, the present invention can adjust the fusion weight of each data source in real time according to the current environment and data characteristics. This dynamic adjustment mechanism makes the fusion process more flexible and can better adapt to the data fusion requirements in different scenarios. Using the dynamic factor D to calculate the new fusion weight can more accurately reflect the importance and reliability of each data source under specific conditions. In this way, the present invention can optimize the fusion results and improve the accuracy and credibility of the final geographic information data. Taking into account that different data sources may have different characteristics and qualities, the present invention sets an independent adjustment coefficient for each data source. , the fusion weight of each data source can be controlled more finely. This helps to make full use of the effective information in the diversified data sources and improve the fusion effect. The method of dynamically adjusting the fusion weight can offset the influence of noise and outliers in the data source on the fusion result to a certain extent. By adjusting the weight in real time, the present invention can reduce the interference of unstable data sources on the final result, thereby enhancing the robustness of the fusion process.

[0126] In a preferred embodiment of the present invention, geographic information data from different sources are fused according to a new fusion weight of each data source to obtain final geographic information data, including:

[0127] According to the new fusion weight of each data source, Merge geographic information data from different sources to obtain the final geographic information data ;

[0128] in, Indicates Geographic information data from various data sources; represents the total number of adjusted weights; The index representing the adjusted weight.

[0129] In the embodiment of the present invention, by introducing a new fusion weight The present invention can more accurately weigh the importance of geographic information data from different data sources in the fusion process. This weighted fusion method fully considers the reliability and relevance of each data source, thereby improving the final fusion data. accuracy. In practical applications, the quality of geographic information data from different data sources often differs. By assigning dynamically adjusted new fusion weights to each data source, the present invention can effectively compensate for these quality differences, ensuring that high-quality data sources account for a larger proportion of the fusion results, thereby improving the overall data quality. The weighted fusion method can reduce the impact of individual abnormal or unstable data sources on the fusion results. When a data source fluctuates or becomes abnormal, its weight will be adjusted accordingly, thereby reducing interference with the final fusion result and maintaining the stability of the fusion result.

[0130] like Figure 2 As shown, an embodiment of the present invention further provides a geographic information data fusion processing system based on cloud computing, comprising:

[0131] An acquisition module is used to scan the target area using a radio wave measurement device, and acquire the original reflection data of the surface and underground structures by transmitting radio waves and receiving the reflected signals;

[0132] A preprocessing module, used for preprocessing the original reflection data to obtain preprocessed reflection data;

[0133] An upload module, used to upload the preprocessed reflection data to the cloud computing platform;

[0134] The fusion module is used to encode the parameters in the reflection data fusion process into genes on the cloud computing platform, randomly generate the initial population, and each individual represents a parameter combination; define the fitness function used to evaluate the quality of each individual;

[0135] The iteration module is used to calculate the fitness value according to the fitness function, and select the corresponding individuals according to the fitness value to perform crossover and mutation operations to generate a new population, and repeat the crossover and mutation operations until the preset number of iterations is reached to obtain the genes of the final individual; the genes of the final individual are decoded into parameters of the reflection data fusion process;

[0136] A calculation module is used to calculate the dynamic factor according to the characteristics of the current reflection data and the historical fusion effect;

[0137] The adjustment module is used to dynamically adjust the fusion weights of data from different sources according to dynamic factors to obtain the final geographic information data.

[0138] It should be noted that the system is a system corresponding to the above method, and all implementation methods in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0139] The embodiment of the present invention further provides a computing device, comprising: a processor, a memory storing a computer program, wherein when the computer program is executed by the processor, the method described above is executed. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0140] The embodiment of the present invention also provides a computer-readable storage medium storing instructions, which, when executed on a computer, enable the computer to execute the method described above. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0141] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A geographic information data fusion processing method based on cloud computing, characterized in that: The method comprises: Step 11, use a radio wave measurement device to scan the target area, and obtain the original reflection data of the surface and underground structures by transmitting radio waves and receiving the reflected signals; Step 12, preprocessing the original reflection data to obtain preprocessed reflection data; Step 13, uploading the preprocessed reflection data to the cloud computing platform; Step 14: On the cloud computing platform, encode the parameters in the reflection data fusion process into genes, randomly generate an initial population, and each individual represents a parameter combination; define a fitness function for evaluating the quality of each individual; Step 15, according to the fitness function, calculate the fitness value, and select the corresponding individual according to the fitness value to perform crossover and mutation operations to generate a new population, repeat the crossover and mutation operations until the preset number of iterations is reached to obtain the genes of the final individual; decode the genes of the final individual into parameters of the reflection data fusion process; Step 16, calculating the dynamic factor according to the characteristics of the current reflection data and the historical fusion effect; Step 17, dynamically adjust the fusion weights of data from different sources according to the dynamic factors to obtain the final geographic information data; calculate the dynamic factors according to the characteristics of the current reflection data and the historical fusion effect, including: According to the characteristics of the current reflection data and the historical fusion effect, Calculate dynamic factor ; in, is the weight coefficient; Indicates the bandwidth of the current network; Indicates the maximum value of network bandwidth under ideal conditions; represents the adjustment coefficient; represents a constant factor; Indicates the distance from the transmitting source to the receiver; Indicates the reference distance; represents the decay index; ； in, is the effective power of the signal, indicating the strength of the target signal itself; is the received signal strength, which indicates the power of the target signal received at the receiving end; is the adjustment coefficient; is the interference strength, which indicates the power of non-target signals in the signal; Indicates the rate of errors that occur during data transmission; represents the adjustment coefficient; where ； in, represents the adjustment coefficient; Indicates The error of the data source; Indicates the number of data sources; Indicates the index of the error.

2. The method for fusion processing of geographic information data based on cloud computing according to claim 1, characterized in that: Upload the preprocessed reflection data to the cloud computing platform, including: Step 131, evaluating the preprocessed reflection data to obtain an evaluation result, and preliminarily determining a data block size range based on the evaluation result; Step 132, initializing the boundary position of the data block as the starting point, calculating the loss function value according to the current data block boundary, and updating the boundary position of the data block until a preset number of iterations is reached to obtain a determined data block boundary; Step 133, performing final segmentation on the preprocessed reflection data according to the determined data block boundaries to obtain segmented data blocks; Step 134, uniquely mark each segmented data block, create an index file, and record the mark, size and location information of each data block in the original data; Step 135 , uploading the marked data blocks one by one to the cloud computing platform according to the size of the data blocks and the network conditions.

3. The method for fusion processing of geographic information data based on cloud computing according to claim 2 is characterized in that: The preprocessed reflection data is evaluated to obtain an evaluation result. According to the evaluation result, the data block size range is preliminarily determined, including: The preprocessed reflection data is Calculate the evaluation value ,in, Indicates the amount of data transferred each time; Indicates the size of the preprocessed reflection data; Indicates the bandwidth of a network, which refers to the amount of data that the network can transmit per second; Indicates the theoretical upper limit of network bandwidth; Indicates the actual transmission capacity of the current network; represents a fixed delay in the data transmission process; Indicates the physical distance that data travels on a network; is the category in the preprocessed reflection data The probability of occurrence, is the number of different categories in the preprocessed reflection data; represents the weight coefficient; Analyze the evaluation values ​​under different data block sizes , according to the evaluation value Based on the changing trend of the data, the size range of the data block can be preliminarily determined.

4. The method for fusion processing of geographic information data based on cloud computing according to claim 3 is characterized in that: According to the fitness value, the corresponding individuals are selected for crossover and mutation operations to generate a new population. The crossover and mutation operations are repeated until the preset number of iterations is reached to obtain the genes of the final individual, including: According to the fitness value, a part of individuals are selected to enter the next generation using the tournament selection algorithm; For the selected individuals, random pairing is performed and crossover operation is performed through multi-point crossover according to the preset crossover rate to generate new individuals; For the new individuals, mutation operation is performed according to the preset mutation rate to obtain the mutated new individuals; The new individuals and the mutated new individuals are combined into a new population, and a check is made to see whether the preset number of iterations has been reached. If reached, the iteration is stopped and the genes of the final individuals are output.

5. The method for fusion processing of geographic information data based on cloud computing according to claim 4 is characterized in that: Dynamically adjust the fusion weights of data from different sources according to dynamic factors to obtain the final geographic information data, including: For each data source, set an initial fusion weight ; Using dynamic factors pass Calculate the new fusion weights for each data source, where It is The adjusted weights of the data sources; is the adjustment factor corresponding to each data source; According to the new fusion weight of each data source, the geographic information data from different sources are fused to obtain the final geographic information data.

6. The method for fusion processing of geographic information data based on cloud computing according to claim 5 is characterized in that: According to the new fusion weight of each data source, the geographic information data from different sources are fused to obtain the final geographic information data, including: According to the new fusion weight of each data source, Merge geographic information data from different sources to obtain the final geographic information data ; in, Indicates Geographic information data from various data sources; represents the total number of adjusted weights; The index representing the adjusted weight.

7. A geographic information data fusion processing system based on cloud computing, characterized in that: Applied to the method according to any one of claims 1 to 6, comprising: An acquisition module is used to scan the target area using a radio wave measurement device, and acquire the original reflection data of the surface and underground structures by transmitting radio waves and receiving the reflected signals; A preprocessing module, used for preprocessing the original reflection data to obtain preprocessed reflection data; An upload module, used to upload the preprocessed reflection data to the cloud computing platform; The fusion module is used to encode the parameters in the reflection data fusion process into genes on the cloud computing platform, randomly generate the initial population, and each individual represents a parameter combination; define the fitness function used to evaluate the quality of each individual; The iteration module is used to calculate the fitness value according to the fitness function, and select the corresponding individuals according to the fitness value to perform crossover and mutation operations to generate a new population, and repeat the crossover and mutation operations until the preset number of iterations is reached to obtain the genes of the final individual; the genes of the final individual are decoded into parameters of the reflection data fusion process; A calculation module is used to calculate the dynamic factor according to the characteristics of the current reflection data and the historical fusion effect; The adjustment module is used to dynamically adjust the fusion weights of data from different sources according to dynamic factors to obtain the final geographic information data.

8. A computing device, characterized in that include: one or more processors; A storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method as claimed in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program, which, when executed by a processor, implements the method according to any one of claims 1 to 6.

Images