Marine surveying and mapping data fusion quality evaluation and optimization processing system

Abstract

The invention discloses a marine surveying and mapping data fusion quality evaluation and optimization processing system, relates to the technical field of data surveying and mapping, and aims to solve the technical problem of low data display authenticity caused by data abnormality after current marine surveying and mapping data fusion. The regional data optimization module and the partition full-inspection module are connected with an optimized data evaluation module, and the optimized data evaluation module evaluates data optimized by the regional data optimization module and determines an output state based on a qualified state of an evaluation result. The coordinate flotation module is designed to acquire the data surveying and mapping range in the fused data, acquire the coordinate value based on the default coordinate distance, and equally divide and select the area based on the coordinate value, so that the speed of secondary selection of other areas is increased, the purpose of rapid detection is achieved, the problem that the fused data cannot be used due to low authenticity is reduced, and the accuracy of data fusion is improved. And meanwhile, the condition that the detection efficiency is reduced due to full detection is prevented.

Description

Technical Field

[0001] This invention relates to the field of data mapping technology, and more specifically, to a system for quality assessment and optimization of marine mapping data fusion. Background Technology

[0002] Marine surveying data is a collection of data formed by measuring and collecting geographical, geological, and hydrological information about the ocean and its related areas (such as coastlines, seabed, and mid-water). It is an important foundation for marine resource development, marine engineering construction, marine environmental protection, and navigation safety. There are many types of marine surveying data, which require different surveying departments to conduct data surveying.

[0003] After mapping different types of marine data, it is necessary to fuse the data. After fusion, multiple data can be viewed intuitively. However, due to the slight differences between different mapping data, the fused data may have anomalies and cannot be matched, resulting in a decrease in the authenticity of the fused data. In view of this, we propose a marine mapping data fusion quality assessment and optimization system. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the existing technology, adapt to the needs of reality, and provide a marine surveying and mapping data fusion quality assessment and optimization processing system to solve the technical problem of low data display authenticity caused by data anomalies after current marine surveying and mapping data fusion.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a marine mapping data fusion quality assessment and optimization processing system, including a coordinate flotation module, which acquires the fused data range, distributes the data evenly based on default coordinate distance, and simultaneously acquires partitioned data based on the coordinates. The coordinate flotation module is connected to a partitioned full inspection module, which is connected to a regional data optimization module. The regional data optimization module optimizes abnormal data in the partitioned full inspection module and expands the area where abnormal data exists in the partitioned full inspection module, simultaneously optimizing the expanded area data. The output position is selected according to the optimization status. The regional data optimization module and the partitioned full inspection module are connected to an optimized data evaluation module, which evaluates the data optimized by the regional data optimization module and determines the output status based on the qualified status of the evaluation result.

[0006] Preferably, the coordinate flotation module partitions the input fused data, obtains the number of partitions of the fused data based on the adjustable default partition distance, and performs noise removal and benchmark unification on the fused data, including time synchronization, coordinate transformation, spatial adjustment and data standard processing.

[0007] Preferably, the partition full inspection module randomly selects partition data output by the coordinate floating module, and performs a full inspection based on the selected area. The inspection content includes grid unification, feature comparison, extension of uncertain data, and intelligent fusion. Abnormal data and normal data are obtained based on the detection differences.

[0008] Preferably, the optimization of abnormal data by the regional data optimization module includes parameter optimization, data reprocessing, algorithm replacement, and manual intervention. The parameter optimization is based on the fusion algorithm and includes difference data adjustment and weight coefficient adjustment.

[0009] Preferably, the data optimization module expands the partition based on abnormal data, with the expansion direction based on the extension direction of the abnormal data in the partition. The expanded area is 125% of the original partition area, and the data optimization of the expanded area is performed simultaneously with the abnormal data optimization, based on the following formula: ; in, As a base point for diffusion, The topological distance is for a single pass. The coordinates represent the diffusion direction, i.e., the base point of diffusion is obtained. Then, based on the single topological distance Sequential diffusion, and diffusion base point Without changing the topological distance, each diffusion will result in a distance greater than the original topological distance. Add one more time, based on Coordinates, whose diffusion direction extends based on these coordinates, default The coefficient is 1.5.

[0010] Preferably, the expansion status of the data optimization module is determined based on the data optimization status of the expanded area, as follows: If abnormal data still exists after the expanded area is optimized by the data optimization module, the area will continue to be expanded and optimized by 125% while extending in the direction of abnormal data until there is no abnormal data or the optimization ends at the adjacent point of random partition. If the expanded area data is optimized by the data optimization module and there is no abnormal data, it will be output to the optimized data evaluation module.

[0011] Preferably, the evaluation content of the optimized data evaluation module includes internal conformity accuracy evaluation, known accuracy comparison, statistical index evaluation, and data consistency evaluation. The internal conformity accuracy evaluation specifically calculates and analyzes the contradictions within the fused and optimized data itself. The known accuracy comparison specifically compares the optimized data with known high-precision fused data or verification data that did not participate in the fusion. The statistical index evaluation specifically performs a statistical evaluation of the error, standard deviation, and confidence interval in the calculation of the optimized data. The data consistency evaluation specifically evaluates the consistency of the optimized data based on spatial features.

[0012] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention obtains the data mapping range from the fused data by designing a coordinate flotation module, obtains coordinate values ​​based on default coordinate distances, and divides the selected areas equally based on these coordinate values. Simultaneously, a partitioned full inspection module randomly selects partitions for full inspection, reducing the increased detection time caused by overall full inspection. Furthermore, a regional data optimization module optimizes abnormal data while expanding the full inspection of that region, facilitating simultaneous detection of nearby abnormal areas and improving detection efficiency. Simultaneous optimization with abnormal data improves consistency during optimization, reduces optimization steps, and prevents the omission of abnormal areas during selection. Extending data anomalies through a single region eliminates the need for secondary selection of expanded data, increasing the speed of secondary selection of other regions and achieving rapid detection. This reduces the problem of unusable data due to low accuracy of the fused data and prevents reduced detection efficiency caused by full inspection. It solves the technical problem of low data display accuracy caused by data anomalies after marine mapping data fusion.

[0013] 2. This invention also optimizes the data evaluation module by designing it to evaluate the optimized data from multiple perspectives, thereby improving the authenticity and feasibility of the optimized data, reducing the possibility of reduced authenticity due to data optimization, preventing other anomalies caused by the fusion of directly optimized data with existing data, and realizing different output states based on the evaluation state, which facilitates multiple evaluation and optimization tests, improves the overall data integrity, reduces the occurrence of data differences in individual areas, and improves optimization performance. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the system structure of the present invention. Detailed Implementation

[0015] like Figure 1As shown, the present invention relates to a marine mapping data fusion quality assessment and optimization processing system, including a coordinate flotation module. The coordinate flotation module acquires the fused data range, distributes the data evenly based on default coordinate distance, and simultaneously acquires partitioned data based on the coordinates. The coordinate flotation module is connected to a partitioned full inspection module, which is connected to a regional data optimization module. The regional data optimization module optimizes abnormal data in the partitioned full inspection module and expands the area where abnormal data exists in the partitioned full inspection module, simultaneously optimizing the expanded area data. The output position is selected according to the optimization status. The regional data optimization module and the partitioned full inspection module are connected to an optimized data evaluation module, which evaluates the data optimized by the regional data optimization module and determines the output status based on the qualified status of the evaluation result.

[0016] This invention designs a coordinate flotation module to obtain the data mapping range from the fused data, obtains coordinate values ​​based on default coordinate distances, and evenly divides the selected areas based on these coordinate values. Simultaneously, a partitioned full-inspection module randomly selects partitions for full inspection, reducing the increased detection time caused by overall full inspection. Furthermore, a regional data optimization module optimizes abnormal data while simultaneously expanding the full inspection of that region, facilitating synchronous detection of nearby abnormal areas and improving detection efficiency. Simultaneous optimization with abnormal data improves consistency, reduces optimization steps, and prevents the omission of abnormal areas during selection. Extending data anomalies through a single region eliminates the need for secondary selection of expanded data, increasing the speed of secondary selection of other regions and achieving rapid detection. This reduces the problem of unusable data due to low accuracy of the fused data and prevents reduced detection efficiency caused by full inspection. It solves the technical problem of low data display accuracy caused by data anomalies after marine mapping data fusion.

[0017] This invention also optimizes the data evaluation module by designing it to evaluate the optimized data from multiple perspectives, thereby improving the authenticity and feasibility of the optimized data, reducing the possibility of reduced authenticity due to data optimization, preventing other anomalies caused by directly merging optimized data with existing data, and realizing different output states based on the evaluation state, which facilitates multiple evaluation and optimization tests, improves the overall data integrity, reduces the occurrence of data differences in individual regions, and improves optimization performance.

[0018] Specifically, the coordinate flotation module partitions the input fused data, obtains the number of partitions of the fused data based on the adjustable default partition distance, and performs noise removal and benchmark unification on the fused data, including time synchronization, coordinate transformation, spatial adjustment and data standardization processing.

[0019] The data partitioning algorithm operates based on fused data. It divides the fused database into several uniform rectangular partitions by a default partitioning distance, based on the database's range. Noise removal and baseline unification, such as Spikes, are then performed on these partitions. All data is transformed to a unified spatial baseline, such as the WGS-84 coordinate system or UTM projection, converting it into the system's internal standard format for easier subsequent fusion processing and improved efficiency. The fast partitioning algorithm is as follows: ; in, This represents the number of coordinate regions in the default distance partition for this area. The default coordinate point The default distance for the region, based on the default coordinates. and default distance This will give you the number of coordinate regions in that partition. .

[0020] Furthermore, the partitioned full inspection module randomly selects partitioned data output by the coordinate floating module and performs comprehensive inspection on the selected area. The inspection content includes grid unification, feature comparison, extension of uncertain data, and intelligent fusion. Abnormal data and normal data are obtained based on the detection differences.

[0021] Comprehensive detection is performed on randomly selected areas. This random selection method reduces the time by about half compared to direct full inspection, improving detection efficiency. At the same time, it is not affected by missed detections caused by incomplete inspection. Based on the above detection content, the performance of abnormal data detection in this area is effectively improved, and multiple detection methods are convenient for outputting data. Grid unification involves interpolating data to a unified grid (such as DEM / DTM) and then fusing the data using algorithms such as Kriging and natural neighbor methods. The algorithms are as follows: ; in, The distance is The number of samples, For sample attribute values, the mutation function The smaller the output, the greater the distance. The stronger the correlation between points, the better; otherwise, there is no significant correlation. This algorithm can quickly identify the state of fused data and improve grid uniformity.

[0022] Furthermore, the regional data optimization module optimizes abnormal data by including parameter optimization, data reprocessing, algorithm replacement, and manual intervention. Parameter optimization is based on the fusion algorithm and includes adjustment of difference data and adjustment of weight coefficients.

[0023] Parameter optimization involves automatically adjusting the parameters of the fusion algorithm, such as the interpolation radius and weighting coefficients; data reprocessing involves feeding back to the preprocessing stage to reprocess the original data in the problem area, such as re-filtering and adjusting the sound velocity model; algorithm replacement involves triggering other fusion algorithms to re-fuse and optimize the unified area and comparing the optimal solution. If the optimal solution cannot be calculated, manual intervention is used, with professionals making judgments and providing guidance for optimization.

[0024] It is worth noting that the data optimization module expands the partition based on abnormal data. The expansion direction is based on the extension direction of the abnormal data in the partition. The expanded area is 125% of the original partition area. Furthermore, the data optimization of the expanded area is performed simultaneously with the abnormal data optimization, based on the following formula: ; in, As a base point for diffusion, The topological distance is for a single pass. The coordinates represent the diffusion direction, i.e., the base point of diffusion is obtained. Then, based on the single topological distance Sequential diffusion, and diffusion base point Without changing the topological distance, each diffusion will result in a distance greater than the original topological distance. Add one more time, based on Coordinates, whose diffusion direction extends based on these coordinates, default The coefficient is 1.5.

[0025] Based on the default 150% diffusion distance, when optimized abnormal data is detected, the coordinates of the abnormal data are used. Extend the scope to optimize the data selection, reduce the number of regions without abnormal data, improve the targeting of data selection, facilitate the rapid optimization of abnormal data, and improve synchronization.

[0026] It is worth noting that the expansion status of the data optimization module is determined based on the data optimization status of the expanded area, as detailed below: If abnormal data still exists after the expanded area is optimized by the data optimization module, the area will continue to be expanded and optimized by 125% while extending in the direction of abnormal data until there is no abnormal data or the optimization ends at the adjacent point of random partition. If the expanded area data is optimized by the data optimization module and there is no abnormal data, it will be output to the optimized data evaluation module.

[0027] This expansion method can quickly and accurately obtain the data status of abnormal data and its extended areas, and optimize them. Subsequent random selection avoids selecting the same extended area twice, reducing the number of selection areas and also reducing the probability of random selection of adjacent areas. Overall, the selection of areas is optimized, thereby effectively saving data optimization time and improving optimization efficiency.

[0028] It is worth noting that the evaluation content of the optimized data evaluation module includes internal conformity accuracy evaluation, known accuracy comparison, statistical indicator evaluation, and data consistency evaluation. Internal conformity accuracy evaluation specifically involves the calculation and analysis of contradictions in the fused and optimized data itself. Known accuracy comparison specifically involves comparing the optimized data with known high-precision fused data or verification data that did not participate in the fusion. Statistical indicator evaluation specifically involves the statistical evaluation of the error, standard deviation, and confidence interval in the calculation of the optimized data. Data consistency evaluation specifically involves the consistency evaluation of the optimized data based on spatial features.

[0029] The optimized data evaluation module covers four dimensions: internal accuracy evaluation, known accuracy comparison, statistical indicator evaluation, and data consistency evaluation. It evaluates the data from different angles, such as the contradictions in the data itself, the differences with external high-precision data, key statistical features, and spatial feature consistency, forming a complete evaluation system that can comprehensively reflect the quality of optimized data and avoid the one-sidedness that may exist in a single evaluation dimension. Multi-dimensional evaluation can accurately pinpoint potential problems in data optimization, such as internal inconsistencies, deviations from benchmark data, stability of statistical characteristics, and consistency of spatial distribution. This information provides clear direction for improving the data optimization process, helps enhance data quality, and allows data users to fully understand data performance, thus enabling them to use data more rationally for subsequent analysis and decision-making.

[0030] Working Principle: This embodiment provides a marine mapping data fusion quality assessment and optimization processing system. During fusion data processing, the fused data is input into a coordinate flotation module. Based on this data, the coordinate flotation module performs time synchronization, coordinate transformation, spatial adjustment, and data standardization processing, ultimately performing a partitioning operation. The data after these operations is output to a partition full-inspection module. This module performs grid unification, feature comparison, uncertain data extension, and intelligent fusion on the fused data. The processed data's status is then determined. Normal data is directly output to the optimization data evaluation module, while abnormal data is output to the regional data optimization module. Abnormal data in this module undergoes data expansion, and parameters are optimized for both the expanded and abnormal data, including difference data adjustment and weight coefficient adjustment. Simultaneously, data reprocessing and algorithm replacement are performed. If an optimal replacement algorithm cannot be obtained, manual intervention is implemented. Optimized abnormal data is then output back to the partition full-inspection module for a loop. Optimized normal data is output to the optimization data evaluation module, which performs internal accuracy assessment, known accuracy comparison, statistical index assessment, and data consistency assessment to determine data quality. Qualified data is directly output for fusion, while unqualified data is output to the regional data optimization module for a loop.

[0031] The embodiments disclosed in this invention are preferred embodiments, but are not limited thereto. Those skilled in the art can easily understand the spirit of this invention based on the above embodiments and make different extensions and variations, but as long as they do not depart from the spirit of this invention, they are all within the protection scope of this invention.

Claims

1. A system for quality assessment and optimization of marine mapping data fusion, characterized in that, The system includes a coordinate floating module, which acquires the fused data range, distributes the data evenly based on default coordinate distances, and acquires partitioned data based on these coordinates. The coordinate floating module is connected to a partitioned full-inspection module, which is connected to a regional data optimization module. The regional data optimization module optimizes abnormal data in the partitioned full-inspection module and expands the region where abnormal data exists, simultaneously optimizing the expanded region's data. The output position is selected based on the optimization status. The regional data optimization module and the partitioned full-inspection module are connected to an optimized data evaluation module, which evaluates the data optimized by the regional data optimization module and determines the output status based on the evaluation result's pass / fail status.

2. The marine mapping data fusion quality assessment and optimization system according to claim 1, characterized in that, The coordinate floating module partitions the input fused data, obtains the number of partitions of the fused data based on the adjustable default partition distance, and performs noise removal and benchmark unification on the fused data, including time synchronization, coordinate transformation, spatial adjustment and data standard processing.

3. The marine mapping data fusion quality assessment and optimization system according to claim 2, characterized in that, The partition full inspection module randomly selects partition data output by the coordinate floating module and performs a comprehensive inspection based on the selected area. The inspection content includes grid unification, feature comparison, extension of uncertain data, and intelligent fusion. Abnormal data and normal data are obtained based on the detection differences.

4. The marine mapping data fusion quality assessment and optimization system according to claim 1, characterized in that, The optimization of abnormal data by the regional data optimization module includes parameter optimization, data reprocessing, algorithm replacement, and manual intervention. The parameter optimization is based on the fusion algorithm and includes difference data adjustment and weight coefficient adjustment.

5. The marine mapping data fusion quality assessment and optimization system according to claim 4, characterized in that, The data optimization module expands the partition based on abnormal data. The expansion direction is based on the extension direction of the abnormal data in the partition. The expanded area is 125% of the original partition area. The data optimization of the expanded area is performed simultaneously with the abnormal data optimization, based on the following formula: ； in, As a base point for diffusion, The topological distance is for a single pass. The coordinates represent the diffusion direction, i.e., the base point of diffusion is obtained. Then, based on the single topological distance Sequential diffusion, and diffusion base point Without changing the topological distance, each diffusion will result in a distance greater than the original topological distance. Add one more time, based on Coordinates, whose diffusion direction extends based on these coordinates, default The coefficient is 1.

5.

6. The marine mapping data fusion quality assessment and optimization processing system according to claim 5, characterized in that, The expansion status of the data optimization module is determined based on the data optimization status of the expanded region, as follows: If abnormal data still exists after the expanded area is optimized by the data optimization module, the area will continue to be expanded and optimized by 125% while extending in the direction of abnormal data until there is no abnormal data or the optimization ends at the adjacent point of random partition. If the expanded area data is optimized by the data optimization module and there is no abnormal data, it will be output to the optimized data evaluation module.

7. The marine mapping data fusion quality assessment and optimization system according to claim 1, characterized in that, The evaluation content of the optimized data evaluation module includes internal conformity accuracy evaluation, known accuracy comparison, statistical indicator evaluation, and data consistency evaluation. The internal conformity accuracy evaluation specifically calculates and analyzes the contradictions within the fused and optimized data itself. The known accuracy comparison specifically compares the optimized data with known high-precision fused data or verification data that did not participate in the fusion. The statistical indicator evaluation specifically performs a statistical evaluation of the error, standard deviation, and confidence interval in the calculation of the optimized data. The data consistency evaluation specifically evaluates the consistency of the optimized data based on spatial features.

Images