A method and system for ocean topography measurement of multi-source heterogeneous data fusion processing

Abstract

The application relates to a marine topographic survey method and system for multi-source heterogeneous data fusion processing, which comprises the following steps: acquiring marine topographic data of different to-be-measured regions to obtain multi-source heterogeneous data; pre-processing the multi-source heterogeneous data to obtain pre-processed data; extracting features from the pre-processed data, and determining the weight of each data source based on the extracted features; performing fusion processing on the data of the multiple data sources to obtain corresponding fusion results; evaluating the fusion results according to a comprehensive evaluation index, taking a first fusion result that passes the evaluation as target marine topographic data; constructing an iterative optimization function, and iteratively optimizing a second fusion result that fails the evaluation according to the comprehensive evaluation index based on the iterative optimization function until the optimized second fusion result passes the evaluation, and taking the optimized second fusion result that passes the evaluation as the target fusion result. The application can improve the accuracy of marine topographic survey and the data processing effect.

Description

Technical Field

[0001] This invention relates to the field of marine surveying and mapping technology, and in particular to a marine topographic surveying method, system, electronic device, and non-transitory computer-readable storage medium for multi-source heterogeneous data fusion processing. Background Technology

[0002] Currently, ocean topography measurement methods primarily rely on a single data source, such as sonar measurement, satellite remote sensing, or underwater laser scanning. Each of these technologies has its advantages and disadvantages. Sonar measurement can provide high-precision underwater topographic data, but it is easily affected by interference in complex marine environments; satellite remote sensing technology has a wide coverage area, but its accuracy in underwater measurements is relatively low; underwater laser scanning can obtain high-resolution three-dimensional data, but its applicability is limited by the transparency and depth of the water. Therefore, a single technology often cannot comprehensively and accurately reflect the complexity of ocean topography.

[0003] However, existing practices are limited by their heavy reliance on data sources, leading to limitations and uncertainties in the measurement results. In complex marine environments, a single data source often fails to fully capture the subtle changes and diversity of topography. Furthermore, the fusion and processing of different data sources has not received sufficient attention, and there is a lack of effective methods to integrate and utilize data from different technologies, thus affecting the overall accuracy and reliability of marine topographic surveys. Summary of the Invention

[0004] This invention addresses the technical problems existing in the prior art by providing a marine topography measurement method, system, electronic device, and non-transitory computer-readable storage medium that can improve the accuracy of marine topography measurement and the data processing effect through multi-source heterogeneous data fusion processing.

[0005] The technical solution of the present invention to solve the above-mentioned technical problems is as follows:

[0006] This invention provides a method for marine topographic measurement by fusing multi-source heterogeneous data, the method comprising:

[0007] Ocean topography data of different areas to be measured are obtained from multiple data sources to obtain multi-source heterogeneous data;

[0008] The multi-source heterogeneous data is preprocessed to obtain preprocessed data;

[0009] Feature extraction is performed on the preprocessed data, and the weight of each data source is determined based on the extracted features;

[0010] Based on the weight of each data source, the data from multiple data sources are fused to obtain the corresponding fusion result;

[0011] The fusion results are evaluated based on comprehensive evaluation indicators, and the first fusion result that passes the evaluation is taken as the target marine topographic data;

[0012] An iterative optimization function is constructed, and the second fusion result that fails the comprehensive evaluation is iteratively optimized based on the iterative optimization function until the optimized second fusion result passes the evaluation. The optimized second fusion result that passes the evaluation is taken as the target fusion result.

[0013] Optionally, the step of acquiring marine topographic data of different areas to be measured from multiple data sources to obtain multi-source heterogeneous data includes:

[0014] First data from deep-water areas were collected using a multibeam sonar system mounted on an unmanned surface vessel.

[0015] Secondary data of shallow water areas were collected by using a drone equipped with a hyperspectral camera.

[0016] Deploy underwater laser scanners in key areas to obtain localized, high-precision third-party data;

[0017] The first data, the second data, and the third data are obtained to obtain the multi-source heterogeneous data.

[0018] Optionally, the multi-source heterogeneous data is preprocessed to obtain preprocessed data, including:

[0019] Noise in the multi-source heterogeneous data is filtered out using an adaptive wavelet transform method to obtain the first intermediate data.

[0020] Perform geometric and radiometric corrections on the first intermediate data to obtain the second intermediate data;

[0021] The second intermediate data from different data sources are unified to the same coordinate system and resolution to obtain the preprocessed data.

[0022] Optionally, the step of extracting features from the preprocessed data and determining the weight of each data source based on the extracted features includes:

[0023] Extract the water depth data and water turbidity data collected from the data source from the preprocessed data;

[0024] Obtain the data quality factor, first adjustment coefficient, second adjustment coefficient, third adjustment coefficient, and fourth adjustment coefficient corresponding to the data source;

[0025] The weight of the data source is determined based on the data quality factor, the first adjustment coefficient, the second adjustment coefficient, the third adjustment coefficient, and the fourth adjustment coefficient corresponding to the data source, as well as the water depth data and water turbidity data collected by the data source.

[0026] Optionally, the step of fusing data from multiple data sources according to the weight of each data source to obtain a corresponding fusion result includes:

[0027] Acquire the observations, local terrain curvature, and slope for each of the aforementioned data sources;

[0028] The observations, local terrain curvature, and slope of each data source are adjusted according to the first and second smoothing factors to obtain the fusion result.

[0029] Optionally, the fusion function is expressed as:

[0030]

[0031] Where F(x,y,z) is the result of the fusion function, U i Let H be the observation value from the i-th data source, H be the local terrain curvature, S be the slope, and μ and ρ be the first and second smoothing factors, respectively.

[0032] Optionally, the step of evaluating the fusion result based on comprehensive evaluation indicators, and using the first fusion result that passes the evaluation as the target marine topographic data, includes:

[0033] Obtain the fusion result corresponding to each sampling point of each of the data sources, and the true value corresponding to the fusion result;

[0034] Obtain the data redundancy of the fusion result;

[0035] The terrain gradient in the fusion result is processed according to the first correction coefficient, and the data redundancy is processed according to the second correction coefficient. The corresponding comprehensive evaluation index is determined by combining the fusion result and the true value.

[0036] If the comprehensive evaluation index of the fusion result is greater than a preset threshold, then the fusion result is determined as the first fusion result that has passed the evaluation.

[0037] Optionally, the comprehensive evaluation index is expressed as:

[0038]

[0039] Where E is the comprehensive evaluation index, and P i T is the fusion result corresponding to the i-th sampling point of each data source. i It is the true value of the fusion result of the i-th sampling point. R is the terrain gradient, R is the data redundancy, ∈ is the first correction factor, θ is the second correction factor, and n is the total number of sampling points.

[0040] Optionally, the iterative optimization of the comprehensive evaluation index for the second fusion result that failed the evaluation based on the iterative optimization function includes:

[0041] Obtain the preset learning rate, convergence control parameters, and the Laplace operator for the optimization variables in the current iteration round;

[0042] Obtain the first fusion function value of the current iteration round and the second fusion function value of the previous iteration round;

[0043] Based on the learning rate, the convergence control parameters, and the Laplacian operator, the first fusion function value and the second fusion function value are iteratively optimized to obtain the optimized second fusion result.

[0044] The present invention also provides a marine topographic surveying system for multi-source heterogeneous data fusion processing, the system comprising:

[0045] The data acquisition module is used to acquire marine topographic data of different areas to be measured from multiple data sources, resulting in multi-source heterogeneous data;

[0046] The data processing module is used to preprocess the multi-source heterogeneous data to obtain preprocessed data;

[0047] The weight determination module is used to extract features from the preprocessed data and determine the weight of each data source based on the extracted features.

[0048] The data fusion module is used to fuse data from multiple data sources according to the weight of each data source to obtain the corresponding fusion result;

[0049] The first processing module is used to evaluate the fusion result according to the comprehensive evaluation index, and to use the first fusion result that passes the evaluation as the target marine topography data;

[0050] The second processing module is used to construct an iterative optimization function, and based on the iterative optimization function, iteratively optimize the second fusion result that fails the comprehensive evaluation index until the optimized second fusion result passes the evaluation, and then take the optimized second fusion result that passes the evaluation as the target fusion result.

[0051] Furthermore, to achieve the above objectives, the present invention also proposes an electronic device, comprising: a memory for storing computer software programs; and a processor for reading and executing the computer software programs, thereby realizing the marine topographic measurement method for multi-source heterogeneous data fusion processing as described above.

[0052] Furthermore, to achieve the above objectives, the present invention also proposes a non-transitory computer-readable storage medium storing a computer software program, which, when executed by a processor, implements a marine topographic measurement method for multi-source heterogeneous data fusion processing as described above.

[0053] The beneficial effects of this invention are:

[0054] (1) This invention employs three data acquisition methods: multibeam sonar, hyperspectral remote sensing, and underwater laser scanning. Through the coordinated operation of unmanned surface vessels, drones, and laser scanners, it achieves comprehensive and detailed coverage of deep water, shallow water, and local areas. Compared with traditional single-source oceanographic surveys, multi-source data acquisition makes measurements more comprehensive in different marine environments, ensuring the integrity and accuracy of topographic data for each region.

[0055] (2) This invention achieves high-precision multi-source data integration by defining weighting coefficients and fusion functions to effectively weight and fuse data from multiple sources. Dynamically adjusting weights based on factors such as water depth and turbidity rationalizes the data contribution values ​​from different regions, helping to reduce data discrepancies and improve the accuracy of the fusion results.

[0056] (3) This invention evaluates the accuracy of the fusion results through a comprehensive evaluation index and makes dynamic corrections based on factors such as data redundancy and terrain gradient. Compared with static measurement methods, this adaptive optimization mechanism significantly enhances the adaptability to data complexity and regional differences, achieves a dynamic balance of accurate measurement, and can adapt to the ever-changing marine environment.

[0057] In summary, this invention achieves a better balance between wide-area coverage and local precision through collaborative acquisition and adaptive fusion of multi-source data, which can significantly improve the overall accuracy and efficiency of marine topographic surveying, making it a preferred solution for conducting refined topographic surveying in complex marine environments. Attached Figure Description

[0058] Figure 1 A scene diagram illustrating a marine topographic measurement method for multi-source heterogeneous data fusion processing provided by this invention;

[0059] Figure 2 A flowchart of a marine topographic measurement method for multi-source heterogeneous data fusion processing provided by the present invention;

[0060] Figure 3 A schematic diagram of the structure of a marine topographic surveying system for multi-source heterogeneous data fusion processing provided by the present invention;

[0061] Figure 4A schematic diagram of the hardware structure of a possible electronic device provided by the present invention;

[0062] Figure 5 This is a schematic diagram of the hardware structure of a possible computer-readable storage medium provided by the present invention. Detailed Implementation

[0063] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0064] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0065] In the description of this invention, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed herein.

[0066] Please see Figure 1 , Figure 1 This is a scene diagram illustrating a marine topographic measurement method for multi-source heterogeneous data fusion processing provided by the present invention. Figure 1 As shown, the terminal and server are connected via a network, such as a wired or wireless network. The terminal can include, but is not limited to, portable devices such as mobile phones and tablets with various network platform applications installed, as well as fixed terminals such as computers, kiosks, and advertising machines. The server provides users with various business services, including service push servers and user recommendation servers.

[0067] It should be noted that, Figure 1The scenario diagram illustrating a marine topographic measurement method for multi-source heterogeneous data fusion processing is merely an example. The terminals, servers, and application scenarios described in this embodiment of the invention are for the purpose of more clearly illustrating the technical solutions of this embodiment and do not constitute a limitation on the technical solutions provided by this embodiment. As those skilled in the art will know, with the evolution of systems and the emergence of new business scenarios, the technical solutions provided by this embodiment of the invention are also applicable to similar technical problems.

[0068] The terminal can be used for:

[0069] Ocean topography data of different areas to be measured are obtained from multiple data sources to obtain multi-source heterogeneous data;

[0070] The multi-source heterogeneous data is preprocessed to obtain preprocessed data;

[0071] Feature extraction is performed on the preprocessed data, and the weight of each data source is determined based on the extracted features;

[0072] Based on the weight of each data source, the data from multiple data sources are fused to obtain the corresponding fusion result;

[0073] The fusion results are evaluated based on comprehensive evaluation indicators, and the first fusion result that passes the evaluation is taken as the target marine topographic data;

[0074] An iterative optimization function is constructed, and the second fusion result that fails the comprehensive evaluation is iteratively optimized based on the iterative optimization function until the optimized second fusion result passes the evaluation. The optimized second fusion result that passes the evaluation is taken as the target fusion result.

[0075] Please see Figure 2 The present invention provides a flowchart of a marine topographic measurement method for multi-source heterogeneous data fusion processing, comprising the following steps:

[0076] Step 201: Obtain marine topographic data of different areas to be tested from multiple data sources to obtain multi-source heterogeneous data.

[0077] In some embodiments, step 201 may include:

[0078] First data from deep-water areas were collected using a multibeam sonar system mounted on an unmanned surface vessel.

[0079] Secondary data of shallow water areas were collected by using a drone equipped with a hyperspectral camera.

[0080] Deploy underwater laser scanners in key areas to obtain localized, high-precision third-party data;

[0081] The first data, the second data, and the third data are obtained to obtain the multi-source heterogeneous data.

[0082] Multibeam sonar systems are detection devices suitable for deep-water environments, measuring underwater topography by emitting and receiving multiple sound beams. They possess strong penetrating power, capable of traversing deep water layers to provide accurate topographic depth data. Unmanned surface vessels (USVs) carrying multibeam sonar can operate autonomously in deep water, reducing the need for manual operation and making them particularly suitable for continuous and large-scale topographic mapping tasks in deep water areas. The primary data includes topographic depth and contour data of deep-water areas, suitable for large-scale deep-water topographic modeling.

[0083] Among these, hyperspectral cameras possess rich spectral information acquisition capabilities, enabling them to distinguish the reflection characteristics of different substances in shallow underwater layers. In shallow water areas, optical data has good penetrating power, allowing hyperspectral cameras to collect detailed information on shallow underwater topography and substance distribution. Unmanned aerial vehicles (UAVs) offer high flexibility, enabling close-range flight in shallow water areas to acquire high-resolution data on shallow underwater topography and water quality. This method is suitable for shallow water areas with complex terrain, reducing optical distortion and interference through low-altitude flight. The second set of data includes surface topography, substance distribution information, and spectral characteristics of the shallow water area, suitable for detailed shallow water topographic mapping and environmental analysis.

[0084] Among them, the underwater laser scanner is a high-precision detection device capable of acquiring microscopic topographic data of local underwater areas. It is typically used for high-precision data acquisition of specific areas, such as measuring seabed structure, rock faults, or sediment characteristics. Third-party data can include high-precision topographic data of local areas, suitable for detailed analysis and 3D reconstruction, and can be used for further processing and model validation.

[0085] It is understandable that the first, second, and third data acquired through the aforementioned three devices form multi-source heterogeneous data. This data includes marine topographic information at different depths (deep water, shallow water) and with varying levels of precision (large-scale, medium precision, local high precision). Through weighted calculations and data fusion models, this multi-source heterogeneous data can overcome the limitations of a single data source, improving the comprehensiveness and accuracy of topographic surveying. This process makes the topographic information from deep water to shallow water, and then to specific key areas, more complete, contributing to the establishment of accurate and hierarchical marine topographic models.

[0086] This invention, through the above three acquisition methods and multi-source data fusion processing, can obtain complete and accurate marine topographic data, providing solid data support for refined topographic surveying and data modeling in complex sea areas.

[0087] Step 202: Preprocess the multi-source heterogeneous data to obtain preprocessed data.

[0088] In some embodiments, step 202 may include:

[0089] Noise in the multi-source heterogeneous data is filtered out using an adaptive wavelet transform method to obtain the first intermediate data.

[0090] Perform geometric and radiometric corrections on the first intermediate data to obtain the second intermediate data;

[0091] The second intermediate data from different data sources are unified to the same coordinate system and resolution to obtain the preprocessed data.

[0092] Wavelet transform is a commonly used signal processing method suitable for noise removal. Adaptive wavelet transform dynamically adjusts filtering parameters based on data characteristics to effectively remove random noise and interference signals from the data. Multi-source heterogeneous data (such as sonar data, hyperspectral data, and laser scanning data) are affected by environmental interference and equipment errors during acquisition. Denoising through wavelet transform can improve data clarity and make data features more prominent. The first intermediate data obtained after denoising is measurement data with significantly reduced noise, laying the foundation for subsequent geometric and radiometric corrections.

[0093] Geometric correction eliminates spatial distortions in measurement data caused by changes in viewpoint or platform attitude, ensuring the data's geometry matches the actual terrain. This process uses mathematical models to restore the data's true location, guaranteeing spatial accuracy. Radiometric correction adjusts the spectral distortion of optical data (such as hyperspectral data) under different lighting conditions, reflecting its actual spectral intensity and distribution. This step ensures consistent spectral characteristics of optical data across various environments. The second intermediate data, obtained after geometric and radiometric corrections, is more accurate in both spatial and radiometric terms, providing a foundation for unified processing of different data sources.

[0094] In practice, different data sources may use different coordinate systems (such as WGS84, UTM, etc.). Transforming all data to a unified coordinate system allows the data to be overlaid and used on the same geographic space. Different data sources may have different spatial resolutions (for example, sonar data may be coarse, while hyperspectral and laser scanning data may be finer). Unifying the data sources to the same resolution ensures consistency in the spatial distribution of the data, thus avoiding scale inconsistencies during fusion processing. After unifying the coordinate system and resolution, the second intermediate data from all data sources form a standardized, preprocessed set of data, laying the foundation for the fusion of multi-source data.

[0095] In summary, this invention successfully eliminates noise and spatial inconsistencies in multi-source data through preprocessing steps such as adaptive wavelet denoising, geometric and radiometric correction, and coordinate system and resolution unification, ensuring the spatial consistency and accuracy of the data. The preprocessed data more accurately reflects the terrain features of different regions, facilitating the smooth progress of subsequent data fusion and thus improving the accuracy and reliability of the final terrain model.

[0096] Step 203: Extract features from the preprocessed data and determine the weight of each data source based on the extracted features.

[0097] In some embodiments, step 203 may include:

[0098] Extract the water depth data and water turbidity data collected from the data source from the preprocessed data;

[0099] Obtain the data quality factor, first adjustment coefficient, second adjustment coefficient, third adjustment coefficient, and fourth adjustment coefficient corresponding to the data source;

[0100] The weight of the data source is determined based on the data quality factor, the first adjustment coefficient, the second adjustment coefficient, the third adjustment coefficient, and the fourth adjustment coefficient corresponding to the data source, as well as the water depth data and water turbidity data collected by the data source.

[0101] In some embodiments, the weight of the data source can be represented as:

[0102] W=α·exp(-β·D)·(1-γ·T)·(1+λ·Q);

[0103] Where W is the weight of the data source, D represents the water depth, T represents the water turbidity, Q represents the data quality factor, and α, β, γ, and λ are the first, second, third, and fourth adjustment coefficients, respectively.

[0104] In practice, W represents the proportion or importance of a specific data source in the multi-source fusion process. A larger W value indicates that the data source has a higher weight in the fusion model, meaning that the data source has a greater impact on the final measurement results. The weighting coefficients are adjusted based on multiple factors such as water depth, turbidity, and data quality to ensure the rationality and accuracy of the data sources under different environments.

[0105] In exp(-β·D), the greater the water depth, the higher the likelihood of the data source being affected by environmental factors, especially for optical and sonar data, which suffer severe data attenuation in deep water. Therefore, this term is adjusted using a negative exponential function. As the water depth D increases, the value of exp(-β·D) gradually decreases, thus gradually reducing the weight of data in deep water regions.

[0106] β adjusts the sensitivity of water depth to weight decay. A larger β value accelerates weight decay and is suitable for unstable measurements in deep water; a smaller β value is suitable for shallow water and makes weight decay more slow.

[0107] In (1-γ·T), water turbidity affects the clarity of the data source, especially for optical imaging. Higher turbidity results in lower underwater visibility and reduced data reliability. Therefore, the weights are linearly decayed using (1-γ·T) to ensure lower data weights in turbid water. As T increases, this term decreases, reducing the data weights.

[0108] The adjustment coefficient γ is used to adjust the degree to which water turbidity affects the data weights. A larger γ value makes the weights more sensitive to turbidity, which is suitable for shallow water measurement environments where high clarity is required.

[0109] In 1+λ·Q, the data quality factor Q represents the objective quality or reliability of the data source, which may be determined by factors such as equipment precision and environmental adaptability. A higher quality factor indicates more reliable data, and its weight increases accordingly. Linear amplification is achieved through 1+λ·Q, thus increasing the weight of high-quality data.

[0110] λ is used to control the influence of the data quality factor on the weights. A larger λ value enhances the role of high-quality data and is suitable for measurement scenarios that require strict data quality assurance.

[0111] α is a global adjustment coefficient used to standardize and unify the overall level of weights. It is typically set according to actual measurement needs to adjust the range of weights. Adjusting the value of α changes the overall level of weights but does not affect the relative effects of the D, T, and Q factors.

[0112] This formula combines three key parameters: water depth, turbidity, and data quality. It takes into account both the environmental adaptability of each data source and the actual data quality. Through exponential decay and linear adjustment, it ensures that each factor has a reasonable impact under different circumstances.

[0113] When the measurement environment changes (such as switching from shallow water to deep water), the weighting coefficients are automatically adjusted to reduce the weight of optical data in deep water and increase the influence of high-quality data, thereby improving the accuracy of the fusion results.

[0114] By adjusting the comprehensive weighting coefficients, we can ensure that the weight allocation of different data sources is reasonable, optimize the overall fusion effect of multi-source data, and adapt to the complex and ever-changing marine environment measurement needs.

[0115] If a data source measures deep water (large D value), has low water turbidity (small T value), and has a high equipment quality factor Q, then the weight W of this data source will decrease due to the exp(-β·D) term, but will be amplified by the (1+λ·Q) term, thus achieving a balance in the weights.

[0116] Conversely, in shallow water areas with high turbidity and average data quality, the weight W will be reduced by the superposition of multiple factors to avoid the negative impact of the data source on the fusion result.

[0117] In summary, this invention enables effective screening and integration of multi-source data, resulting in high reliability and adaptability of the final fusion results in different marine environments.

[0118] Step 204: Based on the weight of each data source, perform data fusion processing on the data from multiple data sources to obtain the corresponding fusion result.

[0119] In some embodiments, step 204 may include:

[0120] Acquire the observations, local terrain curvature, and slope for each of the aforementioned data sources;

[0121] The observations, local terrain curvature, and slope of each data source are adjusted according to the first and second smoothing factors to obtain the fusion result.

[0122] In some embodiments, the fusion function is represented as:

[0123]

[0124] Where F(x, y, z) is the result of the fusion function, U i Let H be the observation value from the i-th data source, H be the local terrain curvature, S be the slope, and μ and ρ be the first and second smoothing factors, respectively.

[0125] In the specific implementation, F(x, y, z) is the fused terrain result, representing the terrain value at a given point (x, y, z). By weighted averaging of multi-source data and incorporating terrain smoothing control and slope information, this result can represent the true terrain features.

[0126] W i It is the weight coefficient of the i-th data source, representing the importance of each data source.

[0127] U i It is the observation value of the i-th data source, which in this scenario is usually the ocean topography measurement value of a certain location by that data source.

[0128] μ and ρ are smoothing factors used to control the smoothness of the fusion results and the influence of slope, respectively adjusting the weights of terrain curvature and slope in the results.

[0129] This is the second-order gradient of the local terrain (i.e., the Laplace operator), used to measure the curvature of the terrain. This term reflects the curvature or undulation of the current terrain surface.

[0130] S represents slope information, indicating the degree of terrain inclination, which affects the intensity of the smoothing process.

[0131] This section implements a weighted average for different data sources. This is achieved by using the observations U from each data source. i By weight (W) i A weighted average is used to ensure that the contribution of different data sources to the final fused value is proportional to their respective weights. The significance of a weighted average is that high-quality data (high weight) contributes more to the final result, while noisier data (low weight) has a smaller impact. This weighting mechanism can improve the accuracy of data fusion and reduce errors caused by low-quality data.

[0132] Smoothing control item This item is used to smooth the weighted average result, so that the fused terrain model is more reasonable in terms of continuity and smoothness.

[0133] Local topographic curvature It reflects the undulation of the terrain surface (e.g., ridges or valleys). The greater the curvature, the more dramatic the terrain changes. By multiplying by the curvature term, areas with large terrain undulations can be smoothed more effectively.

[0134] The slope attenuation term exp(-ρ·S) is used. A larger slope S indicates a more drastic change in terrain. In this case, the intensity of the smoothing process needs to be reduced to preserve detail. Therefore, the smoothness is attenuated by exp(-ρ·S), causing the smoothing factor to decrease as the slope increases. This ensures that over-smoothing is not performed in areas with steep slopes, thus preserving subtle terrain features.

[0135] The smoothing factor μ controls the overall impact of the smoothing term. A larger μ value increases the intensity of the smoothing process, resulting in a smoother terrain; while a smaller μ value reduces the smoothing intensity, more accurately reflecting the changes in the original terrain data.

[0136] The overall purpose of this formula is to generate a comprehensive terrain model based on observations from different data sources. This model considers the weighted contributions of multiple data sources while also achieving good continuity and naturalness through smoothing. Its key mechanisms include:

[0137] Weighted fusion: By using a weighted average, we ensure that different data sources have different impacts on the results based on their quality and applicability, thereby improving the overall reliability of the fused data.

[0138] Adaptive smoothing: Introducing smoothing processing with curvature and slope control, the terrain model remains clear in areas with drastic slope changes, while providing a natural transition in flat areas, avoiding discontinuities or abrupt changes in the model.

[0139] In marine topographic surveying, if multiple data sources (such as sonar data, hyperspectral data, and laser data) provide different observations U at a certain location... i The formula will be based on the weight W i A weighted average is used to ensure accurate fusion.

[0140] In areas with steep slopes or dramatic changes in terrain, the smoothing term... It plays a controlling role, keeping the data smooth without losing detail.

[0141] The fusion function of this invention achieves a balance between data weighting and smoothing control, effectively integrating multi-source data while adaptively adjusting for terrain features to generate an accurate model that conforms to the actual terrain characteristics. By controlling the influence of each term, the formula provides an efficient and accurate data fusion method for complex marine terrain.

[0142] Step 205: Evaluate the fusion results according to the comprehensive evaluation indicators, and take the first fusion result that passes the evaluation as the target marine topography data.

[0143] In some embodiments, step 205 may include:

[0144] Obtain the fusion result corresponding to each sampling point of each of the data sources, and the true value corresponding to the fusion result;

[0145] Obtain the data redundancy of the fusion result;

[0146] The terrain gradient in the fusion result is processed according to the first correction coefficient, and the data redundancy is processed according to the second correction coefficient. The corresponding comprehensive evaluation index is determined by combining the fusion result and the true value.

[0147] If the comprehensive evaluation index of the fusion result is greater than a preset threshold, then the fusion result is determined as the first fusion result that has passed the evaluation.

[0148] In some embodiments, the comprehensive evaluation index is expressed as:

[0149]

[0150] Where E is the comprehensive evaluation index, and P i T is the fusion result corresponding to the i-th sampling point of each data source. i It is the true value of the fusion result of the i-th sampling point. R is the terrain gradient, R is the data redundancy, ∈ is the first correction factor, θ is the second correction factor, and n is the total number of sampling points.

[0151] In practice, E is used as a comprehensive evaluation metric to measure the overall accuracy and reliability of the fusion results. A smaller E value indicates that the error between the fusion result and the true value is small, and the reliability is high.

[0152] P i It is the fusion result of each data source at the i-th sampling point, that is, the measurement value calculated by the fusion model.

[0153] T i It is the true value of the i-th sampling point, that is, the reference value obtained in the actual measurement.

[0154] This is the terrain gradient, used to reflect the rate of change of the terrain surface. Generally, a larger value means drastic terrain changes, while a smaller value indicates relatively gentle terrain.

[0155] R stands for data redundancy, which refers to the amount of redundant data in the fusion result. A higher redundancy means that data from the same area is sampled repeatedly.

[0156] ∈ is the first correction factor, used to adjust the impact of terrain gradient on the evaluation index.

[0157] θ is the second correction factor, used to adjust the impact of data redundancy on the evaluation index.

[0158] This term is the root mean square error (RMSE), used to evaluate the fusion result P. i Compared with the true value T i The deviation between them. By taking the square root of the average of the sum of squared errors of all sampling points, RMSE can reflect the average level of error well.

[0159] The RMSE term represents a basic fusion accuracy assessment, measuring the bias of the fusion model's output data. A smaller value indicates that the fused data is closer to the true value. RMSE is highly sensitive, giving higher weight to points with larger errors, which helps identify regions of abnormal error.

[0160] Terrain gradient correction term According to terrain gradient The RMSE results were corrected. A larger terrain gradient indicates more drastic changes in the terrain surface, which may increase the difficulty of data fusion. Increasing the value of the evaluation indicator E reflects that in regions of rapid change, the impact of error on the results is greater.

[0161] The correction factor ∈ is used to control the intensity of the influence of terrain gradient on evaluation metrics. A larger ∈ value will increase the correction effect of terrain gradient, making it more suitable for scenarios that require strict control of complex terrain.

[0162] In areas with dramatic topographic changes, the E value is increased to emphasize the accuracy requirements of the fusion results in steep or complex terrain, ensuring that the measurement data adapts to complex terrain.

[0163] In the redundancy attenuation term exp(-θ·R), the data redundancy R represents the data repetition. Higher redundancy typically indicates multiple acquisitions of the same terrain data, which can improve the stability of the fusion results. Redundancy is attenuated using a negative exponential function exp(-θ·R). As redundancy increases, this term approaches 0, helping to reduce the E value and thus reflecting the stability brought by high redundancy. The correction coefficient θ controls the degree of influence of redundancy on the evaluation index. A larger θ value causes redundancy to decay more quickly, making the E value decrease more significantly when there is a lot of redundant data. This term encourages data redundancy and increases data stability, especially suitable for data scenarios with a large amount of multi-source repeated acquisitions, making the fusion results more reliable even with increased redundancy.

[0164] This comprehensive evaluation index E integrates multiple factors to fully assess the applicability of the fusion results in terms of accuracy, terrain complexity, and data stability. Specifically: The root mean square error (RMSE) term assesses the basic accuracy of the data fusion. A terrain gradient correction term further tightens the accuracy requirements for complex terrain, reducing accuracy loss due to drastic terrain changes. A data redundancy correction term emphasizes data stability and reliability, resulting in higher accuracy for highly redundant data sources, as indicated by lower evaluation index values ​​in the fusion results.

[0165] If a region has complex terrain with high redundancy, then the terrain gradient... The terrain gradient correction term is relatively large, and the redundancy R is also high. In this case, the terrain gradient correction term... It will increase the E value, but due to the effect of the redundancy attenuation term exp(-θ·R), the overall E value may not increase significantly, thus making the assessment of this region more reflective of the actual accuracy.

[0166] In regions with flat terrain and low redundancy, the gradient and redundancy terms have a smaller impact on E, and the evaluation of the fusion results is mainly dominated by the root mean square error term, making the evaluation index reflect the basic level of accuracy.

[0167] The evaluation index E of this invention combines error accuracy, terrain change, and data redundancy to construct a multi-factor adaptive evaluation method, which makes the fusion results more reflective of the actual accuracy and applicability under different terrain conditions, and helps to identify the differences in reliability and data fusion effects in different regions.

[0168] Step 206: Construct an iterative optimization function, and based on the iterative optimization function, iteratively optimize the second fusion result that fails the comprehensive evaluation index until the optimized second fusion result passes the evaluation. The optimized second fusion result that passes the evaluation is taken as the target fusion result.

[0169] In some embodiments, step 206 may include:

[0170] Obtain the preset learning rate, convergence control parameters, and the Laplace operator for the optimization variables in the current iteration round;

[0171] Obtain the first fusion function value of the current iteration round and the second fusion function value of the previous iteration round;

[0172] Based on the learning rate, the convergence control parameters, and the Laplacian operator, the first fusion function value and the second fusion function value are iteratively optimized to obtain the optimized second fusion result.

[0173] In some embodiments, the iterative optimization function can be expressed as:

[0174]

[0175] Where O(k+1) is the optimization variable of the iterative optimization function after the (k+1)th iteration, η is the learning rate, F(k) represents the first fusion function value of the current iteration, F(k-1) represents the second fusion function value of the previous iteration, and -φ is the convergence control parameter. Let O(k) be the Laplace operator.

[0176] In the implementation, O(k) is the optimization variable for the k-th iteration, representing the fusion optimization value of the current iteration. O(k+1) is the optimization variable after the (k+1)-th iteration, used to calculate the optimization result for the next iteration. η is the learning rate or step size factor, controlling the size of the update in each iteration. F(k) and F(k-1) represent the fusion function values ​​of the current and previous iterations, respectively, used to measure the change after the current data fusion. φ is the convergence control parameter, used to control the influence of the smoothing term on the iteration update step size. It is an O(k) Laplacian operator, namely the second gradient (or curvature) of the optimized value of the current terrain data, used to measure the smoothness of the current data.

[0177] In some embodiments, F(k)-F(k-1) reflects the magnitude of change in the current iterative fusion value. Specifically: when the difference between F(k) and F(k-1) is large, it indicates that the current data fusion state still has considerable room for improvement compared to the ideal state, thus requiring a larger adjustment to the O value. When the difference between F(k) and F(k-1) is small, it indicates that the data fusion is gradually approaching stability, and the update amount should be gradually reduced to converge to an optimal solution.

[0178] This option is used to smooth the update magnitude: The larger the absolute value of , the greater the curvature of the current optimization result, indicating significant discontinuities or drastic changes in the data fusion result. In this case, the value of the exponential decay term will approach zero, thereby reducing the update amplitude and avoiding unstable fluctuations. Correspondingly, when When the value is small, it indicates that the current optimization result is relatively smooth. At this time, the decay effect is small, which can maintain a large update step size and improve the convergence speed.

[0179] The learning rate η controls the basic size of the step size in each iteration. It plays an overall regulatory role. If the value is too large, the optimization may oscillate or diverge; if the value is too small, it may lead to slow convergence and increase computational cost.

[0180] The core purpose of this formula is to gradually adjust the optimization objective based on the previous iteration's optimized value O(k), combined with the changes in the data fusion model's output F and the smoothness of the current optimized value, to make the fused data results more accurate and smooth. Its implementation process includes the following stages:

[0181] In the initial stage, when the iteration just begins, the difference term F(k)-F(k-1) is large, and O(k) is not yet fully smooth. Therefore, the formula updates with a large step size, quickly approaching the target value. As the iteration progresses, the curvature of O(k) gradually decreases, the smoothness increases, and the exponential decay term gradually increases, causing the adjustment step size to gradually decrease, achieving fine optimization. When F(k) and F(k-1) are almost equal and When the value is small, the update amount approaches zero, the optimization process gradually converges, and finally a smooth and accurate fusion result is obtained.

[0182] In summary, by introducing the control of difference terms and the Laplace operator, this invention balances the accuracy and convergence speed of optimization, making the fusion results of multi-source heterogeneous data more stable and accurate, and is particularly suitable for marine measurement scenarios with complex topographic changes.

[0183] Please see Figure 3 , Figure 3This is a schematic diagram of the structure of a marine topographic surveying system for multi-source heterogeneous data fusion processing provided by the present invention.

[0184] like Figure 3 As shown in the figure, the marine topographic surveying system for multi-source heterogeneous data fusion processing proposed in this embodiment of the invention includes:

[0185] Data acquisition module 301 is used to acquire marine topographic data of different areas to be measured from multiple data sources to obtain multi-source heterogeneous data;

[0186] Data processing module 302 is used to preprocess the multi-source heterogeneous data to obtain preprocessed data;

[0187] The weight determination module 303 is used to extract features from the preprocessed data and determine the weight of each data source based on the extracted features.

[0188] The data fusion module 304 is used to fuse data from multiple data sources according to the weight of each data source to obtain a corresponding fusion result;

[0189] The first processing module 305 is used to evaluate the fusion result according to the comprehensive evaluation index, and to use the first fusion result that passes the evaluation as the target marine topography data.

[0190] The second processing module 306 is used to construct an iterative optimization function, and based on the iterative optimization function, iteratively optimize the second fusion result that fails the comprehensive evaluation index until the optimized second fusion result passes the evaluation, and take the optimized second fusion result that passes the evaluation as the target fusion result.

[0191] Please see Figure 4 , Figure 4 This is a schematic diagram illustrating an embodiment of the electronic device provided in this invention. For example... Figure 4 As shown, an embodiment of the present invention provides an electronic device 400, including a memory 410, a processor 420, and a computer program 411 stored in the memory 410 and executable on the processor 420. When the processor 420 executes the computer program 411, it performs the following steps:

[0192] Ocean topography data of different areas to be measured are obtained from multiple data sources to obtain multi-source heterogeneous data;

[0193] The multi-source heterogeneous data is preprocessed to obtain preprocessed data;

[0194] Feature extraction is performed on the preprocessed data, and the weight of each data source is determined based on the extracted features;

[0195] Based on the weight of each data source, the data from multiple data sources are fused to obtain the corresponding fusion result;

[0196] The fusion results are evaluated based on comprehensive evaluation indicators, and the first fusion result that passes the evaluation is taken as the target marine topographic data;

[0197] An iterative optimization function is constructed, and the second fusion result that fails the comprehensive evaluation is iteratively optimized based on the iterative optimization function until the optimized second fusion result passes the evaluation. The optimized second fusion result that passes the evaluation is taken as the target fusion result.

[0198] Please see Figure 5 , Figure 5 This is a schematic diagram illustrating an embodiment of a computer-readable storage medium provided by an embodiment of the present invention. For example... Figure 5 As shown, this embodiment provides a computer-readable storage medium 500 on which a computer program 411 is stored. When the computer program 411 is executed by a processor, it performs the following steps:

[0199] Ocean topography data of different areas to be measured are obtained from multiple data sources to obtain multi-source heterogeneous data;

[0200] The multi-source heterogeneous data is preprocessed to obtain preprocessed data;

[0201] Feature extraction is performed on the preprocessed data, and the weight of each data source is determined based on the extracted features;

[0202] Based on the weight of each data source, the data from multiple data sources are fused to obtain the corresponding fusion result;

[0203] The fusion results are evaluated based on comprehensive evaluation indicators, and the first fusion result that passes the evaluation is taken as the target marine topographic data;

[0204] An iterative optimization function is constructed, and the second fusion result that fails the comprehensive evaluation is iteratively optimized based on the iterative optimization function until the optimized second fusion result passes the evaluation. The optimized second fusion result that passes the evaluation is taken as the target fusion result.

[0205] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0206] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0207] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A system that specifies functions in one or more boxes.

[0208] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including an instruction set implemented in a process. Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0209] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0210] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0211] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for marine topographic surveying that integrates and processes multi-source heterogeneous data, characterized in that, The method includes: Ocean topography data of different areas to be measured are obtained from multiple data sources to obtain multi-source heterogeneous data; The multi-source heterogeneous data is preprocessed to obtain preprocessed data; Feature extraction is performed on the preprocessed data, and the weight of each data source is determined based on the extracted features; Based on the weight of each data source, the data from multiple data sources are fused to obtain a corresponding fusion result, including: acquiring the observation values, local terrain curvature, and slope of each data source; adjusting the observation values, local terrain curvature, and slope of each data source according to a first smoothing factor and a second smoothing factor to obtain the fusion result; the fusion function is expressed as: ;in, It is the result of the fusion function. For the observation value of the i-th data source, S represents the local topographic curvature, and S represents the slope. , These are the first smoothing factor and the second smoothing factor, respectively. It is the weight coefficient of the i-th data source; The fusion results are evaluated based on comprehensive evaluation indicators, and the first fusion result that passes the evaluation is taken as the target marine topographic data; An iterative optimization function is constructed, and the second fusion result that fails the comprehensive evaluation is iteratively optimized based on the iterative optimization function until the optimized second fusion result passes the evaluation. The optimized second fusion result that passes the evaluation is taken as the target fusion result.

2. The marine topographic surveying method for multi-source heterogeneous data fusion processing according to claim 1, characterized in that, The process involves acquiring marine topographic data from multiple data sources for different areas under test, resulting in multi-source heterogeneous data, including: First data from deep-water areas were collected using a multibeam sonar system mounted on an unmanned surface vessel. Secondary data was collected from shallow water areas using a drone equipped with a hyperspectral camera. Deploy underwater laser scanners in key areas to obtain localized, high-precision third-party data; The first data, the second data, and the third data are obtained to obtain the multi-source heterogeneous data.

3. The marine topographic surveying method for multi-source heterogeneous data fusion processing according to claim 2, characterized in that, The multi-source heterogeneous data is preprocessed to obtain preprocessed data, including: Noise in the multi-source heterogeneous data is filtered out using an adaptive wavelet transform method to obtain the first intermediate data. Perform geometric and radiometric corrections on the first intermediate data to obtain the second intermediate data; The second intermediate data from different data sources are unified to the same coordinate system and resolution to obtain the preprocessed data.

4. The marine topographic surveying method for multi-source heterogeneous data fusion processing according to claim 3, characterized in that, The step of extracting features from the preprocessed data and determining the weight of each data source based on the extracted features includes: The water depth data and water turbidity data collected from the data source are extracted from the preprocessed data; Obtain the data quality factor, first adjustment coefficient, second adjustment coefficient, third adjustment coefficient, and fourth adjustment coefficient corresponding to the data source; The weight of the data source is determined based on the data quality factor, the first adjustment coefficient, the second adjustment coefficient, the third adjustment coefficient, and the fourth adjustment coefficient corresponding to the data source, as well as the water depth data and water turbidity data collected by the data source.

5. The marine topographic surveying method for multi-source heterogeneous data fusion processing according to claim 1, characterized in that, The evaluation of the fusion results based on comprehensive evaluation indicators, and the selection of the first fusion result that passes the evaluation as the target marine topography data, includes: Obtain the fusion result corresponding to each sampling point of each of the data sources, and the true value corresponding to the fusion result; Obtain the data redundancy of the fusion result; The terrain gradient in the fusion result is processed according to the first correction coefficient, and the data redundancy is processed according to the second correction coefficient. The corresponding comprehensive evaluation index is determined by combining the fusion result and the true value. If the comprehensive evaluation index of the fusion result is less than a preset threshold, then the fusion result is determined as the first fusion result that has passed the evaluation.

6. The marine topographic surveying method for multi-source heterogeneous data fusion processing according to claim 1, characterized in that, The comprehensive evaluation indicators are expressed as follows: Where E is the comprehensive evaluation index, It is the fusion result corresponding to the i-th sampling point of each data source. It is the true value of the fusion result of the i-th sampling point. R is the terrain gradient, and R is the data redundancy. It is the first correction factor. This is the second correction factor, and n is the total number of sampling points.

7. The marine topographic surveying method for multi-source heterogeneous data fusion processing according to claim 6, characterized in that, The iterative optimization of the comprehensive evaluation index for the second fusion result that failed the evaluation based on the iterative optimization function includes: Obtain the preset learning rate, convergence control parameters, and the Laplace operator for the optimization variables in the current iteration round; Obtain the first fusion function value of the current iteration round and the second fusion function value of the previous iteration round; Based on the learning rate, the convergence control parameters, and the Laplacian operator, the first fusion function value and the second fusion function value are iteratively optimized to obtain the optimized second fusion result.

8. A marine topographic surveying system for multi-source heterogeneous data fusion processing, used to perform the marine topographic surveying method for multi-source heterogeneous data fusion processing as described in any one of claims 1-7, characterized in that, The system includes: The data acquisition module is used to acquire marine topographic data of different areas to be measured from multiple data sources, resulting in multi-source heterogeneous data; The data processing module is used to preprocess the multi-source heterogeneous data to obtain preprocessed data; The weight determination module is used to extract features from the preprocessed data and determine the weight of each data source based on the extracted features. The data fusion module is used to fuse data from multiple data sources according to the weight of each data source to obtain a corresponding fusion result. It is also used to acquire the observation values, local terrain curvature, and slope of each data source; and to adjust the observation values, local terrain curvature, and slope of each data source according to a first smoothing factor and a second smoothing factor to obtain the fusion result. The fusion function is expressed as: ;in, It is the result of the fusion function. For the observation value of the i-th data source, S represents the local topographic curvature, and S represents the slope. , These are the first smoothing factor and the second smoothing factor, respectively. It is the weight coefficient of the i-th data source; The first processing module is used to evaluate the fusion result according to the comprehensive evaluation index, and to use the first fusion result that passes the evaluation as the target marine topography data; The second processing module is used to construct an iterative optimization function, and based on the iterative optimization function, iteratively optimize the second fusion result that fails the comprehensive evaluation index until the optimized second fusion result passes the evaluation, and then take the optimized second fusion result that passes the evaluation as the target fusion result.

Images