Ship water scale detection method and system based on laser scanning, electronic equipment and storage medium

Abstract

The invention belongs to the field of ship detection, and discloses a ship overwater scale detection method and system based on laser scanning, electronic equipment and a storage medium, and the method comprises the following steps: S1, determining the laser scanning density according to the geometric complexity of the surface of a ship body through self-adaptive adjustment of the laser scanning granularity, performing laser scanning on the to-be-detected ship based on the determined laser scanning density to obtain initial point cloud data of the ship; s2, performing down-sampling processing on the initial point cloud data by using a stereoscopic microscope method, and reserving geometric key feature points on the surface of the ship to obtain a down-sampling point cloud data set; s3, performing frequency domain analysis on the reduced sampling point cloud data set by using fast Fourier transform, and extracting low-frequency geometric information to obtain a simplified geometric model of the ship; s4, performing surface reconstruction on the simplified geometric model by using a k-nearest neighbor method, generating a continuous ship surface grid, and obtaining a surface geometric description; and S5, based on the surface geometric description, obtaining final ship overwater scale data.

Description

Technical Field

[0001] The present invention relates to the technical field of ship detection, and in particular to a method, system, electronic equipment and storage medium for detecting the scale of a ship above water based on laser scanning. Background Art

[0002] Ship scale detection on the water is a key link in ship design, construction, and operation, directly affecting the safety, economy, and seaworthiness of ships. Accurately measuring the length, width, height, and surface profile of a ship not only provides data support for optimizing ship performance, but also plays an important role in port scheduling and waterway planning. However, traditional detection methods such as manual measurement or optical photogrammetry have significant limitations. Manual measurement is inefficient and susceptible to human error, while optical photogrammetry is limited by lighting conditions and environmental interference, making it difficult to obtain high-precision data in complex water surface environments. These shortcomings make it difficult for ship scale detection to meet the needs of the modern shipbuilding industry in terms of accuracy and efficiency.

[0003] In the detection method based on laser scanning, obtaining high-precision geometric information of the ship's surface becomes a core challenge. The point cloud data collection of laser scanning needs to be fine enough to capture the tiny structures and deformations on the hull surface. However, the increase in scanning granularity will significantly increase the amount of data, which in turn places higher requirements on data processing efficiency. Although high-granularity data can improve measurement accuracy, processing complex point cloud data requires stronger computing power and more efficient algorithms. If the granularity is insufficient, key geometric details may be missed, resulting in distorted measurement results. This contradiction between fineness and processing efficiency makes how to optimize the data processing process while ensuring high-precision measurement a technical problem that needs to be solved urgently.

[0004] Therefore, how to adjust the granularity of laser scanning and optimize the data processing method to maintain efficient data processing capabilities while measuring the ship's surface dimensions with high precision becomes a key issue in this study. Summary of the Invention

[0005] In order to solve the above technical problems, the present invention provides a method for detecting the scale of a ship above water based on laser scanning, comprising the following steps:

[0006] S1. Adaptively adjust the laser scanning granularity, determine the laser scanning density based on the geometric complexity of the hull surface, and perform laser scanning on the vessel to be inspected based on the determined laser scanning density to obtain initial point cloud data of the vessel;

[0007] S2. downsampling the initial point cloud data using a stereo microscope method, retaining key geometric feature points on the ship surface, and obtaining a downsampled point cloud dataset;

[0008] S3. Performing frequency domain analysis on the downsampled point cloud dataset using fast Fourier transform to extract low-frequency geometric information to obtain a simplified geometric model of the ship;

[0009] S4. Surface reconstruction of the simplified geometric model is performed using a k-nearest neighbor method to generate a continuous ship surface mesh to obtain a surface geometric description;

[0010] S5. Based on the surface geometry description, obtain the final ship water scale data.

[0011] Preferably, the S1 includes:

[0012] Adaptively adjusting the laser scanning granularity, determining the laser scanning density according to the geometric complexity of the hull surface, and using a laser scanner to perform laser scanning on the ship to be inspected based on the determined laser scanning density to obtain original point cloud data;

[0013] Extracting features from the hull surface of the ship to be detected by a geometric analysis algorithm, determining complex areas on the hull surface, and obtaining a distribution of complex areas;

[0014] According to the complex regions, the point cloud data is segmented by a hierarchical clustering method to determine high-density scanning regions, and obtain segmented regional point cloud data;

[0015] Performing data enhancement on the regional point cloud data by an interpolation method, supplementing the point cloud with respect to a low-density area in the regional point cloud data, and obtaining enhanced point cloud data;

[0016] Coordinate calibration is performed on the enhanced point cloud data to obtain the initial point cloud data of the ship.

[0017] Preferably, the S2 includes:

[0018] Calculating the local curvature of each point in the initial point cloud data using a stereo microscope method, and selecting points whose local curvature is higher than a first preset threshold as geometric feature points;

[0019] Based on the calculated local curvature, the initial point cloud data is gridded using a VoxelGrid algorithm to obtain representative points within each grid to obtain a preliminary downsampled point cloud;

[0020] Comparing the spatial distance between the geometric feature point and the preliminary downsampled point cloud, and retaining the geometric feature point if the distance is less than a second preset threshold, to obtain a downsampled point cloud with retained features;

[0021] Performing neighborhood analysis on the feature-retaining downsampled point cloud to determine the neighborhood density of each point, identify and eliminate isolated points, and obtain an optimized downsampled point cloud;

[0022] According to the normal vector distribution of the optimized down-sampled point cloud, the normal vector change trend is obtained, the geometric key feature points of the ship surface are determined, and the geometric key feature points are merged with the optimized down-sampled point cloud to obtain the down-sampled point cloud dataset.

[0023] Preferably, the S3 includes:

[0024] Performing frequency domain conversion on the downsampled point cloud data using fast Fourier transform to obtain a frequency domain signal;

[0025] If the frequency of the frequency domain signal is lower than a third preset threshold, retaining the corresponding low-frequency component and extracting the low-frequency geometric features;

[0026] constructing an initial geometric model based on the low-frequency geometric features to obtain a preliminary geometric structure, and optimizing the preliminary geometric structure using a geometric smoothing algorithm to generate a smoothed geometric model;

[0027] Key geometric features are extracted from the smoothed geometric model, and the simplified geometric model is constructed based on the key geometric features.

[0028] Preferably, the S4 includes:

[0029] The simplified geometric model is processed using a k-nearest neighbor algorithm to determine a neighborhood point set for each point in the point cloud, thereby obtaining a continuous initial surface topology structure.

[0030] Based on the initial surface topology structure, an initial hull network is generated using a Delaunay triangulation algorithm to obtain an initial meshing result, and local mesh optimization is performed on distorted areas in the initial meshing result to obtain an optimized mesh structure;

[0031] Using a surface continuity evaluation method to obtain continuity parameters of the optimized grid structure, and constructing a continuous hull grid based on the continuity parameters;

[0032] Based on the three-dimensional reconstruction accuracy analysis, the geometric error distribution of the continuous hull grid is obtained, and it is determined whether the error distribution is uniform, so as to obtain the surface geometric description.

[0033] The present invention also provides a ship water scale detection system based on laser scanning, wherein the detection system applies any of the above-mentioned detection methods and comprises: a point cloud acquisition module, a downsampling module, a geometric model construction module, a surface reconstruction module and a data calculation module;

[0034] The point cloud acquisition module adaptively adjusts the laser scanning granularity, determines the laser scanning density according to the geometric complexity of the hull surface, and performs laser scanning on the ship to be inspected based on the determined laser scanning density to obtain initial point cloud data of the ship;

[0035] The downsampling module downsamples the initial point cloud data using a stereo microscope method, retains geometric key feature points on the ship surface, and obtains a downsampled point cloud data set;

[0036] The geometric model building module uses fast Fourier transform to perform frequency domain analysis on the downsampled point cloud data set to extract low-frequency geometric information and obtain a simplified geometric model of the ship;

[0037] The surface reconstruction module uses the k-nearest neighbor method to perform surface reconstruction on the simplified geometric model, generates a continuous ship surface mesh, and obtains a surface geometric description;

[0038] The data calculation module obtains final ship water-scale data based on the surface geometric description.

[0039] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and runnable on the processor, wherein when the processor executes the program, the laser scanning-based ship water scale detection system method described above is implemented.

[0040] The present invention also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed, it implements any of the above-mentioned laser scanning-based ship water scale detection system methods.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] This method uses adaptive laser scanning to acquire initial point cloud data, employs a stereomicroscope algorithm to downsample and retain key feature points, uses a fast Fourier transform to extract low-frequency geometric information to create a simplified model, and uses a k-nearest neighbor algorithm to reconstruct a continuous surface mesh, ultimately obtaining the ship's surface-scale data. This method, combining multiple algorithms, transforms complex point cloud data into a simplified geometric model, effectively extracting key surface features and reconstructing a continuous surface, significantly improving the accuracy and efficiency of ship scale measurements. This method can be widely applied in fields such as ship design, manufacturing, and testing, providing important data support for ship performance evaluation and optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0044] Figure 1Schematic diagram of a method flow in an embodiment of the present invention;

[0045] Figure 2 Schematic diagram of the structure of an electronic device according to an embodiment of the present invention.

[0046] Description of reference numerals:

[0047] 1010 , processor; 1020 , memory; 1030 , input / output interface; 1040 , communication interface; 1050 , bus. DETAILED DESCRIPTION

[0048] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0049] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present disclosure should have the usual meanings understood by people with ordinary skills in the field to which the present disclosure belongs. The "first", "second" and similar words used in the embodiments of the present disclosure do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative position relationships. When the absolute position of the described object changes, the relative position relationship may also change accordingly.

[0050] Example 1

[0051] In this embodiment, if Figure 1 As shown, the ship water scale detection method based on laser scanning includes the following steps:

[0052] S1. Adaptively adjust the laser scanning granularity, determine the laser scanning density according to the geometric complexity of the hull surface, and perform laser scanning on the ship to be inspected based on the determined laser scanning density to obtain the initial point cloud data of the ship.

[0053] S1 includes: adaptively adjusting the laser scanning granularity, determining the laser scanning density according to the geometric complexity of the hull surface, and using a laser scanner to perform laser scanning on the ship to be inspected based on the determined laser scanning density to obtain original point cloud data; extracting features from the hull surface of the ship to be inspected through a geometric analysis algorithm, judging the complex areas on the hull surface, and obtaining the distribution of complex areas; using a hierarchical clustering method to segment the point cloud data according to the complex areas, determining the high-density scanning areas, and obtaining segmented regional point cloud data; enhancing the regional point cloud data through an interpolation method, supplementing the low-density areas in the regional point cloud data with point clouds, and obtaining enhanced point cloud data; and calibrating the coordinates of the enhanced point cloud data to obtain the initial point cloud data of the ship.

[0054] In this embodiment, an adaptive adjustment mechanism determines the scanning granularity based on the geometric complexity of the hull surface. Complex areas, such as the bow or propeller area, have large variations in curvature and require higher-density scanning. The scanning step size is dynamically adjusted by analyzing the curvature differences between adjacent points in the point cloud. For example, the scanning granularity is adjusted to 20 points per square centimeter for the bow area, which has high curvature, while the flat hull side remains at 5 points per square centimeter. This approach improves the efficiency of capturing detail in complex areas and reduces data redundancy. A high-precision laser scanner is then used to scan the hull surface of a cargo ship, generating an initial point cloud dataset containing millions of points at a resolution of approximately 10 points per square centimeter. If the initial point cloud density falls below a preset threshold, for example, fewer than 8 points per square centimeter, a geometric analysis algorithm is used to extract features and identify complex areas. Specifically, the algorithm calculates the point cloud's normal vectors and curvature to identify areas of high curvature, such as hull welds or near portholes, thereby forming a distribution of complex areas. Based on the complex distribution of the area, a hierarchical clustering algorithm was used for regional segmentation to identify high-density scan areas. This generated point cloud data for the segmented areas. Curvature-based spline interpolation was then used to enhance the low-density areas within the point cloud data, increasing the point cloud density to 10 points per square centimeter. This enhancement increased the overall density of the point cloud dataset by approximately 30%, providing greater detail and avoiding holes in subsequent modeling. If the enhanced point cloud dataset met quality thresholds, such as point cloud uniformity and density, a point cloud registration algorithm was used for coordinate calibration, with the calibration error controlled within 0.5 mm, to obtain the initial point cloud data for the vessel.

[0055] S2. Use the stereo microscope method to downsample the initial point cloud data, retain the geometric key feature points of the ship surface, and obtain the downsampled point cloud dataset.

[0056] S2 includes: using a stereo microscope method to calculate the local curvature of each point in the initial point cloud data, and selecting points with local curvature higher than a first preset threshold as geometric feature points; based on the calculated local curvature, using the VoxelGrid algorithm to grid the initial point cloud data, obtaining representative points in each grid, and obtaining a preliminary downsampled point cloud; comparing the spatial distance between the geometric feature points and the preliminary downsampled point cloud, if the distance is less than a second preset threshold, retaining it to obtain a downsampled point cloud with retained features; performing neighborhood analysis on the downsampled point cloud with retained features, judging the neighborhood density of each point, determining and eliminating isolated points, and obtaining an optimized downsampled point cloud; calculating the normal vector distribution of the point cloud based on the optimized downsampled point cloud, obtaining the normal vector change trend, determining the geometric key feature points of the ship surface, and merging the geometric key feature points with the optimized downsampled point cloud to obtain a downsampled point cloud data set.

[0057] In this embodiment, the stereo microscope algorithm consists of two algorithms: a binocular stereo vision algorithm and a neighborhood covariance algorithm. The stereo microscope algorithm first calculates the local curvature of each point. Points with high curvature typically correspond to sharp corners or complex curved surfaces on the hull surface, such as the bow or propeller mounting area. High-curvature points are identified by analyzing the spatial relationship between each point in the point cloud and its neighboring points, calculating the curvature value. If the curvature value of a point exceeds a first preset threshold of 0.05, it is marked as a geometric feature point. Based on the calculated curvature value, the VoxelGrid algorithm is used to grid the initial point cloud data, setting the grid side length to 0.1 meters. The algorithm divides the point cloud into multiple cubic grids, selects a representative point (such as the centroid point) within each grid, and further downsamples it to obtain a preliminary downsampled point cloud. The spatial distance between the geometric feature points and the initial downsampled point cloud is then compared. If the distance is less than a second preset threshold, the point is retained, resulting in a downsampled point cloud with retained features. A neighborhood analysis is then performed on this downsampled point cloud, searching for a neighborhood with a radius of 0.2 meters. If a point has fewer than three points in its neighborhood, it is considered an outlier. These outliers are identified and removed to obtain the optimized downsampled point cloud. Finally, the normal vector distribution of the optimized downsampled point cloud is calculated: the normal vector reflects the surface orientation of a point. Areas of significant variation typically correspond to areas with significant geometric features, such as welds or corners on a ship's hull. When the normal vector angle varies by more than 30 degrees, it is marked as a geometric key feature point. These key points are then merged with the optimized downsampled point cloud to generate the final downsampled point cloud dataset, which contains approximately 495,000 points.

[0058] S3. Use fast Fourier transform to perform frequency domain analysis on the downsampled point cloud dataset, extract low-frequency geometric information, and obtain a simplified geometric model of the ship.

[0059] S3 includes: using fast Fourier transform to perform frequency domain conversion on the downsampled point cloud data to obtain a frequency domain signal; if the frequency of the frequency domain signal is lower than a third preset threshold, retaining the corresponding low-frequency component and extracting the low-frequency geometric features; constructing an initial geometric model based on the low-frequency geometric features to obtain a preliminary geometric structure, optimizing the preliminary geometric structure through a geometric smoothing algorithm to generate a smoothed geometric model; extracting key geometric features from the smoothed geometric model, and constructing a simplified geometric model based on the key geometric features.

[0060] In this embodiment, downsampled point cloud data is transformed into the frequency domain using a fast Fourier transform (FFT). The spatial coordinates of the point cloud are converted into frequency components. A third preset threshold of 0.1 Hz is set, and low-frequency components below this threshold are retained. These components correspond to smooth areas on the hull surface. Low-frequency geometric features of these areas are extracted, and an initial geometric model is constructed based on these low-frequency geometric features to obtain a preliminary geometric structure. This preliminary geometric model may contain significant noise, especially in areas with large curvature variations. This preliminary geometric structure is then optimized using a geometric smoothing algorithm (such as Laplace smoothing) to generate a smoothed geometric model. Subsequently, by analyzing the variation trend of the normal vector, turning points or edge points on the hull surface are identified, key geometric features are obtained, and a simplified geometric model is constructed based on these key geometric features.

[0061] S4. Use the k-nearest neighbor method to reconstruct the surface of the simplified geometric model, generate a continuous ship surface mesh, and obtain a surface geometric description.

[0062] S4 includes: using the k-nearest neighbor algorithm to process the simplified geometric model, determining the neighborhood point set of each point in the point cloud, and obtaining a continuous initial surface topology structure; based on the initial surface topology structure, using the Delaunay triangulation algorithm to generate the initial hull network, obtaining the initial meshing result, and performing local mesh optimization on the distorted area in the initial meshing result to obtain an optimized mesh structure; using the surface continuity evaluation method to obtain the continuity parameters of the optimized mesh structure, and constructing a continuous hull mesh based on the continuity parameters; based on the three-dimensional reconstruction accuracy analysis, obtaining the geometric error distribution of the continuous hull mesh, determining whether the error distribution is uniform, and obtaining a surface geometric description.

[0063] In this embodiment, when determining the neighborhood point set for each point in the point cloud using the k-nearest neighbor algorithm, the k value is selected based on the point cloud density. Since the point cloud density is 10 points per square centimeter, the initial k value is set to 8. The Euclidean distance from each point to its eight nearest neighbors is calculated to generate a neighborhood point set, forming a first surface topology. If the first surface topology contains discontinuous areas, the k value is adjusted through neighborhood point selection and the inter-point distances are recalculated to obtain a continuous initial surface topology. When generating the initial hull mesh using the Delaunay triangulation algorithm, a triangular mesh is generated based on the initial surface topology. Assuming the point cloud covers a hull length of 10 meters, the algorithm generates approximately 5,000 triangular facets, resulting in an initial meshing result. Geometric distortion is detected by examining the changes in the normal vectors of the triangular facets in the initial meshing result. If the normal vector angle in a region exceeds 30 degrees, it is determined to be a distorted area. In this distorted area, the mesh is optimized through local re-meshing or point position adjustment to reduce sharp triangles, improve mesh quality, and obtain an optimized mesh structure. Surface continuity assessment methods are used to obtain the continuity parameters of the optimized mesh structure. Specifically, surface continuity assessment is achieved by calculating the curvature parameters of the optimized mesh structure. The continuity threshold is a curvature change rate of less than 0.1 per centimeter. Mesh areas below this threshold are considered continuous, and a continuous hull mesh is constructed based on the obtained continuity parameters. Based on 3D reconstruction accuracy analysis, the geometric error distribution of the continuous hull mesh is obtained by comparing the point-to-surface distance distribution of the mesh with the original point cloud. The uniformity of the error distribution is determined, and a surface geometry description is obtained.

[0064] S5. Based on the surface geometry description, the final ship water scale data is obtained.

[0065] It should be noted that the method of the embodiments of the present disclosure can be performed by a single device, such as a computer or server. The method of the embodiments of the present disclosure can also be applied in a distributed scenario, where multiple devices cooperate to perform the method. In such a distributed scenario, one of the multiple devices may only perform one or more steps of the method of the embodiments of the present disclosure, and the multiple devices will interact with each other to complete the method.

[0066] It should be noted that the above describes some embodiments of the present disclosure. Other embodiments are within the scope of the appended claims. In some cases, it should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution, and the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention. The actions or steps recorded in the claims can be performed in an order different from that in the above embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-tasking and parallel processing are also possible or may be advantageous.

[0067] Example 2

[0068] In this embodiment, the laser scanning-based ship surface scale detection system includes: a point cloud acquisition module, a downsampling module, a geometric model construction module, a surface reconstruction module and a data calculation module.

[0069] The point cloud acquisition module adaptively adjusts the laser scanning granularity, determines the laser scanning density according to the geometric complexity of the hull surface, and performs laser scanning on the ship to be inspected based on the determined laser scanning density to obtain the initial point cloud data of the ship.

[0070] The workflow of the point cloud acquisition module includes: adaptively adjusting the laser scanning granularity, determining the laser scanning density according to the geometric complexity of the hull surface, and using a laser scanner to perform laser scanning on the ship to be inspected based on the determined laser scanning density to obtain the original point cloud data; extracting features from the hull surface of the ship to be inspected through a geometric analysis algorithm, judging the complex areas on the hull surface, and obtaining the distribution of complex areas; segmenting the point cloud data according to the complex areas using a hierarchical clustering method, determining the high-density scanning areas, and obtaining the segmented regional point cloud data; enhancing the regional point cloud data through the interpolation method, supplementing the low-density areas in the regional point cloud data with point clouds, and obtaining the enhanced point cloud data; and calibrating the coordinates of the enhanced point cloud data to obtain the initial point cloud data of the ship.

[0071] The downsampling module uses the stereo microscope method to downsample the initial point cloud data, retaining the geometric key feature points of the ship surface to obtain a downsampled point cloud dataset.

[0072] The workflow of the downsampling module includes: using the stereo microscope method to calculate the local curvature of each point in the initial point cloud data, and selecting points with local curvature higher than a first preset threshold as geometric feature points; based on the calculated local curvature, the VoxelGrid algorithm is used to grid the initial point cloud data, and representative points in each grid are obtained to obtain a preliminary downsampling point cloud; the spatial distance between the geometric feature points and the preliminary downsampling point cloud is compared, and if the distance is less than a second preset threshold, it is retained to obtain a downsampling point cloud with retained features; neighborhood analysis is performed on the downsampling point cloud with retained features to determine the neighborhood density of each point, identify and eliminate isolated points to obtain an optimized downsampling point cloud; based on the normal vector distribution of the optimized downsampling point cloud, the normal vector change trend is obtained, the geometric key feature points of the ship surface are determined, and the geometric key feature points are merged with the optimized downsampling point cloud to obtain a downsampling point cloud dataset.

[0073] The geometric model building module uses fast Fourier transform to perform frequency domain analysis on the downsampled point cloud dataset, extracts low-frequency geometric information, and obtains a simplified geometric model of the ship.

[0074] The workflow of the geometric model construction module includes: using fast Fourier transform to perform frequency domain conversion on the downsampled point cloud data to obtain a frequency domain signal; if the frequency of the frequency domain signal is lower than a third preset threshold, retaining the corresponding low-frequency component and extracting the low-frequency geometric features; constructing an initial geometric model based on the low-frequency geometric features to obtain a preliminary geometric structure, optimizing the preliminary geometric structure through a geometric smoothing algorithm to generate a smoothed geometric model; extracting key geometric features from the smoothed geometric model, and constructing a simplified geometric model based on the key geometric features.

[0075] The surface reconstruction module uses the k-nearest neighbor method to reconstruct the surface of the simplified geometric model, generate a continuous ship surface mesh, and obtain a surface geometric description.

[0076] The workflow of the surface reconstruction module includes: using the k-nearest neighbor algorithm to process the simplified geometric model, determining the neighborhood point set of each point in the point cloud, and obtaining a continuous initial surface topology structure; based on the initial surface topology structure, using the Delaunay triangulation algorithm to generate the initial hull network, obtaining the initial meshing result, and performing local mesh optimization on the distorted areas in the initial meshing result to obtain an optimized mesh structure; using the surface continuity evaluation method to obtain the continuity parameters of the optimized mesh structure, and constructing a continuous hull mesh based on the continuity parameters; based on the three-dimensional reconstruction accuracy analysis, obtaining the geometric error distribution of the continuous hull mesh, determining whether the error distribution is uniform, and obtaining a surface geometric description.

[0077] The data calculation module obtains the final ship water-scale data based on the surface geometry description.

[0078] The system of the above embodiment is used to implement the corresponding laser scanning-based ship water scale detection method in any of the above embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0079] It should be noted that the above-mentioned laser scanning-based ship water scale detection system is embodied in the form of functional units. The term "module" here can be implemented in the form of software and / or hardware, and is not specifically limited to this.

[0080] For example, a "module" may be a software program, a hardware circuit, or a combination of the two that implements the aforementioned functionality. The hardware circuit may include an application specific integrated circuit (ASIC), an electronic circuit, a processor (e.g., a shared processor, a dedicated processor, or a group of processors) and memory for executing one or more software or firmware programs, combined logic circuits, and / or other suitable components that support the described functionality.

[0081] Example 3

[0082] Based on the same inventive concept, corresponding to any of the above-mentioned embodiments and methods, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and runnable on the processor. When the processor executes the program, it implements the laser scanning-based ship water scale detection method described in any of the above embodiments.

[0083] Figure 2 10 is a schematic diagram showing a more specific hardware structure of an electronic device provided in this embodiment. The device may include: a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040 are communicatively connected to each other within the device via the bus 1050.

[0084] The processor 1010 can be implemented using a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0085] The memory 1020 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage devices, dynamic storage devices, etc. The memory 1020 can store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1020 and is called and executed by the processor 1010.

[0086] The input / output interface 1030 is used to connect input / output modules to implement information input and output. The input / output modules can be configured as components within the device (not shown in the figure) or can be externally connected to the device to provide corresponding functions. Input devices may include a keyboard, mouse, touch screen, microphone, various sensors, etc., and output devices may include a display, speaker, vibrator, indicator light, etc.

[0087] The communication interface 1040 is used to connect to a communication module (not shown) to enable communication between the device and other devices. The communication module can communicate via a wired method (e.g., USB (Universal Serial Bus), network cable, etc.) or a wireless method (e.g., mobile network, WIFI (Wireless Fidelity), Bluetooth, etc.).

[0088] The bus 1050 comprises a path for transmitting information between the various components of the device (eg, the processor 1010 , the memory 1020 , the input / output interface 1030 , and the communication interface 1040 ).

[0089] It should be noted that although the above device only shows the processor 1010, the memory 1020, the input / output interface 1030, the communication interface 1040, and the bus 1050, in a specific implementation, the device may also include other components necessary for normal operation. In addition, it will be understood by those skilled in the art that the above device may only include the components necessary to implement the embodiments of this specification, and does not necessarily include all the components shown in the figure.

[0090] The system of the above embodiment is used to implement the corresponding laser scanning-based ship water scale detection method in any of the above embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0091] Example 4

[0092] Based on the same inventive concept, corresponding to any of the above-mentioned embodiment methods, the present invention also provides a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable the computer to execute the laser scanning-based ship water scale detection method as described in any of the above embodiments.

[0093] The computer-readable media of this embodiment include permanent and non-permanent, removable and non-removable media that can be used to store information by any method or technology. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device.

[0094] The computer instructions stored in the storage medium of the above embodiment are used to enable the computer to execute the ship water scale detection method based on laser scanning as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0095] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present disclosure (including the claims) is limited to these examples. Within the scope of the present disclosure, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the embodiments of the present disclosure as described above, which are not provided in detail for the sake of simplicity.

[0096] In addition, to simplify the description and discussion, and so as not to obscure the embodiments of the present disclosure, known power / ground connections to integrated circuit (IC) chips and other components may or may not be shown in the provided figures. In addition, devices may be shown in the form of block diagrams to avoid obscuring the embodiments of the present disclosure, and this also takes into account the fact that the details of the implementation of these block diagram devices are highly dependent on the platform on which the embodiments of the present disclosure are to be implemented (i.e., these details should be fully within the purview of those skilled in the art). Where specific details (e.g., circuits) are set forth to describe exemplary embodiments of the present disclosure, it will be apparent to those skilled in the art that the embodiments of the present disclosure may be implemented without these specific details or with variations in these specific details. Therefore, these descriptions should be considered illustrative rather than restrictive.

[0097] Although the present disclosure has been described in conjunction with specific embodiments thereof, many alternatives, modifications, and variations of these embodiments will be apparent to those skilled in the art based on the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may use the embodiments discussed.

[0098] Therefore, the units of each example described in the embodiments of this application can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0099] The embodiments of the present disclosure are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present disclosure should be included in the scope of protection of the present disclosure.

Claims

1. A method for detecting ship scale on water based on laser scanning, characterized in that: The following steps are involved: S1. Adaptively adjust the laser scanning granularity, determine the laser scanning density based on the geometric complexity of the hull surface, and perform laser scanning on the vessel to be inspected based on the determined laser scanning density to obtain initial point cloud data of the vessel; S2. downsampling the initial point cloud data using a stereo microscope method, retaining key geometric feature points on the ship surface, and obtaining a downsampled point cloud dataset; S3. Performing frequency domain analysis on the downsampled point cloud dataset using fast Fourier transform to extract low-frequency geometric information to obtain a simplified geometric model of the ship; S4. Surface reconstruction of the simplified geometric model is performed using a k-nearest neighbor method to generate a continuous ship surface mesh to obtain a surface geometric description; S5. Based on the surface geometry description, obtain the final ship water scale data.

2. The method for detecting ship scale above water based on laser scanning according to claim 1, characterized in that: Said S1 comprises: Adaptively adjusting the laser scanning granularity, determining the laser scanning density according to the geometric complexity of the hull surface, and using a laser scanner to perform laser scanning on the ship to be inspected based on the determined laser scanning density to obtain original point cloud data; Extracting features from the hull surface of the ship to be detected by a geometric analysis algorithm, determining complex areas on the hull surface, and obtaining a distribution of complex areas; According to the complex regions, the point cloud data is segmented by a hierarchical clustering method to determine high-density scanning regions, and obtain segmented regional point cloud data; Performing data enhancement on the regional point cloud data by an interpolation method, supplementing the point cloud with respect to a low-density area in the regional point cloud data, and obtaining enhanced point cloud data; Coordinate calibration is performed on the enhanced point cloud data to obtain the initial point cloud data of the ship.

3. The method for detecting ship scale above water based on laser scanning according to claim 1, characterized in that: The S2 includes: Calculating the local curvature of each point in the initial point cloud data using a stereo microscope method, and selecting points whose local curvature is higher than a first preset threshold as geometric feature points; Based on the calculated local curvature, the initial point cloud data is gridded using a VoxelGrid algorithm to obtain representative points within each grid to obtain a preliminary downsampled point cloud; Comparing the spatial distance between the geometric feature point and the preliminary downsampled point cloud, and retaining the geometric feature point if the distance is less than a second preset threshold, to obtain a downsampled point cloud with retained features; Performing neighborhood analysis on the feature-retaining downsampled point cloud to determine the neighborhood density of each point, identify and eliminate isolated points, and obtain an optimized downsampled point cloud; According to the normal vector distribution of the optimized down-sampled point cloud, the normal vector change trend is obtained, the geometric key feature points of the ship surface are determined, and the geometric key feature points are merged with the optimized down-sampled point cloud to obtain the down-sampled point cloud dataset.

4. The method for detecting ship scale above water based on laser scanning according to claim 1, characterized in that: The S3 includes: Performing frequency domain conversion on the downsampled point cloud data using fast Fourier transform to obtain a frequency domain signal; If the frequency of the frequency domain signal is lower than a third preset threshold, retaining the corresponding low-frequency component and extracting the low-frequency geometric features; constructing an initial geometric model based on the low-frequency geometric features to obtain a preliminary geometric structure, and optimizing the preliminary geometric structure using a geometric smoothing algorithm to generate a smoothed geometric model; Key geometric features are extracted from the smoothed geometric model, and the simplified geometric model is constructed based on the key geometric features.

5. The method for detecting ship scale above water based on laser scanning according to claim 1, characterized in that: The S4 includes: The simplified geometric model is processed using a k-nearest neighbor algorithm to determine a neighborhood point set for each point in the point cloud, thereby obtaining a continuous initial surface topology structure. Based on the initial surface topology structure, an initial hull network is generated using a Delaunay triangulation algorithm to obtain an initial meshing result, and local mesh optimization is performed on distorted areas in the initial meshing result to obtain an optimized mesh structure; Using a surface continuity evaluation method to obtain continuity parameters of the optimized grid structure, and constructing a continuous hull grid based on the continuity parameters; Based on the three-dimensional reconstruction accuracy analysis, the geometric error distribution of the continuous hull grid is obtained, and it is determined whether the error distribution is uniform, so as to obtain the surface geometric description.

6. A ship surface scale detection system based on laser scanning, wherein the detection system applies the detection method according to any one of claims 1 to 5, characterized in that: include: Point cloud acquisition module, downsampling module, geometric model building module, surface reconstruction module and data calculation module; The point cloud acquisition module adaptively adjusts the laser scanning granularity, determines the laser scanning density according to the geometric complexity of the hull surface, and performs laser scanning on the ship to be inspected based on the determined laser scanning density to obtain initial point cloud data of the ship; The downsampling module downsamples the initial point cloud data using a stereo microscope method, retains geometric key feature points on the ship surface, and obtains a downsampled point cloud data set; The geometric model building module uses fast Fourier transform to perform frequency domain analysis on the downsampled point cloud data set to extract low-frequency geometric information and obtain a simplified geometric model of the ship; The surface reconstruction module uses the k-nearest neighbor method to perform surface reconstruction on the simplified geometric model, generates a continuous ship surface mesh, and obtains a surface geometric description; The data calculation module obtains final ship water-scale data based on the surface geometric description.

7. An electronic device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for detecting the ship's surface scale based on laser scanning is implemented.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed, the laser scanning-based ship water scale detection system method according to any one of claims 1 to 5 is implemented.

Images