A ship body damage detection method and device, electronic equipment and storage medium

Abstract

The application provides a ship body damage detection method and device, electronic equipment and storage medium, the ship body damage detection method comprises the following steps: acquiring acoustic emission detection data and visual image data; based on the space-time characteristics of the acoustic emission detection data and visual image data, the acoustic emission detection data and the visual image data are associated; based on the acoustic emission detection data and the corresponding visual image data respectively, the acoustic emission feature damage coefficient and the corresponding visual feature damage coefficient are obtained; and based on the acoustic emission feature damage coefficient and the visual feature damage coefficient, the risk level of the ship body damage is determined. The application simultaneously acquires acoustic emission detection data and visual image data, and performs space-time association on the two, judges the risk level of the ship body damage based on the acoustic emission feature damage coefficient and the corresponding visual feature damage coefficient, and improves the integrity and comprehensiveness of the ship body damage information acquisition.

Description

Technical Field

[0001] This invention relates to the field of ship structural health monitoring technology, specifically to a method, device, electronic equipment, and storage medium for detecting ship hull damage. Background Technology

[0002] Ships endure complex marine environmental conditions during long-term service, such as wave loads, wind pressure, temperature variations, and seawater corrosion. These factors can easily lead to defects in the hull structure, such as cracks, corrosion pits, and fatigue damage. Failure to detect and address these defects in a timely manner will directly impact the ship's navigation safety and structural lifespan. Therefore, establishing a real-time monitoring and damage early warning mechanism for hull structures has become an important research direction in the field of ship maintenance. Currently, hull structural health monitoring mainly relies on non-destructive testing (NDT) technology, with commonly used methods including ultrasonic testing, magnetic particle testing, eddy current testing, acoustic emission testing, and visual inspection. Among these, acoustic emission testing, by collecting transient elastic wave signals generated when the structure is under load or cracks propagate, enables dynamic monitoring of the location and time of damage, and is one of the most actively used online monitoring methods. Existing visual inspection methods mainly rely on surface image features to identify cracks.

[0003] For example, Chinese patent CN 118150706 A discloses a damage detection method, device, electronic device, and medium based on acoustic emission. In the sensor detection module, a finite element model of the wellhead device is established to determine the propagation law of the damage acoustic emission signal. Sensors are arranged according to the propagation law, and the damage acoustic emission signal of the wellhead device is collected through the sensors. In the data acquisition and processing module, the damage acoustic emission signal is preprocessed to obtain the characteristic data of the damage acoustic emission signal.

[0004] In existing research and engineering practice, acoustic emission detection and visual inspection are usually operated as independent systems. Acoustic signals reflect the dynamic acoustic events of crack occurrence, while visual images record static surface morphology. The two differ in data dimensions, temporal response, and spatial correspondence, resulting in scattered detection results and making it difficult to achieve complementary verification. There is a problem that a single signal source cannot fully reflect the damage state, which can easily lead to false detections or missed detections. Summary of the Invention

[0005] The purpose of this invention is to overcome the above-mentioned technical deficiencies and propose a method for detecting ship hull damage, thereby solving the technical problem in the prior art that a single signal source is difficult to fully reflect the damage state, which can easily lead to false detection or missed detection.

[0006] To achieve the above-mentioned technical objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a method for detecting ship hull damage, comprising the following steps: Acquire acoustic emission detection data and visual image data; Based on the spatiotemporal characteristics of the acoustic emission detection data and the visual image data, the acoustic emission detection data and the visual image data are associated; Based on the acoustic emission detection data and the corresponding visual image data, the acoustic emission feature impairment coefficient and the corresponding visual feature impairment coefficient are obtained respectively; and Based on the acoustic emission characteristic damage coefficient and the visual characteristic damage coefficient, the risk level of hull damage is determined.

[0007] In some embodiments, the method for obtaining the acoustic emission characteristic damage coefficient is as follows: Extract the acoustic emission event count, average amplitude, signal energy, and count rate from the acoustic emission detection data; The acoustic emission characteristic damage coefficient is calculated based on the acoustic emission event count, average amplitude, signal energy, and count rate.

[0008] In some embodiments, the formula for calculating the acoustic emission characteristic damage coefficient is: ; Where N is the acoustic emission event count, A is the average amplitude, E is the signal energy, C is the count rate, and w1, w2, w3, and w4 are the acoustic emission characteristic weighting coefficients.

[0009] In some embodiments, the method for obtaining the visual feature impairment coefficient is as follows: The visual image data is input into a pre-trained visual detection model to obtain visual detection results, wherein the visual detection results include at least crack length, crack width, and crack detection confidence. The visual feature damage coefficient is calculated based on the crack length, crack width, and crack detection confidence level.

[0010] In some embodiments, the formula for calculating the visual feature impairment coefficient is: ; in, The length of the crack. The width of the crack; For crack detection confidence level, These are the visual feature weighting coefficients.

[0011] In some embodiments, determining the risk level of hull damage based on the acoustic emission characteristic damage coefficient and the visual characteristic damage coefficient includes: Based on the acoustic emission feature damage coefficient and the visual feature damage coefficient, a preset weighted fusion model is used to calculate the fusion damage index. Based on the fusion damage index and the preset risk level classification strategy, the risk level of the hull damage is determined.

[0012] In some embodiments, the formula for calculating the fusion damage index is: ; in, , These are the weighting coefficients. The characteristic damage coefficient of acoustic emission. The visual feature impairment coefficient; The preset risk level classification method is as follows: When the fusion damage index is within the threshold range of 0-0.25, it is classified as a safe level where no obvious damage is detected. When the fusion damage index is in the threshold range of 0.25-0.50, it is classified as a low-risk level with the presence of microcracks. When the fusion damage index is in the threshold range of 0.50-0.75, it is classified as a medium risk level with a tendency for crack propagation. When the fusion damage index is within the threshold range of 0.75-1, it is classified as a high-risk level with significant crack propagation.

[0013] Secondly, the present invention also provides a hull damage detection device, comprising: The damage detection module is used to acquire acoustic emission detection data and visual image data; The damage feature association module is used to associate the acoustic emission detection data and the visual image data based on the spatiotemporal features of the acoustic emission detection data and the visual image data; The damage result output module is used to obtain the acoustic emission detection result and the corresponding visual detection result based on the acoustic emission detection data and the corresponding visual image data, respectively; and The damage level assessment module is used to determine the risk level of hull damage based on the acoustic emission detection results and visual inspection results.

[0014] Thirdly, the present invention also provides an electronic device, comprising: a processor and a memory; The memory stores computer programs that can be executed by the processor; When the processor executes the computer program, it implements the steps in the hull damage detection method as described in any of the above.

[0015] Fourthly, the present invention also provides a computer-readable storage medium storing computer program instructions, which, when executed by a processor, implement the hull damage detection method as described in any of the above claims.

[0016] Compared with existing technologies, the hull damage detection method provided by this invention improves the completeness and comprehensiveness of hull damage information acquisition by simultaneously acquiring acoustic emission detection data and visual image data, performing spatiotemporal correlation between the two, and determining the risk level of hull damage based on the acoustic emission characteristic damage coefficient and the corresponding visual characteristic damage coefficient. Attached Figure Description

[0017] Figure 1 This is a flowchart of the hull damage detection method provided in the embodiments of the present invention; Figure 2 This is a flowchart of the method for obtaining acoustic emission characteristic damage coefficients provided in an embodiment of the present invention; Figure 3 This is a flowchart of the method for obtaining visual feature damage coefficients provided in an embodiment of the present invention; Figure 4 This is a flowchart of the fusion analysis method provided in the embodiments of the present invention; Figure 5 This is a flowchart of another embodiment of the hull damage detection method provided in this invention; Figure 6 This is a schematic block diagram of the hull damage detection device provided in an embodiment of the present invention; Figure 7 This is a schematic block diagram of the electronic device provided in the embodiments of the present invention; Figure 8 This is a flowchart of the visual inspection process provided in an embodiment of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0019] To address the technical problem that a single signal source cannot fully reflect the damage state, easily leading to false detections or missed detections, this invention provides a hull damage detection method. This method can simultaneously acquire acoustic emission signals and hull surface image information, and perform joint analysis on the two. This allows the detection results to simultaneously include internal hull damage activity characteristics and surface crack morphology characteristics, improving the completeness and comprehensiveness of hull damage information acquisition and reducing the risk of false detections or missed detections.

[0020] Please see Figure 1 This invention provides a method for detecting ship hull damage, which includes the following steps: S101. Acquire acoustic emission detection data and visual image data; S102. Based on the spatiotemporal characteristics of the acoustic emission detection data and the visual image data, associate the acoustic emission detection data and the visual image data; S103. Based on the acoustic emission detection data and the corresponding visual image data, respectively, obtain the acoustic emission feature impairment coefficient and the corresponding visual feature impairment coefficient; and S104. Based on the acoustic emission characteristic damage coefficient and the visual characteristic damage coefficient, determine the risk level of hull damage.

[0021] The implementation process of each step is illustrated below as an example.

[0022] Step S101: Acquire acoustic emission detection data and visual image data.

[0023] For example, acoustic emission (AE) detection is first used to sense damage activity within the ship's structure, acquiring AE detection data. AE sensors are deployed in critical load-bearing components or areas prone to fatigue damage. The acoustic matching layer of the AE sensors is made of a polymer composite material with controllable thickness. Its acoustic impedance is between that of the ship's metallic material and the piezoelectric sensitive element, used to reduce sound wave reflection loss at the interface. The thickness of the acoustic matching layer is designed according to the main frequency range of the acoustic emission signal to meet an approximate quarter-wavelength matching condition, thereby improving the transmission efficiency of sound waves in the target frequency band. AE detection data is then acquired based on the AE sensors.

[0024] For example, visual inspection is used to simultaneously capture cracks on the hull surface and acquire visual image data. Specifically, an intelligent image acquisition robot can be used to acquire images of the hull surface. Based on the hull surface material and working environment conditions, a suitable adhesion method is selected between the robot and the hull surface, allowing it to move on the hull surface and capture images of the cracks using a camera mounted on it. The camera uses a ring light source.

[0025] S102. Based on the spatiotemporal characteristics of the acoustic emission detection data and the visual image data, associate the acoustic emission detection data and the visual image data.

[0026] For example, the acoustic emission detection data and visual image data are first synchronized in time, and the feature parameters in the acoustic emission detection data acquired within the same monitoring time window are associated with the hull surface image in the visual image data acquired at the corresponding time. Then, the acoustic emission detection data is used as the trigger information for damage activity. When the feature parameters show abnormal changes in a certain monitoring area, the visual image data of the corresponding area is retrieved first to verify whether there are cracks and the crack morphology in that area.

[0027] S103. Based on the acoustic emission detection data and the corresponding visual image data, respectively, obtain the acoustic emission feature damage coefficient and the corresponding visual feature damage coefficient.

[0028] For example, acoustic emission event counts, average amplitude, signal energy, and count rate are extracted from the acquired acoustic emission signal. By normalizing the above characteristic parameters, an acoustic emission damage index is constructed, and the acoustic emission characteristic damage coefficient is calculated.

[0029] For example, visual image data is captured by an image acquisition robot and input into a crack detection model based on YOLOv5 network structure and combined with CBAM attention mechanism for analysis. The model outputs visual image results, i.e., whether there are cracks and hull cracks. The visual image results include at least crack length, crack width and crack detection confidence. By normalizing the above feature parameters, a visual feature damage index is constructed and the visual feature damage coefficient is calculated.

[0030] S104. Based on the acoustic emission characteristic damage coefficient and the visual characteristic damage coefficient, determine the risk level of hull damage.

[0031] For example, based on the acoustic emission characteristic damage coefficient and the visual characteristic damage coefficient, a weighted fusion model is used to perform fusion analysis to determine the risk level of hull damage. A preset weighted fusion model is used to calculate the fusion damage index; based on the fusion damage index and a preset risk level classification strategy, the risk level of the hull damage is determined.

[0032] Understandably, acoustic emission characteristics are primarily used to reflect the intensity and development trend of damage activity, while visual characteristics are used to characterize the spatial location, size, and morphological changes of cracks on the hull surface. By fusing multi-source information, visual detection results can be verified by combining the changing trends of acoustic emission signals when cracks are still in their early stages, thereby reducing the probability of false positives and false negatives. Based on the determined risk level of hull damage, operators can more easily prevent and control crack development.

[0033] In one embodiment, to obtain the acoustic emission characteristic damage coefficient, please refer to... Figure 2 The method for obtaining the acoustic emission characteristic damage coefficient is as follows: S201. Extract the acoustic emission event count, average amplitude, signal energy, and count rate from the acoustic emission detection data; S202. Based on the acoustic emission event count, average amplitude, signal energy, and count rate, calculate the acoustic emission characteristic damage coefficient.

[0034] Furthermore, the formula for calculating the acoustic emission characteristic damage coefficient is as follows: ; Among them, acoustic emission event count , which is the number of acoustic emission events detected per unit time; average amplitude , where is the average peak amplitude of the acoustic emission signal; Signal energy , is the cumulative energy value of the acoustic emission signal; Count rate , which is the number of rings counted per unit time; and, , , These are the acoustic emission characteristic weighting coefficients. Based on experimental statistics... =0.30, =0.25, =0.30, =0.15.

[0035] Acoustic emission detection data can be collected by acoustic emission sensors installed on the hull. These sensors are deployed on critical load-bearing components or in areas prone to fatigue damage to collect transient elastic wave signals generated by crack initiation, propagation, and localized plastic deformation during hull operation or loading. After necessary amplification, filtering, and digitization processing, acoustic parameters characterizing damage activity are extracted, including but not limited to signal energy, amplitude, and duration.

[0036] Understandably, acoustic emission signal energy reflects the magnitude of elastic energy released during crack propagation; higher energy values ​​indicate more intense damage activity within the material per unit time. Signal amplitude reflects the intensity of the acoustic emission source and can be used to distinguish between micro-crack activity and macro-crack propagation events. Simultaneously, by statistically analyzing the frequency and trend of acoustic emission events per unit time, the evolution of damage activity from sporadic to continuous can be characterized. When acoustic emission energy and event count rate show a continuous upward trend, it indicates that cracks within the hull structure are in an accelerated propagation stage; when relevant characteristic parameters tend to stabilize or decrease, it indicates that damage activity has temporarily slowed down. Time-series analysis of the above acoustic emission characteristics can reflect the intensity level and evolution trend of damage activity within the hull structure.

[0037] It should be noted that in the acoustic emission detection step, the acoustic emission characteristic parameters used to characterize the damage activity inside the hull structure can be selected and combined according to actual needs, and the arrangement of acoustic emission sensors and signal processing methods can also be adjusted accordingly. The purpose of all these is to obtain acoustic information reflecting the evolution process of damage inside the hull.

[0038] In one embodiment, to obtain the visual feature impairment coefficient, please refer to... Figure 3 The method for obtaining the visual feature impairment coefficient is as follows: S301. Input the visual image data into a pre-trained visual detection model to obtain visual detection results, wherein the visual detection results include at least crack length, crack width, and crack detection confidence. S302. Based on the crack length, crack width, and crack detection confidence, calculate the visual feature damage coefficient.

[0039] The formula for calculating the visual feature impairment coefficient is as follows: ; in, The length of the crack; The crack width; Confidence level for crack detection. These are the visual feature weighting coefficients. Based on experimental statistics... , .

[0040] The visual image data can be acquired by existing, mature image acquisition robots. These robots can move stably on the ship's surface and complete image acquisition tasks under different posture conditions. Based on the ship's surface material and working environment conditions, a suitable adsorption method is selected to ensure the robot's stability and safety. The robot utilizes existing mature equipment; correspondingly, different models of robots with different structures that move on the ship's hull can be used, employing vacuum adsorption, magnetic adsorption, biomimetic adsorption, and negative pressure adsorption. No single limitation or elaborated description is provided here.

[0041] Understandably, when acoustic emission detection data is abnormal, the system can control an intelligent image acquisition robot to move to the corresponding detection area and acquire images of the hull surface. The acquired images are input into a crack detection model based on a YOLOv5 network structure combined with a CBAM attention mechanism for analysis. The model outputs the hull crack detection results. Crack identification and crack feature extraction are performed on the preprocessed hull surface images. Damage crack features include visual characteristics of the spatial location, size, and distribution on the hull surface. Crack features are used to characterize the intuitive morphology of hull surface damage.

[0042] For details, please refer to Figure 8 The visual inspection process is as follows: First, the camera is initialized and its parameters are set. Then, the crack image information is captured by the industrial camera and the image information is preprocessed. The image to be inspected is then detected by the deployed deep learning model to determine whether there is a crack and output the detection result. This process of acquisition to result output is repeated.

[0043] It should be noted that in visual crack detection, the crack recognition model is not limited to a specific network structure, and the image acquisition method can also be replaced according to the application scenario. As long as effective image information reflecting the crack morphology on the hull surface can be obtained, it is an alternative implementation of the present invention.

[0044] It should be noted that existing ship damage detection methods, primarily based on visual inspection, typically only detect damage when cracks have expanded to a certain size and become clearly identifiable to the naked eye or through imaging equipment, easily missing the crack initiation and early propagation stages. This invention utilizes the high sensitivity of acoustic emission detection to the activity of micro-cracks within materials, capturing abnormal acoustic emission signals before cracks become apparent on the hull surface, and combining this with visual inspection results to confirm the damage status. Compared to existing technologies, this invention effectively advances the detection time of hull damage, improving early warning capabilities. To improve the accuracy of crack identification, the acquired hull surface images undergo preprocessing. This process includes image denoising, brightness equalization, and contrast enhancement to reduce the interference of complex backgrounds and ambient lighting on crack identification. The processing steps are existing mature technologies and will not be elaborated upon here.

[0045] In one embodiment, please refer to Figure 4 The risk level of hull damage is determined based on the acoustic emission characteristic damage coefficient and the visual characteristic damage coefficient. A weighted fusion model is used to perform a fusion analysis of the acoustic emission detection results and the visual detection results, including: S401. Based on the acoustic emission feature damage coefficient and the visual feature damage coefficient, a preset weighted fusion model is used to calculate the fusion damage index. S402. Based on the fusion damage index and the preset risk level classification strategy, determine the risk level of the hull damage.

[0046] In this embodiment, the formula for calculating the fusion damage index is: ; in, , This represents the weighting coefficient. In practical engineering, cracks often first occur inside the material, and their early stages are difficult to identify directly through surface inspection methods. However, acoustic emission can reflect the early damage activity of cracks inside the hull. , .

[0047] The preset risk level classification method is as follows: When the fusion damage index is within the threshold range of 0-0.25, it is classified as a safe level where no obvious damage is detected. When the fusion damage index is in the threshold range of 0.25-0.50, it is classified as a low-risk level with the presence of microcracks. When the fusion damage index is in the threshold range of 0.50-0.75, it is classified as a medium risk level with a tendency for crack propagation. When the fusion damage index is within the threshold range of 0.75-1, it is classified as a high-risk level with significant crack propagation.

[0048] Furthermore, when the determined damage risk level reaches the warning risk level threshold, a warning message is output. This involves determining the hull damage state based on the feature fusion analysis results of acoustic emission and visual detection, i.e., the fused damage index. When the fused damage index reaches the corresponding threshold range, the corresponding warning message is output.

[0049] Understandably, test results and warning information can be displayed through a human-machine interface.

[0050] In one embodiment, please refer to Figure 5 The method further includes: S501. Determine the acoustic emission detection state and the visual detection state based on the acoustic emission damage coefficient and the preset acoustic emission anomaly threshold, the visual damage coefficient and the preset visual detection anomaly threshold, respectively. S502. When the acoustic emission detection status is abnormal and the visual detection status is normal, execute the continuous monitoring strategy. S503. When the acoustic emission detection state is normal and the visual detection state is abnormal, execute the image re-inspection strategy.

[0051] If the acoustic emission detection status is abnormal while the visual detection status is normal (i.e., the damage coefficients satisfy AEI > 0.5 and VDI < 0.25), it can be determined as suspected early internal damage or mechanical noise interference, and a continuous monitoring strategy will be implemented. At this time, the system enters continuous monitoring mode, performs multiple acoustic emission detection data acquisitions on the area within a set time window, recalculates the AEI, and slightly increases the AEI weighting coefficient. If abnormal signals are detected in multiple consecutive monitoring cycles, visual detection is restarted and a risk level assessment is performed.

[0052] If the acoustic emission detection is normal and the visual detection is abnormal (i.e., the damage coefficients VDI > 0.5 and AEI < 0.25), the crack may be a surface scratch, corrosion, or a non-structural crack. In this case, the system re-inspects the area through multiple image acquisitions and detections, and slightly increases the VDI weighting coefficient to confirm the authenticity of the crack.

[0053] Understandably, noise and random interference are ubiquitous in real-world engineering environments. By setting a mechanism that prevents a single anomaly from triggering a final decision, these random noises can be effectively filtered out, allowing the system to react only to persistent or coordinated anomalies, thereby minimizing the false alarm rate. For occasional single anomalies, the system can simply store them locally or mark them with low priority, without immediately triggering high-priority interrupt handling, uploading complex waveforms in their entirety, or issuing alarm messages. This saves computing power on the central controller, reduces the burden on network communication, and allows system resources to focus on processing truly important data.

[0054] It should be noted that in the weighted fusion analysis of acoustic emission and visual data, different information fusion strategies can be used to comprehensively determine the detection results. All of the above fusion methods are based on the premise of simultaneously utilizing acoustic emission detection data and visual graphic data. Their technical effect is to improve the accuracy and reliability of hull damage assessment. That is, in damage assessment and subsequent early warning, the warning format and threshold setting method can be adjusted according to specific application needs, but all are based on the fusion analysis results to indicate the risk of hull damage.

[0055] According to the hull damage detection method provided in this disclosure, acoustic emission detection data and visual image data are acquired; based on the spatiotemporal characteristics of the acoustic emission detection data and visual image data, the acoustic emission detection data and the visual image data are associated; based on the acoustic emission detection data and the corresponding visual image data, acoustic emission feature damage coefficients and corresponding visual feature damage coefficients are obtained respectively; based on the acoustic emission feature damage coefficients and visual feature damage coefficients, the risk level of hull damage is determined.

[0056] By simultaneously acquiring acoustic emission detection data and visual image data, and performing spatiotemporal correlation between the two, the risk level of hull damage is determined based on the acoustic emission characteristic damage coefficient and the corresponding visual characteristic damage coefficient, thereby improving the completeness and comprehensiveness of hull damage information acquisition.

[0057] This invention also provides a hull damage detection device 600, please refer to [link / reference]. Figure 6The hull damage detection device 600 includes a damage detection module 601, a damage feature association module 602, a damage result output module 603, and a damage level judgment module 604. The damage detection module 601 acquires acoustic emission detection data and visual image data; the damage feature association module 602 associates the acoustic emission detection data and the visual image data based on their spatiotemporal characteristics; the damage result output module 603 obtains acoustic emission feature damage coefficients and corresponding visual feature damage coefficients based on the acoustic emission detection data and the corresponding visual image data, respectively; and the damage level judgment module 604 determines the risk level of hull damage based on the acoustic emission feature damage coefficients and visual feature damage coefficients.

[0058] In this embodiment, in order to obtain the acoustic emission characteristic damage coefficient, the damage result output module 603 extracts the acoustic emission event count, average amplitude, signal energy, and count rate from the acoustic emission detection data obtained by the damage detection module 601; and then calculates the acoustic emission characteristic damage coefficient based on the acoustic emission event count, average amplitude, signal energy, and count rate.

[0059] In this embodiment, in order to obtain the visual feature damage coefficient, the damage result output module 603 extracts the visual image data acquired by the damage detection module 601 and inputs the visual image data into a pre-trained visual detection model to obtain the visual detection result. The visual detection result includes at least crack length, crack width, and crack detection confidence. Based on the crack length, crack width, and crack detection confidence, the visual feature damage coefficient is calculated.

[0060] In this embodiment, the damage level judgment module 604 extracts the visual feature damage coefficient and acoustic emission feature damage coefficient output by the damage result output module 603. Based on the acoustic emission feature damage coefficient and the visual feature damage coefficient, a preset weighted fusion model is used to calculate the fusion damage index. Based on the fusion damage index and the preset risk level classification strategy, the risk level of the hull damage is determined.

[0061] As can be seen, the pre-defined risk levels are as follows: When the fusion damage index is within the threshold range of 0-0.25, it is classified as a safe level where no obvious damage is detected. When the fusion damage index is in the threshold range of 0.25-0.50, it is classified as a low-risk level with the presence of microcracks. When the fusion damage index is in the threshold range of 0.50-0.75, it is classified as a medium risk level with a tendency for crack propagation. When the fusion damage index is within the threshold range of 0.75-1, it is classified as a high-risk level with significant crack propagation.

[0062] The damage level assessment module 604 determines and outputs the corresponding risk level data based on a threshold range.

[0063] The aforementioned hull damage detection device 600 can be implemented as a computer program, which can, for example, Figure 7 It runs on the electronic device shown.

[0064] Please see Figure 7 , Figure 7 This is a schematic block diagram of an electronic device provided in an embodiment of the present invention. The electronic device 700 is a host computer or a server.

[0065] See Figure 7 The electronic device 700 includes a processor 702, a memory, and a network interface 705 connected via a device bus 701. The memory may include a storage medium 703 and internal memory 704.

[0066] The storage medium 703 may store an operating system 7031 and a computer program 7032. When the computer program 7032 is executed, it enables the processor 702 to perform a hull damage detection method.

[0067] The processor 702 provides computing and control capabilities to support the operation of the entire electronic device 700.

[0068] The internal memory 704 provides an environment for the operation of the computer program 7032 in the storage medium 703. When the computer program 7032 is executed by the processor 702, the processor 702 can perform the hull damage detection method.

[0069] This network interface 705 is used for network communication, such as providing data transmission. Those skilled in the art will understand that... Figure 7 The structure shown is merely a block diagram of a portion of the structure related to the present invention and does not constitute a limitation on the electronic device 700 to which the present invention is applied. The specific electronic device 700 may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0070] The processor 702 is used to run a computer program 7032 stored in a memory to implement the ship damage detection method disclosed in the embodiments of the present invention.

[0071] Those skilled in the art will understand that Figure 7The embodiments of the computer device shown do not constitute a limitation on the specific configuration of the computer device. In other embodiments, the computer device may include more or fewer components than illustrated, or combine certain components, or have different component arrangements. For example, in some embodiments, the computer device may include only memory and a processor. In such embodiments, the structure and function of the memory and processor are different from those shown. Figure 5 The embodiments shown are consistent and will not be described again here.

[0072] It should be understood that, in this embodiment of the invention, the processor 702 may be a Central Processing Unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0073] In another embodiment of the present invention, a computer-readable storage medium is provided. This computer-readable storage medium may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. The computer-readable storage medium stores a computer program, wherein the computer program, when executed by a processor, implements the hull damage detection method disclosed in the embodiments of the present invention.

[0074] Those skilled in the art will readily understand that, for the sake of convenience and brevity, the specific working processes of the devices, apparatuses, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art 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 invention.

[0075] In the embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Units with the same function may be grouped into one unit. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, or may be electrical, mechanical, or other forms of connection.

[0076] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of the present invention, depending on actual needs.

[0077] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0078] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause an electronic device (which may be a personal computer, a backend server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), magnetic disks, or optical disks.

[0079] The specific embodiments of the present invention described above do not constitute a limitation on the scope of protection of the present invention. Any other corresponding changes and modifications made in accordance with the technical concept of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. A method for detecting ship hull damage, characterized in that, Includes the following steps: Acquire acoustic emission detection data and visual image data; Based on the spatiotemporal characteristics of the acoustic emission detection data and the visual image data, the acoustic emission detection data and the visual image data are associated; Based on the acoustic emission detection data and the corresponding visual image data, the acoustic emission feature impairment coefficient and the corresponding visual feature impairment coefficient are obtained respectively. as well as Based on the acoustic emission characteristic damage coefficient and the visual characteristic damage coefficient, the risk level of hull damage is determined.

2. The method for detecting ship damage according to claim 1, characterized in that, The method for obtaining the acoustic emission characteristic damage coefficient is as follows: Extract the acoustic emission event count, average amplitude, signal energy, and count rate from the acoustic emission detection data; The acoustic emission characteristic damage coefficient is calculated based on the acoustic emission event count, average amplitude, signal energy, and count rate.

3. The method for detecting ship damage according to claim 2, characterized in that, The formula for calculating the acoustic emission characteristic damage coefficient is as follows: ； Where N is the acoustic emission event count, A is the average amplitude, E is the signal energy, C is the count rate, and w1, w2, w3, and w4 are the acoustic emission characteristic weighting coefficients.

4. The method for detecting ship damage according to claim 1, characterized in that, The method for obtaining the visual feature impairment coefficient is as follows: The visual image data is input into a pre-trained visual detection model to obtain visual detection results, wherein the visual detection results include at least crack length, crack width, and crack detection confidence. The visual feature damage coefficient is calculated based on the crack length, crack width, and crack detection confidence level.

5. The method for detecting ship hull damage according to claim 4, characterized in that, The formula for calculating the visual feature impairment coefficient is as follows: ； in, The length of the crack. The crack width; For crack detection confidence level, These are the visual feature weighting coefficients.

6. The method for detecting ship hull damage according to claim 1, characterized in that, The determination of the risk level of hull damage based on the acoustic emission characteristic damage coefficient and the visual characteristic damage coefficient includes: Based on the acoustic emission feature damage coefficient and the visual feature damage coefficient, a preset weighted fusion model is used to calculate the fusion damage index. Based on the fusion damage index and the preset risk level classification strategy, the risk level of the hull damage is determined.

7. The method for detecting ship damage according to claim 6, characterized in that, The formula for calculating the fusion damage index is as follows: ； in, , These are the weighting coefficients. The characteristic damage coefficient of acoustic emission. The visual feature impairment coefficient; The preset risk level classification method is as follows: When the fusion damage index is within the threshold range of 0-0.25, it is classified as a safe level where no obvious damage is detected. When the fusion damage index is in the threshold range of 0.25-0.50, it is classified as a low-risk level with the presence of microcracks. When the fusion damage index is in the threshold range of 0.50-0.75, it is classified as a medium risk level with a tendency for crack propagation. When the fusion damage index is within the threshold range of 0.75-1, it is classified as a high-risk level with significant crack propagation.

8. A hull damage detection device, characterized in that, include: The damage detection module is used to acquire acoustic emission detection data and visual image data; The damage feature association module is used to associate the acoustic emission detection data and the visual image data based on the spatiotemporal features of the acoustic emission detection data and the visual image data; The damage result output module is used to obtain the acoustic emission detection result and the corresponding visual detection result based on the acoustic emission detection data and the corresponding visual image data, respectively. as well as The damage level assessment module is used to determine the risk level of hull damage based on the acoustic emission detection results and visual inspection results.

9. An electronic device, characterized in that, include: Processor and memory; The memory stores computer programs that can be executed by the processor; When the processor executes the computer program, it implements the steps in the hull damage detection method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer program instructions, which, when executed by a processor, implement the hull damage detection method as described in any one of claims 1-7.

Images