Ship inner cabin coating monitoring method based on pixel scanning and laser scanning

Abstract

The invention relates to the technical field of corrosion monitoring, and particularly discloses a ship inner cabin coating monitoring method based on pixel scanning and laser scanning, and the method comprises the steps: determining a plurality of composite information collection positions based on the structure information of a ship inner cabin, the sum of the view ranges of all the composite information acquisition positions covers all to-be-detected ranges of the ship inner cabin; generating an inner cabin image of the ship inner cabin based on the column pixel scanning information; generating point cloud information of the ship inner cabin based on the two-dimensional laser scanning information; and analyzing a coating corrosion state of the ship inner cabin based on the inner cabin image and the point cloud information, and generating a coating monitoring result. Through intelligent planning of a composite information acquisition position and a multi-modal data synchronous acquisition mechanism, the problems of image distortion, detail loss and the like caused by complex structure and insufficient light of a ship inner cabin in ship inner cabin coating monitoring are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of ship monitoring, and particularly relates to a method for monitoring the coating of a ship's inner cabin based on pixel scanning and laser scanning. Background Art

[0002] The inner cabins of ships (including cargo holds, ballast tanks, fuel tanks, etc.) are in corrosive environments such as humidity, salt spray, and chemical media for a long time, which leads to the corrosion of metal structures and thus poses a threat to the safety of ships. To protect these critical areas and extend their service life, protective coatings are usually applied to the inner cabin surfaces. However, in practical applications, these coatings may fail due to problems such as rust, so it is crucial to regularly monitor the coatings of ship inner cabins.

[0003] Traditional coating condition monitoring mainly relies on manual visual inspection or local sensor detection. This method is inefficient, has a limited coverage range, and is highly subjective. Especially in the complex-structured inner cabins of ships, it is difficult to comprehensively capture all corrosion situations through manual inspection.

[0004] With the continuous progress of image recognition technology, technicians have begun to use this technology to detect ship inner cabins. For example, dome camera imaging devices are used to collect images of the inner cabin and perform rust recognition. However, in practical applications, the existing technology faces several challenges:

[0005] Firstly, since the inner cabin of a ship is a three-dimensional structure, when using an ordinary camera to collect images, image distortion or perspective deformation is likely to occur, which will cause errors in the assessment of the actual rust situation and thus reduce the accuracy of the images. Secondly, the light in the inner cabin is usually weak, and ordinary cameras are greatly interfered when collecting images in this environment, resulting in the image clarity and collection range not meeting the actual requirements. In addition, the area of some rust regions is very small, which limits the accuracy and recognition ability of coating rust monitoring. Summary of the Invention

[0006] To overcome the above technical problems existing in the prior art, an embodiment of the present invention provides a method for monitoring the coating of a ship's inner cabin based on pixel scanning and laser scanning, which is applied to a coating monitoring device for a ship's cabin. The method includes: determining a plurality of composite information collection positions based on the structural information of the ship's inner cabin, and the sum of the field of view ranges of all composite information collection positions covers all the areas to be detected in the ship's inner cabin; simultaneously collecting column pixel scanning information and two-dimensional laser scanning information at each composite information collection position; generating an inner cabin image of the ship's inner cabin based on the column pixel scanning information; generating point cloud information of the ship's inner cabin based on the two-dimensional laser scanning information; analyzing the corrosion state of the coating of the ship's inner cabin based on the inner cabin image and the point cloud information to generate a coating monitoring result.

[0007] Preferably, determining multiple composite information acquisition positions based on the structural information of the ship's inner cabin includes: determining the spatial parameters and occlusion data of the ship's inner cabin based on the structural information; determining the information acquisition area based on the spatial parameters and the occlusion data; determining the aperture of the punching, and determining the strength influence factor of the ship's inner cabin when punching in the information acquisition area based on the aperture; determining the composite information acquisition positions according to the strength influence factor, a preset strength threshold, a preset quantity threshold, and a preset position constraint.

[0008] Preferably, the cabin coating monitoring device includes an integrally arranged industrial line array camera and a two-dimensional laser scanning radar. The acquisition of column pixel scanning information and two-dimensional laser scanning information includes: determining the information acquisition trajectory; determining the preset shooting speed of the industrial line array camera at the current composite information acquisition position and the acquisition pitch angle of the two-dimensional laser scanning radar at the current composite information acquisition position; controlling the industrial line array camera to perform corresponding column pixel scanning operations based on the information acquisition trajectory and the preset shooting speed to obtain column pixel scanning information, where the column pixel scanning information includes all pixel information within the field of view corresponding to the current composite information acquisition position; controlling the two-dimensional laser scanning radar to perform corresponding laser scanning operations at the current composite information acquisition position based on the information acquisition trajectory and the acquisition pitch angle to obtain two-dimensional laser scanning information, where the two-dimensional laser scanning information includes all point cloud information within the field of view corresponding to the current composite information acquisition position.

[0009] Preferably, determining the information acquisition trajectory includes: determining the optical axis direction of the line array camera at the current composite information acquisition position; obtaining the cabin wall normal field, determining the centering position based on the cabin wall normal field, and determining the configured position of the line array camera based on the centering position; determining the shooting trajectory of the line array camera based on the field of view at the current composite information acquisition position, the optical axis direction, and the configured position.

[0010] Preferably, the method further includes: obtaining the spatial depth information of the ship's inner cabin based on the two-dimensional laser scanning information; determining the first included angle between each part of the ship's inner cabin and the optical axis direction based on the spatial depth information; adjusting the preset shooting speed based on the first included angle to generate a first adjusted speed, where the first adjusted speed is inversely proportional to the first included angle; controlling the industrial line array camera to perform corresponding column pixel scanning operations based on the information acquisition trajectory and the first adjusted speed.

[0011] Preferably, the method further includes: determining a scanning direction of the two-dimensional lidar; determining a plane angle of a current scanning plane in the ship's inner cabin based on the two-dimensional lidar scanning information; determining a second included angle between the scanning direction and the plane angle; determining whether the second included angle is greater than a preset included angle value; if so, determining a first jitter frequency and a first jitter amplitude based on the second included angle; controlling the two-dimensional lidar to perform a compensation scanning operation on the current scanning plane based on the first jitter frequency and the first jitter amplitude to generate compensated lidar scanning information; and optimizing the two-dimensional lidar scanning information based on the compensated lidar scanning information to generate optimized two-dimensional lidar scanning information.

[0012] Preferably, the analyzing the coating corrosion state of the ship's inner cabin based on the inner cabin image and the point cloud information to generate a coating monitoring result includes: performing a fusion process on the inner cabin image and the point cloud information to obtain fused information; performing a coating corrosion state analysis on the ship's inner cabin based on the fused information to generate corrosion height information and corrosion area information; and generating a coating monitoring result based on the corrosion height information and the corrosion area information.

[0013] Preferably, the performing a fusion process on the inner cabin image and the point cloud information to obtain fused information includes: extracting depth information of the ship's inner cabin based on the point cloud information; performing a three-dimensional conversion on the inner cabin image based on the depth information to generate fused information; or: determining a plurality of scanning planes of the ship's inner cabin based on the point cloud information; performing a corrosion analysis on each scanning plane to generate preliminary corrosion information; and highlighting the inner cabin image based on the preliminary corrosion information to generate fused information.

[0014] Preferably, the method further includes: after generating the coating monitoring result, determining a suspected corrosion area based on the coating monitoring result; adjusting the preset shooting speed based on the suspected corrosion area to generate a second adjusted speed, and determining a second jitter frequency and a second jitter amplitude of the two-dimensional lidar based on the suspected corrosion area, the second jitter frequency being greater than the first jitter frequency and the second jitter amplitude being greater than the first jitter amplitude; controlling the industrial line array camera to perform a corresponding column pixel scanning operation based on the information acquisition trajectory and the second adjusted speed; and controlling the two-dimensional lidar to perform a compensation scanning operation on the current scanning plane based on the first jitter frequency and the first jitter amplitude.

[0015] Preferably, the method further includes: obtaining historical maintenance measures and their corresponding historical maintenance times and historical maintenance locations; determining a maintenance method effectiveness coefficient according to the historical maintenance measures; determining a maintenance time decay coefficient according to the historical maintenance measures and their corresponding historical maintenance times; determining a spatial influence coefficient according to the historical maintenance locations; determining a corrosion risk prediction value according to the maintenance method effectiveness coefficient, the maintenance time decay coefficient, the spatial influence coefficient, and the coating monitoring results; and generating a corrosion risk prediction map according to the corrosion risk prediction value.

[0016] Through the technical solution provided by the present invention, the present invention has at least the following technical effects: By means of the intelligent planning of the composite information acquisition location and the multi-modal data synchronous acquisition mechanism in the embodiments of the present invention, the problems of image distortion, incomplete coverage, and detail loss caused by complex three-dimensional structures and insufficient light in traditional in-cabin coating monitoring of ships are solved.

[0017] Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent specific implementation part. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The drawings are used to provide a further understanding of the embodiments of the present invention, and constitute a part of the specification, and are used to explain the embodiments of the present invention together with the following specific implementation manners, but do not constitute a limitation to the embodiments of the present invention. In the drawings:

[0019] Figure 1 is a flowchart of a method for monitoring the coating of a ship's inner cabin based on pixel scanning and laser scanning provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0020] The following will describe in detail the specific implementation manners of the embodiments of the present invention with reference to the drawings. It should be understood that the specific implementation manners described herein are only used to illustrate and explain the embodiments of the present invention, and are not used to limit the embodiments of the present invention.

[0021] The terms "system" and "network" in the embodiments of the present invention can be used interchangeably. "Plural" means two or more. In view of this, in the embodiments of the present invention, "plural" can also be understood as "at least two". "And / or" describes the association relationship of associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " generally represents an "or" relationship between the associated objects before and after, unless otherwise specified. In addition, it should be understood that in the description of the embodiments of the present invention, the terms such as "first" and "second" are only used for the purpose of distinguishing descriptions, and cannot be understood as indicating or implying relative importance, nor can they be understood as indicating or implying order.

[0022] Please refer toFigure 1 , an embodiment of the present invention provides a method for monitoring the coating of the inner cabin of a ship based on pixel scanning and laser scanning, which is applied to a monitoring device for the coating of the cabin. The method includes:

[0023] Step 1, determine a plurality of composite information acquisition positions based on the structural information of the inner cabin of the ship, and the sum of the visual field ranges of all the composite information acquisition positions covers all the areas to be detected in the inner cabin of the ship.

[0024] When monitoring the inner cabin of a ship, it is necessary to first determine the acquisition positions in order to drill holes at the corresponding positions to allow the scanning device to penetrate and obtain laser scanning data and pixel scanning data. Due to the complexity of the environment in the inner cabin of the ship, it is impossible to obtain sufficient data by collecting the scanning data of the inner cabin of the ship at only one position. Therefore, it is necessary to obtain the scanning data at multiple positions simultaneously and perform comprehensive analysis on it. The prior art usually randomly determines the acquisition positions. However, in the actual application process, on the one hand, due to the complex structure of the inner cabin of the ship, pipelines, corners, equipment, etc. may be randomly distributed. Therefore, there are problems such as coverage blind spots or structural safety hazards in the acquisition positions selected by the random method. On the other hand, drilling holes at positions such as the top of the inner cabin of the ship will damage its force-bearing structure and reduce its physical performance. Therefore, holes cannot be randomly drilled on the inner cabin of the ship, and the positions and quantities of the holes should be determined in combination with its actual physical strength requirements. Especially in the case of complex cabin layouts and dynamic occlusion environments, the prior methods lack the comprehensive optimization of structural strength and visual field.

[0025] In the embodiment of the present invention, determining a plurality of composite information acquisition positions based on the structural information of the inner cabin of the ship includes: determining the spatial parameters and occlusion data of the inner cabin of the ship based on the structural information; determining the information acquisition area based on the spatial parameters and the occlusion data; determining the aperture of the hole, and determining the strength influence factor of the inner cabin of the ship when drilling holes in the information acquisition area based on the aperture; determining the composite information acquisition positions according to the strength influence factor, a preset strength threshold, a preset quantity threshold, and a preset position constraint.

[0026] In a possible embodiment, the structural information of the inner cabin of a ship is obtained by analyzing the drawings of the inner cabin, and then the spatial parameters and occlusion data are extracted therefrom. The spatial data may include, but is not limited to, information such as the length, width, height, and three-dimensional position of the inner cabin of the ship. The occlusion data may include, but is not limited to, data such as pipeline data, corner data, and equipment data (such as the size data and spatial position data of the equipment) with occlusion effects in the inner cabin of the ship. Based on these spatial parameters and occlusion data, the areas where information can be collected are determined, and a plurality of initial composite information collection positions are generated on these areas. For example, in the embodiment of the present invention, the cabin coating monitoring device includes an integrally arranged industrial line array camera and a two-dimensional laser scanning radar. By opening holes in the top of the inner cabin of the ship, the industrial line array camera and the two-dimensional laser scanning radar are inserted into the inner cabin of the ship and scanning operations are performed. The areas where information can be collected are located, for example, in the upper left corner area and the upper right corner area of the top of the inner cabin of the ship.

[0027] When specifically determining the composite information collection positions, using the ray casting algorithm, the viewing ranges corresponding to different initial collection positions can be calculated. Subsequently, the hole diameter is determined according to the size of the cabin coating monitoring device, and the influence factor of the hole diameter on the strength of the inner cabin of the ship is evaluated by combining finite element analysis. On this basis, a multi-objective optimization model can be constructed, aiming to minimize the number of collection positions, maximize the viewing range, and minimize the influence factor on the strength of the inner cabin of the ship. Under the consideration of preset strength thresholds, quantity thresholds, and position constraint conditions, the genetic algorithm is used to solve the multi-objective optimization model, so as to obtain the number and precise coordinates of the composite information collection positions. Among them, the preset strength thresholds, quantity thresholds, and position constraints are set by technicians according to experience and actual needs and are not specifically limited in the embodiment of the present invention. For example, considering that the top of the inner cabin of the ship is close to the deck and is convenient for maintenance, it can be set that the image acquisition window is located at the top of the inner cabin of the ship. Since the image acquisition window is usually not opened in the middle position of the inner cabin of the ship, it can be set that the image acquisition window is located at the edge of the top of the inner cabin. For example, 4 image acquisition windows can be set in the inner cabin of a large ship, 2 in the inner cabin of a medium-sized ship, and 1 in the inner cabin of a small ship.

[0028] The embodiment of the present invention can effectively scan all scene information in the complex scene of the inner cabin of the ship by integrating structural strength constraints and visual field optimization, and at the same time can minimize the number of holes opened in the inner cabin of the ship, minimize the impact of the monitoring process on the structure of the inner cabin of the ship as much as possible, and ensure its service life. Compared with the random point distribution method, the multi-objective optimization model based on the ray casting algorithm and finite element analysis can accurately calculate the effective viewing range of each collection point and the mechanical influence of the hole opening on the cabin wall, eliminate the coverage blind area and avoid the high stress risk area.

[0029] After determining the composite information acquisition positions, start scanning the information in the inner cabin of the ship. In the prior art, using ordinary cameras to acquire images of the inner cabin of a ship is affected and interfered by a large number of factors, such as light intensity, oil stains, occlusion, etc. Therefore, the accuracy and accuracy of its information acquisition are relatively low and cannot meet the actual requirements. At the same time, traditional scanning devices for monitoring the coatings in the inner cabin of ships mostly rely on a single sensor, such as a camera or a lidar, and there is a problem of single data dimension.

[0030] Step 2, simultaneously acquire column pixel scan information and two-dimensional laser scan information at each composite information acquisition position.

[0031] After determining the composite information acquisition positions, control the cabin coating monitoring device to scan the inner cabin of the ship. Specifically, control the industrial line array camera and the two-dimensional laser scanning radar to scan simultaneously to obtain the corresponding scan information. In an embodiment of the present invention, the acquisition of the column pixel scan information and the two-dimensional laser scan information includes: determining the information acquisition trajectory; determining the preset shooting speed of the industrial line array camera at the current composite information acquisition position and the acquisition pitch angle of the two-dimensional laser scanning radar at the current composite information acquisition position; controlling the industrial line array camera to perform the corresponding column pixel scan operation based on the information acquisition trajectory and the preset shooting speed to obtain the column pixel scan information, where the column pixel scan information includes all pixel information within the field of view corresponding to the current composite information acquisition position; controlling the two-dimensional laser scanning radar to perform the corresponding laser scan operation based on the information acquisition trajectory and the acquisition pitch angle at the current composite information acquisition position to obtain the two-dimensional laser scan information, where the two-dimensional laser scan information includes all point cloud information within the field of view corresponding to the current composite information acquisition position.

[0032] In a possible embodiment, connect the cabin coating monitoring device (including an integrally arranged industrial line array camera and a two-dimensional laser scanning radar) to one end of the telescopic rod through a pan-tilt, and then connect the other end of the telescopic rod to the suspension bracket located outside the opening. In this way, the telescopic rod can drive the cabin coating monitoring device to vertically pass through the opening and penetrate deep into the inner cabin of the ship for scanning. According to the determined multiple composite information acquisition positions, the acquisition trajectories of the cabin coating monitoring device at each composite information acquisition position can be determined in advance, and the acquisition trajectories include the acquisition trajectory of the industrial line array camera and the acquisition trajectory of the two-dimensional laser scanning radar.

[0033] In the embodiments of the present invention, the determination of the information acquisition trajectory includes: determining the optical axis direction of the line array camera at the current composite information acquisition position; obtaining the bulkhead normal field, determining the centering position based on the bulkhead normal field, and determining the configuration position of the line array camera based on the centering position; determining the shooting trajectory of the line array camera based on the field of view range, the optical axis direction, and the configuration position at the current composite information acquisition position; determining the scanning trajectory of the two-dimensional laser scanning radar based on the shooting trajectory; and generating the information acquisition trajectory based on the shooting trajectory and the scanning trajectory.

[0034] In a possible implementation manner, first, determine the optical axis direction of the line array camera at the current composite information acquisition position. This optical axis direction is, for example, the direction of the plane extending outward along the axis coaxial with the cabin coating monitoring device, and this plane is perpendicular to the side wall of the ship's inner cabin. Since in the application process, although the industrial line array camera has the effects of high precision and strong anti-interference ability, when it acquires images, it is still affected by the angle between the object to be acquired and the industrial line array camera. Therefore, in an ideal situation, keeping the angle between the object to be acquired and the industrial line array camera close to 90° can make the quality of the acquired images the best. Therefore, in the embodiments of the present invention, further obtain the bulkhead normal field, determine the centering position based on the bulkhead normal field, and determine the configuration position of the line array camera based on the centering position. At this configuration position, the industrial line array camera can achieve the lowest angle with all the bulkheads in the ship's inner cabin to ensure the best overall quality of the acquired images. Finally, determine the shooting trajectory of the line array camera according to the field of view range, the optical axis direction, and the configuration position at the current composite information acquisition position. This shooting trajectory can ensure that the industrial line array camera acquires all the pixel information in the field of view range corresponding to the current composite information acquisition position. On this basis, the scanning trajectory of the two-dimensional laser scanning radar can be further determined, that is, the scanning trajectory of the two-dimensional laser scanning radar is also determined based on the configuration position of the industrial line array camera. Specifically, it is determined based on the configuration position of the industrial line array camera, the field of view range at the current composite information acquisition position, and its own scanning angle range. The information acquisition trajectory can be generated according to the above shooting trajectory and scanning trajectory.

[0035] In the embodiments of the present invention, by combining the actual physical characteristics of the industrial line array camera during the scanning process, the information acquisition trajectory is correspondingly designed to ensure that the industrial line array camera has the smallest distortion during image acquisition, effectively improving the accuracy of the entire ship's inner cabin image acquisition and the precision of subsequent corrosion analysis. On this basis, determining the scanning trajectory of the two-dimensional laser scanning radar can ensure that both the industrial line array camera and the two-dimensional laser scanning radar can independently complete the independent information scanning work of the current field of view, thereby providing high-precision and reliable data support for subsequent accurate corrosion analysis.

[0036] At this time, further determine the preset shooting speed of the industrial line array camera, for example, determine the preset shooting speed of the industrial line array camera according to the actual clarity requirements; and determine the acquisition pitch angle of the two-dimensional lidar. After determining the above parameters, the industrial line array camera and the two-dimensional lidar both independently collect their own information (pixel information and point cloud information) and are used for subsequent independent analysis to improve the monitoring accuracy.

[0037] During the scanning process, control the cabin coating monitoring device to rotate circumferentially along the predetermined information acquisition trajectory, and make the industrial line array camera scan at the set shooting speed to obtain column pixel scan information. At the same time, control the two-dimensional lidar to scan at the set pitch angle to obtain two-dimensional lidar scan information.

[0038] In another embodiment, when the vertical field of view range of the composite information acquisition position exceeds the vertical field of view of the industrial line array camera, the field of view range corresponding to the composite information acquisition position can be horizontally divided into two or more sub-field of view ranges, so as to obtain two or more information acquisition trajectories. In this way, the data trajectories of all areas inside the ship's cabin can be adaptively segmented to cover the complex structure of the cabin wall, and the scanning blind area in the vertical direction can be completely eliminated through the multi-sub-field of view coverage strategy. At the same time, synchronously adjust the shooting speed and the pitch angle of the lidar to ensure seamless matching of the data ranges of the two sensors, and significantly improve the detection range and consistency of coating micro-defects.

[0039] In the embodiment of the present invention, through the integrated industrial line array camera and two-dimensional lidar, on the one hand, by utilizing the excellent performance of the industrial line array camera, effectively solve the problems of light interference, oil stain interference, etc. in the complex scene inside the ship's cabin, realize the accurate acquisition of images inside the ship's cabin, and facilitate the subsequent reliable distinction between oil stains and corrosion, achieving higher monitoring accuracy; on the other hand, through the combined coordinated control and acquisition parameters (information acquisition trajectory, acquisition pitch angle, etc.) of the industrial line array camera and the two-dimensional lidar, realize the efficient and blind-free automatic scanning inside the ship's cabin, ensure the spatio-temporal synchronization of image data and point cloud data, and avoid the complexity of repeated calibration required by traditional split devices.

[0040] In the actual application process, industrial line cameras adopt a fixed shooting speed. Although they can complete data acquisition, in the face of the complex geometric structures inside the ship's cabin, such as curved bulkheads, when the fixed shooting speed faces scanning surfaces at different angles, the number of image pixels collected per unit time is inconsistent. It cannot be dynamically adjusted according to the curvature of the cabin surface. When the included angle between the optical axis of the industrial line camera and the normal direction of the cabin surface increases, such as when scanning a concave area, the sampling interval between adjacent pixels at a fixed speed becomes larger, and the image produces stretching distortion, affecting the geometric measurement range of coating defects. At the same time, redundant data is generated due to uniform scanning in flat areas, while in high-curvature or key areas, such as welds and corrosion points, details are lost due to insufficient sampling density, and multiple supplementary scans are required, resulting in low efficiency.

[0041] To solve the above technical problems, in an embodiment of the present invention, the method further includes: determining a first included angle between each location inside the ship's cabin and the optical axis direction based on the two-dimensional laser scanning information; adjusting the preset shooting speed based on the first included angle to generate a first adjusted speed, where the first adjusted speed is inversely proportional to the first included angle; and controlling the industrial line camera to perform corresponding column pixel scanning operations based on the information acquisition trajectory and the first adjusted speed.

[0042] In a possible implementation manner, through two-dimensional laser scanning information, a normal vector estimation algorithm (such as PCA) is used to calculate the local surface normal vector of each point, and then the actual included angle between this point and the optical axis of the industrial line camera is determined. At the same time, the surface normal vectors of each region are calculated to determine the real-time included angle with the camera's optical axis; the moving speed of the line camera is dynamically adjusted based on the size of the included angle. When the included angle is small (for example, at a flat bulkhead), high-speed scanning is maintained to improve efficiency, while when the included angle increases (for example, in a curved concave area), the speed is automatically reduced to increase the pixel sampling density, thereby suppressing image perspective distortion; the spatial coordinates of the laser point cloud and the camera image texture are synchronously fused, and the residual deformation is eliminated through a geometric correction model to achieve high-accuracy seamless detection of coating defects throughout the cabin.

[0043] The embodiment of the present invention adaptively adjusts the acquisition speed of the industrial line camera according to the actual spatial distribution inside the ship's cabin, improves the pixel density, thereby while ensuring data integrity, significantly enhancing the accuracy and efficiency of pixel information acquisition, and effectively suppressing image perspective distortion.

[0044] In the actual application process, due to the limitation of equipment procurement costs, the number of acquisition points on the two-dimensional lidar often only has the accuracy to meet the general use. For example, for ordinary lidar, its angular resolution is generally at the level of 1°. Because the scanning beam is sparse, such as only multiple array single-line scanning, the sparse data volume is difficult to meet the coating monitoring requirements. If its angular resolution is increased, the equipment procurement cost will increase greatly; if the angular resolution is not increased, it cannot meet the monitoring accuracy requirements in large-space scenarios such as the inner cabin of a ship. Especially in complex bulkhead structures, there is a plane angle between the scanning plane of the two-dimensional lidar and the inner wall of the ship's cabin, which further reduces the accuracy of the point cloud data scanned by the lidar.

[0045] In an embodiment of the present invention, the method further includes: determining the scanning direction of the two-dimensional lidar; determining the plane angle of the current scanning plane of the inner cabin of the ship based on the two-dimensional laser scanning information; determining a second included angle between the scanning direction and the plane angle; judging whether the second included angle is greater than a preset included angle value; if so, determining a first jitter frequency and a first jitter amplitude based on the second included angle; controlling the two-dimensional lidar to perform a compensation scanning operation on the current scanning plane based on the first jitter frequency and the first jitter amplitude to generate compensation laser scanning information; optimizing the two-dimensional laser scanning information based on the compensation laser scanning information to generate optimized two-dimensional laser scanning information.

[0046] In a possible embodiment, a vertical jitter device is arranged on the cabin coating monitoring device. For example, the vertical jitter device can be configured by a motor + reducer to accurately control its jitter amplitude, and then it is applied to drive the two-dimensional lidar to jitter in the vertical direction through the vertical jitter device to achieve scanning compensation. For example, if the angular resolution of the two-dimensional lidar is 1°, during jitter compensation, the two-dimensional lidar can be controlled to scan downward by 1° first, then scan upward by 0.5°, and then based on the same principle, scan downward until the laser scanning of the entire screen is completed, thereby realizing the compensation of scanning accuracy. Therefore, the corresponding compensation parameters need to be determined according to the scanning accuracy required in actual situations.

[0047] Specifically, based on the two-dimensional laser scanning information, the least squares method is used to fit the plane normal of the current scanning plane, calculate the angle between the plane normal and the gravity direction, and use this angle as the plane angle. Further, calculate the deviation angle between the scanning direction (i.e., the pointing direction of the radar beam) and the plane angle, that is, the second angle. If this deviation angle exceeds the preset threshold, it is determined that the vertical coverage is insufficient and jitter compensation is required. At this time, the jitter frequency and jitter amplitude of the lidar in the vertical direction are dynamically set according to the size of the second angle. For example, the larger the angle, the higher the jitter frequency and the smaller the step size of the stepper motor. By controlling the two-dimensional lidar to perform high-frequency and small-amplitude swings in the vertical direction, the effective scanning beam can be extended, the point cloud density of the scanning on the inner wall of the cabin can be increased, and the data accuracy can be improved.

[0048] In this embodiment, through the dynamic jitter compensation technology, the scanning beam of the two-dimensional lidar is extended to multiple times of the original to match the field of view of the line array camera, eliminate the vertical blind area, increase the point cloud coverage rate, and solve the inherent hardware difference problem between the two-dimensional lidar and the line array camera in the vertical field of view, providing a reliable solution for low-cost and high-efficiency full-dimensional detection of the cabin.

[0049] Step 3, generate the inner cabin image of the ship's inner cabin based on the column pixel scanning information.

[0050] Step 4, generate the point cloud information of the ship's inner cabin based on the two-dimensional laser scanning information.

[0051] After the corresponding column pixel scanning information and two-dimensional laser scanning information are collected at each composite information acquisition position, on the one hand, all the column pixel scanning information can be integrated and processed to generate the inner cabin image of the ship's inner cabin; on the other hand, all the two-dimensional laser scanning information can be integrated and processed to generate the point cloud information of the ship's inner cabin. At this time, the coating corrosion state analysis is performed according to the above inner cabin image and point cloud information.

[0052] Step 5, analyze the coating corrosion state of the ship's inner cabin based on the inner cabin image and the point cloud information, and generate a coating monitoring result.

[0053] In the embodiment of the present invention, the analysis of the coating corrosion state of the ship's inner cabin based on the inner cabin image and the point cloud information to generate a coating monitoring result includes: performing fusion processing on the inner cabin image and the point cloud information to obtain the fused information; performing coating corrosion state analysis on the ship's inner cabin based on the fused information to generate corrosion height information and corrosion area information; generating a coating monitoring result based on the corrosion height information and the corrosion area information.

[0054] In a possible implementation, in order to overcome the technical problem in the prior art that the accuracy of corrosion analysis of a single data source is relatively low and cannot meet the actual requirements, the collected inner cabin images and point cloud information are fused to perform corrosion analysis from multiple dimensions and improve the analysis accuracy. Specifically, first, the inner cabin images and the point cloud data are aligned in space and time, the texture information is extracted from the inner cabin images, and the geometric information is extracted from the point cloud data. Then, through a feature matching algorithm, this information is mapped into a unified three-dimensional space to generate a composite data model that integrates color, depth, and surface topology information. Next, based on this composite data model, the three-dimensional deformation and two-dimensional diffusion characteristics of corrosion are analyzed synchronously. The local curvature and normal offset of the point cloud are used to quantify the depression depth of the coating, and combined with image segmentation technology, the boundary of the rust area is extracted to calculate the actual diffusion area. Finally, the height and area data of corrosion are integrated to construct a multi-dimensional evaluation index, and a hierarchical monitoring result including corrosion height and corrosion area information is generated, providing a comprehensive diagnosis basis from local to overall for coating maintenance.

[0055] In traditional ship coating detection, the fusion of images and point cloud data usually adopts simple superposition or static plane-by-plane analysis, resulting in low data utilization efficiency. For example, the coordinate system differences between two-dimensional images and three-dimensional point clouds make it difficult to accurately align texture and geometric information, and the apparent corrosion characteristics are separated from the three-dimensional deformation, making it impossible to comprehensively quantify the corrosion impact; while the method of segmenting the scanning plane with a fixed threshold does not consider the dynamic geometric characteristics of the curved cabin wall, resulting in missed detection of local corrosion diffusion paths and making it difficult to support accurate maintenance decisions.

[0056] In the embodiment of the present invention, the fusion process of the inner cabin images and the point cloud information to obtain the fused information includes: extracting the depth information of the ship's inner cabin based on the point cloud information; performing three-dimensional conversion on the inner cabin images based on the depth information to generate the fused information; or: determining multiple scanning planes of the ship's inner cabin based on the point cloud information; performing corrosion analysis on each scanning plane to generate preliminary corrosion information; highlighting the inner cabin images based on the preliminary corrosion information to generate the fused information.

[0057] In a possible implementation, a triangular mesh model of the ship's inner cabin is constructed by using point cloud data, and the three-dimensional coordinates and normal vectors corresponding to each pixel are extracted; the RGB texture of the inner cabin images is mapped to the mesh vertices to generate a three-dimensional texture model, and the fused information covers color, depth, and geometric attributes at the same time.

[0058] In this embodiment, through the mapping of depth information, the coordinate system deviation between two-dimensional images and three-dimensional point clouds is eliminated, realizing the seamless fusion of texture and geometric features, and significantly improving the accuracy of corrosion quantitative analysis.

[0059] In another possible implementation, according to the curvature and density of the point cloud, the bulkhead is segmented into multiple scanning planes, which include flat planes, arc planes, etc.; the corrosion analysis algorithm is independently executed for each scanning plane, including but not limited to the point cloud height difference threshold, image color clustering, etc., to identify the preliminary corrosion area, such as a height loss ≥ 0.2 mm or a color deviation from the standard value exceeding 15%; the coordinates of the preliminary corrosion area are projected onto the inner cabin image, and the fused information is generated through edge enhancement and pseudo-color highlighting (such as red marking) to visually display the location and severity of the corrosion.

[0060] This embodiment can dynamically segment the scanning plane based on geometric characteristics, accurately capture the corrosion diffusion law in complex areas such as curved surfaces and welds, thereby reducing missed detections and misjudgments.

[0061] In the traditional method, the corrosion area indicated by the coating detection result is directly determined as the final corrosion area. However, due to interference such as water mist and oil stains in the cabin, problems such as unclear images and laser failure will occur, affecting the coating detection accuracy. Technicians expect to further improve the accuracy of monitoring data based on the existing monitoring equipment and monitoring accuracy to assist in further improving the accuracy of monitoring results.

[0062] In the embodiment of the present invention, the method further includes: after generating the coating monitoring result, determining a suspected corrosion area based on the coating monitoring result; adjusting the preset shooting speed based on the suspected corrosion area to generate a second adjustment speed, and determining a second jitter frequency and a second jitter amplitude of the two-dimensional laser scanning radar based on the suspected corrosion area, the second jitter frequency being greater than the first jitter frequency, and the second jitter amplitude being greater than the first jitter amplitude; controlling the industrial line array camera to perform a corresponding column pixel scanning operation based on the information acquisition trajectory and the second adjustment speed; controlling the two-dimensional laser scanning radar to perform a compensation scanning operation on the current scanning plane based on the first jitter frequency and the first jitter amplitude.

[0063] In a possible implementation, areas suspected of corrosion are screened out based on the coating monitoring results, and their spatial distribution and risk levels are marked; for the identified suspected corrosion areas, the scanning speed is significantly reduced on the basis of the established shooting speed to form a second adjusted speed. At this time, the industrial line array camera can be controlled to perform a second and focused scan of the suspected corrosion area at this second adjusted speed, thereby increasing the image density of the suspected corrosion area. At the same time, on the basis of the first jitter frequency and the first jitter amplitude, the vertical jitter frequency and the jitter amplitude are synchronously increased to generate a second jitter frequency and a second jitter amplitude, so as to further expand the scanning beam density. At this time, the two-dimensional lidar can be controlled to perform a second and focused scan of the suspected corrosion area to obtain denser point cloud data, so as to further improve the corrosion identification accuracy and further improve the identification accuracy of the corrosion area, meeting the actual requirements.

[0064] In this embodiment, by controlling the line array camera to perform low-speed and high-precision scanning of the suspected corrosion area along the trajectory, and controlling the lidar to generate dense point clouds with wide-angle and high-frequency jitter, more detailed features of the suspected corrosion area can be obtained; by spatially and temporally aligning the local detailed features with the preliminary inspection results, the reflection and shadow artifacts can be eliminated, and a high-confidence coating corrosion detection result can be output.

[0065] In addition, the existing methods for assessing the coating corrosion risk in the inner cabin of ships mainly rely on the instant monitoring data of the current coating state, such as point cloud data and / or image data, while ignoring the long-term impact of historical maintenance measures on the coating performance, resulting in the prediction results deviating from the actual corrosion process.

[0066] In the embodiment of the present invention, the method further includes: obtaining historical maintenance measures and their corresponding historical maintenance times and historical maintenance locations; determining a maintenance method effectiveness coefficient according to the historical maintenance measures; determining a maintenance time attenuation coefficient according to the historical maintenance measures and their corresponding historical maintenance times; determining a spatial influence coefficient according to the historical maintenance location; determining a corrosion risk prediction value according to the maintenance method effectiveness coefficient, the maintenance time attenuation coefficient, the spatial influence coefficient and the coating monitoring results; and generating a corrosion risk prediction map according to the corrosion risk prediction value.

[0067] In a possible implementation, historical operation and maintenance data is obtained. The historical operation and maintenance data includes historical operation and maintenance time and historical operation and maintenance location. The time interval between the historical operation and maintenance time and the current time is calculated. When the time interval is less than a set value, this location is determined as a low-risk area, and a predicted corrosion risk map is generated based on the low-risk area. The predicted corrosion risk map generated in this way determines the risk only based on the time interval, ignoring the differences in the corrosion inhibition effects of different maintenance measures. For example, temporary painting (with a short validity period) and full coating replacement (long-term protection) are treated equally under the same time interval, resulting in misjudgment of high-risk areas. For example, a temporary maintenance area (such as 6 months after paint repair) is marked as low-risk due to a short time interval, but the actual coating has failed, and the missed inspection rate is as high as 30%; the non-linear degradation of the coating performance is ignored; the protection range of the maintenance measures for the surrounding area is ignored. Local repair welding can only inhibit corrosion near the repair point, but the traditional method regards it as a global low-risk; the spatial influence caused by coating bridging is ignored. For example, the location that needs to be maintained is a certain area, but during the recoating or spraying process, the paint may level off naturally or splash to adjacent areas, thereby reducing the corrosion risk of the adjacent areas.

[0068] To solve the above problems, in the embodiments of the present invention, the historical operation and maintenance data includes historical maintenance measures and their corresponding historical maintenance time and historical maintenance location. Generating a predicted corrosion risk map based on the historical operation and maintenance data includes: determining a maintenance method effectiveness coefficient according to the historical maintenance measures; determining a maintenance time decay coefficient according to the historical maintenance measures and their corresponding historical maintenance time; determining a spatial influence coefficient according to the historical maintenance location; determining a predicted risk corrosion value according to the maintenance method effectiveness coefficient, the maintenance time decay coefficient, and the spatial influence coefficient; and generating a predicted corrosion risk map according to the predicted risk corrosion value.

[0069] Specifically, in the scenario of coating maintenance in the inner cabin of a ship, a mapping relationship table of maintenance measures - maintenance protection effectiveness is established (for example, the protection effectiveness coefficient corresponding to full coating replacement is 1, the effectiveness coefficient of local mechanical grinding and recoating is 0.75, the efficiency coefficient of temporary spraying is 0.5, and the effectiveness coefficient is obtained by experimental fitting. The effectiveness coefficient of coating cleaning and maintenance is 0.1). The maintenance method effectiveness coefficient is determined according to the mapping relationship table of maintenance measures - maintenance protection effectiveness and the historical maintenance measures; a time decay function of the maintenance measure effectiveness is established, and the maintenance time decay coefficient is determined according to the time decay function, the historical maintenance measures, and their corresponding historical maintenance time. The time decay function is obtained by experimental fitting; a mapping relationship table of maintenance measure intervals - spatial influence coefficients is established (for example, for full ship recoating, the spatial influence coefficient is 1, for large-scale local mechanical grinding and recoating is 1.1, and for small-scale local mechanical grinding and recoating is 1.3). The spatial influence coefficient is determined according to the mapping relationship table of maintenance measure intervals - spatial influence coefficients and the historical maintenance location.

[0070] In the embodiments of the present invention, the relationships among maintenance measures, maintenance effectiveness, time, and space are comprehensively considered, thoroughly solving the problem in the prior art that the long-term impact of historical maintenance measures on the coating performance is ignored, resulting in the prediction result deviating from the actual corrosion process.

[0071] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the embodiments provided in the present application can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0072] The optional implementation manners of the embodiments of the present invention have been described in detail above in conjunction with the accompanying drawings. However, the embodiments of the present invention are not limited to the specific details in the above implementation manners. Within the technical concept scope of the embodiments of the present invention, various simple modifications can be made to the technical solutions of the embodiments of the present invention, and these simple modifications all belong to the protection scope of the embodiments of the present invention.

[0073] In addition, it should be noted that, among the various specific technical features described in the above specific implementation manners, they can be combined in any appropriate manner without conflict. To avoid unnecessary repetition, the embodiments of the present invention will not separately describe various possible combination manners.

[0074] In addition, any combination can be made among various different implementation manners of the embodiments of the present invention, as long as it does not violate the idea of the embodiments of the present invention, and it should also be regarded as the content disclosed by the embodiments of the present invention.

Claims

1. A method for monitoring the coating of the inner cabin of a ship based on pixel scanning and laser scanning, characterized in that, Applied to a cabin coating monitoring device, the method includes: Determine a plurality of composite information acquisition positions based on the structural information of the ship's inner cabin, and the sum of the field of view ranges of all the composite information acquisition positions covers all the areas to be detected in the ship's inner cabin; Simultaneously acquire column pixel scan information and two-dimensional laser scan information at each composite information acquisition position; Generate an inner cabin image of the ship's inner cabin based on the column pixel scan information; Generate point cloud information of the ship's inner cabin based on the two-dimensional laser scan information; Analyze the coating corrosion state of the ship's inner cabin based on the inner cabin image and the point cloud information to generate a coating monitoring result.

2. The method for monitoring the coating inside a ship's cabin based on pixel scanning and laser scanning according to claim 1, characterized in that The determining a plurality of composite information acquisition positions based on the structural information of the ship's inner cabin includes: Determine the spatial parameters and occlusion data of the ship's inner cabin based on the structural information; Determine the information acquisition area based on the spatial parameters and the occlusion data; Determine the aperture of the hole, and based on the aperture, determine the strength influence factor of the ship's inner cabin when drilling holes in the information acquisition area; Determine the composite information acquisition positions according to the strength influence factor, a preset strength threshold, a preset quantity threshold, and a preset position constraint.

3. The method for monitoring the coating inside a ship's cabin based on pixel scanning and laser scanning according to claim 1, characterized in that The cabin coating monitoring device includes an integrally arranged industrial line array camera and a two-dimensional laser scanning radar. The acquiring column pixel scan information and two-dimensional laser scan information includes: Determine the information acquisition trajectory; Determine the preset shooting speed of the industrial line array camera at the current composite information acquisition position, and the acquisition pitch angle of the two-dimensional laser scanning radar at the current composite information acquisition position; Control the industrial line array camera to perform corresponding column pixel scanning operations based on the information acquisition trajectory and the preset shooting speed to obtain column pixel scan information, and the column pixel scan information includes all pixel information in the field of view range corresponding to the current composite information acquisition position; Control the two-dimensional laser scanning radar to perform corresponding laser scanning operations at the current composite information acquisition position based on the information acquisition trajectory and the acquisition pitch angle to obtain two-dimensional laser scan information, and the two-dimensional laser scan information includes all point cloud information in the field of view range corresponding to the current composite information acquisition position.

4. A method for monitoring the coating of the inner cabin of a ship based on pixel scanning and laser scanning according to claim 3, characterized in that, The determining the information acquisition trajectory includes: Determine the optical axis direction of the line array camera at the current composite information acquisition position; Obtain the cabin wall normal field, determine the centering position based on the cabin wall normal field, and determine the configuration position of the line array camera based on the centering position; Determine the shooting trajectory of the line array camera based on the field of view range, the optical axis direction, and the configuration position of the current composite information acquisition position; Determine the scanning trajectory of the two-dimensional laser scanning radar based on the shooting trajectory; Generate an information acquisition trajectory based on the shooting trajectory and the scanning trajectory.

5. A method for monitoring the coating inside a ship's cabin based on pixel scanning and laser scanning according to claim 4, characterized in that, The method further includes: Obtain the spatial depth information of the ship's inner cabin based on the two-dimensional laser scan information; Determine the first included angle between each part of the ship's inner cabin and the optical axis direction based on the spatial depth information; Adjust the preset shooting speed based on the first included angle to generate a first adjusted speed, and the first adjusted speed is inversely proportional to the first included angle; Based on the information acquisition trajectory and the first adjusted speed, control the industrial line array camera to perform corresponding column pixel scanning operations.

6. The method for monitoring the coating of the inner cabin of a ship based on pixel scanning and laser scanning according to claim 4, wherein The method further includes: Determine the scanning direction of the two-dimensional lidar; Based on the two-dimensional laser scanning information, determine the plane angle of the current scanning plane inside the ship's cabin; Determine the second included angle between the scanning direction and the plane angle; Judge whether the second included angle is greater than a preset included angle value; If so, determine the first jitter frequency and the first jitter amplitude based on the second included angle; Based on the first jitter frequency and the first jitter amplitude, control the two-dimensional lidar to perform compensation scanning operations on the current scanning plane to generate compensated laser scanning information; Optimize the two-dimensional laser scanning information based on the compensated laser scanning information to generate optimized two-dimensional laser scanning information.

7. A method for monitoring the coating inside a ship's cabin based on pixel scanning and laser scanning according to claim 1, characterized in that, The analysis of the coating corrosion state of the ship's inner cabin based on the inner cabin image and the point cloud information to generate a coating monitoring result includes: Perform fusion processing on the inner cabin image and the point cloud information to obtain fused information; Based on the fused information, perform coating corrosion state analysis on the ship's inner cabin to generate corrosion height information and corrosion area information; Generate a coating monitoring result based on the corrosion height information and the corrosion area information.

8. A method for monitoring the coating inside a ship's cabin based on pixel scanning and laser scanning according to claim 7, characterized in that, The performing fusion processing on the inner cabin image and the point cloud information to obtain fused information includes: Extract the depth information of the ship's inner cabin based on the point cloud information; Perform three-dimensional conversion on the inner cabin image based on the depth information to generate fused information; or: Determine multiple scanning planes of the ship's inner cabin based on the point cloud information; Perform corrosion analysis on each scanning plane to generate preliminary corrosion information; Highlight the inner cabin image based on the preliminary corrosion information to generate fused information.

9. A method for monitoring the coating of the inner cabin of a ship based on pixel scanning and laser scanning according to claim 6, characterized in that, The method further includes: After generating the coating monitoring result, determine a suspected corrosion area based on the coating monitoring result; Adjust the preset shooting speed based on the suspected corrosion area to generate a second adjusted speed, and determine the second jitter frequency and the second jitter amplitude of the two-dimensional lidar based on the suspected corrosion area, where the second jitter frequency is greater than the first jitter frequency and the second jitter amplitude is greater than the first jitter amplitude; Based on the information acquisition trajectory and the second adjusted speed, control the industrial line array camera to perform corresponding column pixel scanning operations; Based on the first jitter frequency and the first jitter amplitude, control the two-dimensional lidar to perform compensation scanning operations on the current scanning plane.

10. A method for monitoring the coating of the inner cabin of a ship based on pixel scanning and laser scanning according to any one of claims 1-9, characterized in that, The method further includes: Obtain historical maintenance measures and their corresponding historical maintenance times and historical maintenance locations; Determine the maintenance method effectiveness coefficient according to the historical maintenance measures; Determine the maintenance time decay coefficient according to the historical maintenance measures and their corresponding historical maintenance times; Determine the space influence coefficient according to the historical maintenance location; Determine the corrosion risk prediction value according to the maintenance method effectiveness coefficient, the maintenance time decay coefficient, the space influence coefficient and the coating monitoring result; Generate a corrosion risk prediction map according to the corrosion risk prediction value.

Images