Container defect detection method, system, device, and medium

Abstract

The application provides a container defect detection method, system, device and medium, which can be applied to the field of intelligent technology of port containers. Based on a work starting signal, the application obtains multiple surface images and ranging data of a target container, then performs structured identification and segmentation based on a second surface image obtained through preprocessing to obtain a mask image, performs candidate detection according to the ranging data and the mask image to obtain defect candidate points, performs defect identification and segmentation on the defect candidate points to obtain defect identification results and defect contours, performs defect risk assessment according to the defect identification results and the defect contours to obtain a risk assessment result, so that the accuracy and real-time performance of container defect detection can be effectively improved, and then based on the risk assessment result, the surface image, the mask image and video data are packaged and processed, and the visual display content corresponding to the target container is controlled, so that relevant personnel can timely and accurately understand the defect information of the container.

Description

Technical Field

[0001] This invention relates to the field of intelligent port container technology, and in particular to a container defect detection method, system, equipment and medium. Background Technology

[0002] Traditional container defect detection methods primarily rely on manual visual inspection or random checks. These methods are highly subjective and lack consistent evidence, leading to low accuracy. Furthermore, images acquired in complex environments (such as low light, rain, fog, or backlighting) suffer from poor quality, further reducing the accuracy of the identification results. While existing systems can capture images or videos of containers, these images or videos are currently only used for post-event analysis, preventing relevant personnel from timely understanding of defects within the containers. Summary of the Invention

[0003] The main objective of this invention is to provide a method, system, equipment, and medium for detecting container defects, which can effectively improve the accuracy and real-time performance of container defect detection and enable relevant personnel to promptly understand container defect information.

[0004] To achieve the above objectives, one embodiment of the present invention provides a method for detecting defects in containers, the method comprising the following steps:

[0005] Obtain the operation start signal for the area where the target container is located;

[0006] Based on the operation start signal, acquire multiple first surface images of the target container and distance measurement data of the target container;

[0007] The first surface images are preprocessed to obtain the corresponding second surface images;

[0008] The second surface image is subjected to structured recognition and segmentation to obtain a mask image;

[0009] Based on the ranging data and the mask image, candidate detection is performed to obtain candidate defect points;

[0010] Defect candidate points are identified and segmented to obtain defect identification results and defect contours;

[0011] Based on the defect identification results and the defect profile, a defect risk assessment is performed to obtain the risk assessment results.

[0012] Based on the risk assessment results, the first surface image, the mask image, and the video data corresponding to the risk assessment results are packaged and processed to obtain a structured event;

[0013] The structured event controls the visualization of the cargo handling control system corresponding to the target container.

[0014] In some embodiments, acquiring multiple first surface images of the target container based on the operation start signal includes:

[0015] The image acquisition device is controlled according to the operation start signal to acquire multiple frames of images of multiple surfaces of the target container, thereby obtaining multiple frames of surface images;

[0016] The multiple surface images are filtered to obtain multiple images of the first surface.

[0017] In some embodiments, preprocessing the plurality of first surface images to obtain corresponding second surface images includes:

[0018] Multiple images of the first surface are sequentially subjected to distortion correction, dehazing, image stabilization, and high dynamic range synthesis to obtain the corresponding second surface image.

[0019] In some embodiments, the step of performing structured recognition and segmentation on the second surface image to obtain a mask image includes:

[0020] The container number, orientation, hazard markings, and seal status of the target container are identified based on the second surface image.

[0021] The mask image is determined in the second surface image based on the box number information, orientation information, hazard sign information, and seal status information.

[0022] In some embodiments, the step of performing candidate detection based on the ranging data and the mask image to obtain candidate defect points includes:

[0023] Generate the region of interest corresponding to the second surface image based on the mask image;

[0024] Based on the ranging data, the target area is located in the second surface image to obtain the target area, which includes a corner area, a weld area, a door seal area, or a panel area.

[0025] Based on the region of interest, candidate detection is performed in the target region to obtain defect candidate points.

[0026] In some embodiments, the step of performing a defect risk assessment based on the defect identification result and the defect profile to obtain a risk assessment result includes:

[0027] Count the number of defects corresponding to the defect identification results within the defect contour;

[0028] Obtain the confidence level of the defect identification results;

[0029] The defect severity score is calculated based on the geometric data of the defect profile, the number of defects, and the confidence level.

[0030] Based on the severity score of the defect, the defect level is classified to obtain the risk assessment result.

[0031] In some embodiments, controlling the visualization display content of the tallying control system corresponding to the target container based on the structured event includes:

[0032] The structured event is pushed to the review end;

[0033] Receive the review result returned by the review terminal;

[0034] Based on the verification results and the structured event control, the visualization display content of the cargo handling control system corresponding to the target container is adjusted.

[0035] Another embodiment of the present invention provides a container defect detection system, the system comprising:

[0036] The first acquisition module is used to acquire the operation start signal of the area where the target container is located;

[0037] The second acquisition module is used to acquire multiple first surface images of the target container and distance measurement data of the target container according to the operation start signal;

[0038] The preprocessing module is used to preprocess multiple first surface images respectively to obtain corresponding second surface images;

[0039] The first recognition and segmentation module is used to perform structured recognition and segmentation on the second surface image to obtain a mask image;

[0040] The detection module is used to perform candidate detection based on the ranging data and the mask image to obtain candidate defect points;

[0041] The second identification and segmentation module is used to identify and segment the defect candidate points to obtain defect identification results and defect contours.

[0042] The assessment module is used to perform a defect risk assessment based on the defect identification results and the defect profile, and obtain the risk assessment results.

[0043] The packaging module is used to package the first surface image, the mask image, and the video data corresponding to the risk assessment result based on the risk assessment result to obtain a structured event;

[0044] The control module is used to control the visual display content of the cargo handling control system corresponding to the target container based on the structured events.

[0045] Another aspect of the present invention provides an electronic device, the electronic device including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the above-described method.

[0046] Another aspect of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0047] The present invention provides the following beneficial effects: In this embodiment, after obtaining the operation start signal of the area where the target container is located, multiple first surface images of the target container and the distance measurement data of the target container are acquired. Then, the multiple first surface images are preprocessed to obtain corresponding second surface images. The second surface images are then structured and segmented to obtain mask images. Based on the distance measurement data and mask images, candidate detection is performed to obtain defect candidate points. Defect candidate points are then identified and segmented to obtain defect identification results and defect contours. Based on the defect identification results and defect contours, defect risk assessment is performed to obtain risk assessment results. This can effectively improve the accuracy and real-time performance of container defect detection. Then, based on the risk assessment results, the first surface images, mask images, and video data corresponding to the risk assessment results are packaged and processed to obtain structured events. Based on the structured events, the visualization display content of the cargo handling control system corresponding to the target container is controlled, thereby enabling relevant personnel to understand the defect information of the container in a timely and accurate manner. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0049] Figure 1 This is a flowchart of a container defect detection method provided in an embodiment of this application;

[0050] Figure 2 This is a schematic diagram of a container defect detection system provided in an embodiment of this application. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0052] It should be noted that, in the description of this application, the terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element. The terms "upper," "lower," etc., indicating orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, are used only for the convenience of describing this application and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application. Unless otherwise expressly specified and limited, the terms "installed," "connected," and "linked" should be interpreted broadly, for example, as a fixed connection, a detachable connection, or an integral connection; a mechanical connection or an electrical connection; a direct connection or an indirect connection through an intermediate medium; or a connection within two elements. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0053] The terms "first," "second," etc., used in this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class, without limiting the number of objects; for example, a first object can be one or more. Furthermore, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects have an "or" relationship.

[0054] In related technologies, traditional container defect detection methods mainly rely on manual visual inspection or random checks. This method is highly subjective and lacks consistent evidence, leading to low accuracy. Furthermore, images acquired in complex environments (such as low light, rain, fog, or backlight) have poor quality, further reducing the accuracy of the identification results. While existing systems can acquire images or videos of containers, currently, these images or videos are only used for post-event analysis, preventing relevant personnel from timely understanding of defects within the containers.

[0055] In view of this, embodiments of this application provide a container defect detection method, system, equipment, and medium, which can effectively improve the accuracy and real-time performance of container defect detection, and enable relevant personnel to promptly understand container defect information.

[0056] The embodiments of this application will be described in detail below with reference to the accompanying drawings:

[0057] Reference Figure 1 This application provides a method for detecting defects in containers, which includes, but is not limited to, steps S110 to S190:

[0058] Step S110: Obtain the operation start signal for the area where the target container is located;

[0059] Step S120: Acquire multiple first surface images of the target container and distance measurement data of the target container according to the operation start signal;

[0060] Step S130: Preprocess the multiple first surface images to obtain the corresponding second surface images;

[0061] Step S140: Perform structured recognition and segmentation on the second surface image to obtain a mask image;

[0062] Step S150: Perform candidate detection based on ranging data and mask image to obtain candidate defect points;

[0063] Step S160: Perform defect identification and segmentation on the defect candidate points to obtain the defect identification results and defect contours;

[0064] Step S170: Based on the defect identification results and defect profiles, conduct a defect risk assessment to obtain the risk assessment results;

[0065] Step S180: Based on the risk assessment results, the first surface image, mask image and video data corresponding to the risk assessment results are packaged and processed to obtain structured events;

[0066] Step S190: Based on the structured event control, control the visualization display content of the cargo handling control system corresponding to the target container.

[0067] It is understood that the operation start signal in this embodiment can refer to the lifting signal of the target container, the spreader working signal, or the trolley change status signal within the area where the target container is located. The lifting signal and the spreader working signal can be acquired in real time by a programmable logic controller (PLC). The trolley change status signal can be acquired in real time by an inertial measurement unit (IMU).

[0068] It is understood that in this embodiment, the acquisition of the operation start signal indicates that loading or unloading of the target container is required at the current time. Both loading and unloading processes involve moving the target container; if defects in the target container are not detected in time, the safety of the container during loading, unloading, and transportation will be affected. Therefore, in this embodiment, after acquiring the operation start signal, the image acquisition device can be controlled to acquire multiple frames of images of multiple surfaces of the target container, obtaining multiple surface images. These multiple frames are then filtered to obtain multiple first surface images, thus maintaining time synchronization between the loading / unloading process and the image acquisition process. Specifically, the image acquisition device in this embodiment can be a multi-view camera array. This multi-view camera array can acquire images of multiple surfaces of the target container, for example, acquiring images of the four sides and the top of the target container. In this embodiment, when acquiring surface images, multiple frames of surface images are acquired simultaneously, and the images are filtered based on their clarity, resulting in multiple frames of surface images with higher clarity as the first surface images.

[0069] It is understood that, in this embodiment, during the acquisition of multiple surface images, a ranging device is also controlled to measure the distance to the target container to obtain the distance data of the container corresponding to each surface image. This distance data includes, but is not limited to, the distance between target containers, the height of the target container's location, or the length, width, and height of the target container itself, thereby providing effective data support for subsequent defect localization.

[0070] It is understood that in this embodiment, after obtaining multiple first surface images, each first surface image is preprocessed to obtain a corresponding second surface image. Specifically, in this embodiment, the multiple first surface images can be sequentially subjected to distortion correction, dehazing, image stabilization, and high dynamic range synthesis to obtain the corresponding second surface image. Distortion correction calculates the distortion direction and degree of each pixel in each first surface image to obtain the ideal position of that point, effectively correcting radial and tangential distortion and restoring the first surface image to a near-realistic state. Dehazing uses physical models (such as atmospheric scattering theory) and algorithms (such as transmission models and illumination estimation) to eliminate problems such as reduced contrast and color distortion caused by fog in the first surface image. Image stabilization refers to the process of eliminating or reducing image sequence blur or jitter caused by camera or shooting equipment shake through technical means to obtain a clearer surface image. High dynamic range (HDR) synthesis processing combines multiple first surface illumination images with different exposure levels (typically taken by shooting the same scene at different apertures) and uses algorithms to merge the information from these images to generate a single image containing more detail and a wider brightness range. This embodiment, by employing different enhancement methods on the first surface images, can effectively improve the clarity of the surface image and highlight the feature information carried within it.

[0071] It is understood that in this embodiment, after obtaining the second surface image, a mask image is obtained by structured recognition and segmentation of the second surface image. Specifically, this embodiment can identify the container number, orientation, hazard sign, and seal status information of the target container based on the second surface image, and then determine the mask image in the second surface image based on the container number, orientation, hazard sign, and seal status information. The mask image defines the region of interest using a binary matrix, thereby enabling local image manipulation. This embodiment generates a corresponding mask image after identifying the container number, orientation, hazard sign, or seal status information carried in the second surface image, thereby providing geometric prior information for subsequent defect detection.

[0072] It is understood that, in this embodiment, after obtaining the corresponding mask image, candidate detection is performed based on the ranging data and the mask image to obtain candidate defect points. Specifically, this embodiment can generate a region of interest corresponding to the second surface image based on the mask image. Simultaneously, after locating the target region in the second surface image based on the ranging data, candidate detection is performed within the target region based on the region of interest to obtain candidate defect points. The target region includes corner fitting areas, weld seam areas, door seal areas, or panel areas.

[0073] It is understandable that corner fittings are key components in the container structure, primarily used for supporting, handling, and securing the container. Corner fittings are installed at the four corners of the container, serving as the intersection points of the uprights (corner posts) connecting the top and bottom corner fittings with the connecting components (lower end beams) at the ends of the container body. The area where the corner fittings are located in the second surface image is shown. Container welds are metal connections formed during the welding process, used to connect two or more metal components. In container manufacturing, welds are a crucial part of the container structure, ensuring overall strength and sealing. The weld area refers to the area where the welds are located in the second surface image. Container door seals are safety devices installed on container doors, primarily used to prevent unauthorized opening of goods during transport. The door seal area refers to the area where the door seal is located in the second surface image. Container panels refer to the actual plate-like outer surfaces and geometric properties of the six major panels that make up the container body, including the top panel, side panels, front panel, door panel, bottom panel, and the outer surface of the underframe structure. The panel area refers to the area where the panel is located in the second surface image.

[0074] It is understood that this embodiment obtains defect candidate points and performs defect identification on these candidate points. For example, this embodiment can determine the type of defect present at the defect candidate point based on the acquired image, such as surface depressions, surface cracks, surface corrosion, paint peeling, etc. The defect type identified at each defect candidate point can be one, two, or more defect types. After obtaining the defect identification result at each defect candidate point, this embodiment delineates the defect contour corresponding to that defect candidate point on the second surface image.

[0075] It is understood that in this embodiment, after obtaining the defect identification result and defect contour for each defect candidate point, a defect risk assessment is performed based on the defect identification result and defect contour to obtain the risk assessment result. Specifically, this embodiment can calculate the defect severity score by counting the number of defects corresponding to the defect identification result within the defect contour, obtaining the confidence level of the defect identification result, and then calculating the defect severity score based on the geometric data of the defect contour, the number of defects, and the confidence level. The defect severity score can be calculated using the following formula:

[0076] ;

[0077] In the formula, R represents the defect severity score, A represents the area corresponding to the defect contour, L represents the length corresponding to the defect contour, W represents the width corresponding to the defect contour, I represents the number of defects contained in the defect contour, conf represents the confidence level corresponding to the defect identification result within the defect contour, and α, β, γ, δ and ε all represent weighting coefficients, which can be adjusted according to the actual defect detection needs.

[0078] It is understood that in this embodiment, after obtaining the defect severity score, the defect level is classified according to the defect severity score to obtain the risk assessment result. Specifically, this embodiment can compare the calculated defect severity score with the level threshold. For example, when the defect severity score is less than the first level threshold, a defect may exist within the current defect contour, and a prompt message can be generated to alert the user; when the defect severity score is greater than or equal to the first level threshold and less than the second level threshold, a defect exists within the current defect contour, but the defect is not severe, so a yellow warning message is generated; when the defect severity score is greater than or equal to the second level threshold, a defect exists within the current defect contour and the defect is relatively severe, requiring attention from relevant personnel, so a red warning message is generated. The first level threshold is less than the second level threshold, and the size of the two level thresholds can be set according to actual detection needs. This embodiment, after classifying defects based on the defect severity score, generates a risk assessment result containing handling suggestions to provide relevant personnel with effective defect information and related suggestions.

[0079] It is understood that, after obtaining the risk assessment results, when the defect level is determined to be high based on the risk assessment results, the first surface image (e.g., an image containing five surfaces), the mask image, the related video data, and the keyframe image corresponding to the defect can be packaged and processed to generate a structured event (JSON) for subsequent data retrieval in the application process.

[0080] It is understood that, after obtaining a structured event, this embodiment can control the visualization display content of the corresponding tallying control system for the target container based on the structured event. Specifically, this embodiment can push the structured event to the review end, where relevant personnel can manually confirm or correct the structured event and write a review result. After receiving the review result returned by the review end, this embodiment controls the visualization display content of the corresponding tallying control system for the target container based on the review result and the structured event, so that relevant personnel can promptly understand the container's defect information and related handling suggestions. In addition, this embodiment can also store the structured event and the review result in a database, thereby using the data stored in the database to iteratively train the model corresponding to this implementation method, and continuously adjust the model parameters and detection thresholds to effectively improve the accuracy of defect detection.

[0081] For example, when the method of this embodiment detects losses during the unloading process based on the top view and four side views collected during implementation, if a dent is found on the surface of the large plate, a structured event is automatically packaged and pushed to manual review. When the method of this embodiment detects the loading door seal based on the top view and four side views collected during implementation, if an abnormality is found in the door seal texture, a structured event is automatically generated and pushed to manual review for confirmation before being entered into the warehouse.

[0082] As described above, the method in this embodiment, by automatically triggering multi-view automatic acquisition of container surface images, can effectively improve the accuracy and timeliness of defect detection results. Furthermore, by combining defect severity scores with confidence levels, automatic adjustment of risk levels can be achieved. Simultaneously, an automated closed-loop operation from acquisition to archiving is formed, reducing manual intervention and continuously optimizing the relevant models through a self-learning mechanism, thereby further improving the accuracy of defect detection results.

[0083] Reference Figure 2 This application provides a container defect detection system, the system comprising:

[0084] The first acquisition module is used to acquire the operation start signal of the area where the target container is located;

[0085] The second acquisition module is used to acquire multiple first surface images of the target container and distance measurement data of the target container based on the operation start signal;

[0086] The preprocessing module is used to preprocess multiple first surface images to obtain corresponding second surface images;

[0087] The first recognition and segmentation module is used to perform structured recognition and segmentation on the second surface image to obtain a mask image;

[0088] The detection module is used to perform candidate detection based on ranging data and mask images to obtain candidate defect points;

[0089] The second identification and segmentation module is used to identify and segment defect candidate points to obtain defect identification results and defect contours.

[0090] The assessment module is used to perform defect risk assessment based on defect identification results and defect profiles, and obtain risk assessment results.

[0091] The packaging module is used to package the first surface image, mask image and video data corresponding to the risk assessment results based on the risk assessment results, and obtain structured events.

[0092] The control module is used to control the visual display content of the cargo handling control system corresponding to the target container based on structured events.

[0093] It is understood that the content of the above method embodiments is applicable to this system embodiment. The specific functions implemented in this system embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0094] This application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement... Figure 1 The method shown.

[0095] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0096] This application provides a computer-readable storage medium storing a computer program, which is implemented when executed by a processor. Figure 1 The method shown.

[0097] It is understood that the content of the above method embodiments is applicable to this medium embodiment. The specific functions implemented in this medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0098] The device embodiments described above are merely illustrative. 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 modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0099] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0100] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical coding feature maps; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting defects in containers, characterized in that, The method includes the following steps: Obtain the operation start signal for the area where the target container is located; Based on the operation start signal, acquire multiple first surface images of the target container and distance measurement data of the target container; The first surface images are preprocessed to obtain the corresponding second surface images; The second surface image is subjected to structured recognition and segmentation to obtain a mask image; Based on the ranging data and the mask image, candidate detection is performed to obtain candidate defect points; Defect candidate points are identified and segmented to obtain defect identification results and defect contours; Based on the defect identification results and the defect profile, a defect risk assessment is performed to obtain the risk assessment results. Based on the risk assessment results, the first surface image, the mask image, and the video data corresponding to the risk assessment results are packaged and processed to obtain a structured event; The structured event controls the visualization of the cargo handling control system corresponding to the target container.

2. The method according to claim 1, characterized in that, The step of acquiring multiple first surface images of the target container based on the operation start signal includes: The image acquisition device is controlled according to the operation start signal to acquire multiple frames of images of multiple surfaces of the target container, thereby obtaining multiple frames of surface images; The multiple surface images are filtered to obtain multiple images of the first surface.

3. The method according to claim 1, characterized in that, The step of preprocessing multiple first surface images to obtain corresponding second surface images includes: Multiple images of the first surface are sequentially subjected to distortion correction, dehazing, image stabilization, and high dynamic range synthesis to obtain the corresponding second surface image.

4. The method according to claim 1, characterized in that, The step of performing structured recognition and segmentation on the second surface image to obtain a mask image includes: The container number, orientation, hazard markings, and seal status of the target container are identified based on the second surface image. The mask image is determined in the second surface image based on the box number information, orientation information, hazard sign information, and seal status information.

5. The method according to claim 1, characterized in that, The step of performing candidate detection based on the ranging data and the mask image to obtain candidate defect points includes: Generate the region of interest corresponding to the second surface image based on the mask image; Based on the ranging data, the target area is located in the second surface image to obtain the target area, which includes a corner area, a weld area, a door seal area, or a panel area. Based on the region of interest, candidate detection is performed in the target region to obtain defect candidate points.

6. The method according to claim 1, characterized in that, The step of performing a defect risk assessment based on the defect identification result and the defect profile to obtain a risk assessment result includes: Count the number of defects corresponding to the defect identification results within the defect contour; Obtain the confidence level of the defect identification results; The defect severity score is calculated based on the geometric data of the defect profile, the number of defects, and the confidence level. Based on the severity score of the defect, the defect level is classified to obtain the risk assessment result.

7. The method according to claim 1, characterized in that, The step of controlling the visualization display content of the tallying control system corresponding to the target container based on the structured event includes: The structured event is pushed to the review end; Receive the review result returned by the review terminal; Based on the verification results and the structured event control, the visualization display content of the cargo handling control system corresponding to the target container is adjusted.

8. A container defect detection system, characterized in that, The system includes: The first acquisition module is used to acquire the operation start signal of the area where the target container is located; The second acquisition module is used to acquire multiple first surface images of the target container and distance measurement data of the target container according to the operation start signal; The preprocessing module is used to preprocess multiple first surface images respectively to obtain corresponding second surface images; The first recognition and segmentation module is used to perform structured recognition and segmentation on the second surface image to obtain a mask image; The detection module is used to perform candidate detection based on the ranging data and the mask image to obtain candidate defect points; The second identification and segmentation module is used to identify and segment the defect candidate points to obtain defect identification results and defect contours. The assessment module is used to perform a defect risk assessment based on the defect identification results and the defect profile, and obtain the risk assessment results. The packaging module is used to package the first surface image, the mask image, and the video data corresponding to the risk assessment result based on the risk assessment result to obtain a structured event; The control module is used to control the visual display content of the cargo handling control system corresponding to the target container based on the structured events.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 7.

Images