A container intelligent crossing inspection system and method

Abstract

The present invention relates to the field of electronic technology applications, and discloses a container intelligent crossing inspection system and method. The container intelligent crossing inspection system includes: an image acquisition module for obtaining images of three sides of the container. A type acquisition module is used to obtain the container type. A travel sensor is used to obtain the x-axis coordinate of the container. A distance sensor is used to obtain the y and z coordinates of the container. An image stitching module is used to obtain the overall container surface map. A spatial construction module is used to form a three-dimensional container map and the three-dimensional coordinates of the container. A component classification module is used to obtain the component name, three-dimensional coordinates and send an abnormal signal. An outlier determination module is used to obtain the defect type. A microscopic camera is used to obtain the defect size and defect characteristics. A repair determination module is used to obtain repair measures. By constructing a coordinate system to locate each component of the container and comparing them through an exposure model, defects are accurately acquired and determined with high accuracy.

Description

Technical Field

[0001] The present invention relates to the field of electronic technology applications, and in particular to a container intelligent crossing damage inspection system and method. Background Art

[0002] In recent years, container throughput at major ports has shown a continuous upward trend, making intelligent port container operations a pressing task. Container damage inspection, as a crucial component of container operations, is an essential research area, yet it remains a largely understudied area. Currently, container inspection at major port gates primarily relies on manual inspection; at a few gates, staff utilize auxiliary equipment, but this is still essentially manual inspection. Manual inspection presents numerous challenges, such as requiring high levels of experience and a hazardous working environment. Summary of the Invention

[0003] In order to solve the above technical problems, the present invention provides a container intelligent crossing inspection system, the container intelligent crossing inspection system comprising:

[0004] There are several image acquisition modules, which are used to capture the two sides and the top of the container to obtain images of the three sides of the container;

[0005] A type acquisition module is used to receive an image of a container and identify the image identifier on the container to obtain the container type;

[0006] A travel sensor is used to sense the x-axis travel of the container and obtain the x-axis coordinate of the container;

[0007] A distance sensor, which is used to sense the outline of the container and obtain the y and z coordinates of the container;

[0008] An image stitching module, which is used to receive images of the container and stitch the images of the container to obtain an overall container surface image;

[0009] A spatial construction module is used to receive the overall container surface map and the container surface contour coordinates, perform standardization on the overall container surface map, and combine and construct a three-dimensional container map and the three-dimensional coordinates of the container;

[0010] The component classification module is used to receive the 3D coordinates and 3D container map of the container, analyze them to obtain component features and component maps; search a pre-set feature-component information table to obtain the component name; search a pre-set name, type, and component model library to obtain the exposed model; determine whether the component map and the exposed model are the same; if not, obtain the component name and 3D coordinates and send an abnormal signal;

[0011] The outlier determination module is used to receive the abnormal signal and the component diagram, search a pre-set component diagram-defect library, and obtain the defect type;

[0012] A microscopic camera is used to receive abnormal signals, amplify the abnormal three-dimensional coordinates, and obtain a magnified image of the defect; and measure the magnified image of the defect to obtain the defect size and defect characteristics;

[0013] The repair determination module is used to receive the defect size and defect characteristics, search a preset size, characteristic-measure information table, and obtain the repair measures.

[0014] Preferably, the container intelligent crossing inspection system further includes:

[0015] Time acquisition module, which counts the current time point to obtain the time point T i ;

[0016] Time determination module, receiving time point T i and time point T i-1 , time point T i-1 Get the time obtained by the last time acquisition module and get the time point T i-1 The corresponding image of the coordinate and bound to the time period T i-1 、T i Picture of coordinates.

[0017] Preferably, the micro camera is mounted on the second support frame, the sliding drive member is mounted on the second support frame and connected to the micro camera, and the sliding drive member is used to drive the micro camera to slide to the abnormal three-dimensional coordinates of the container.

[0018] Preferably: the sliding drive component includes a sliding groove, a motor and a screw rod, the sliding groove is opened on the inner side of the second support frame, the screw rod is rotatably arranged inside the sliding groove, a mounting seat is slidingly arranged inside the sliding groove, the mounting seat is nested in the screw rod, the motor is installed on the second support frame, and the output shaft of the motor is coaxially fixedly connected to the screw rod.

[0019] Preferably, a guide rod is fixedly provided on the second support frame, the direction of the guide rod is consistent with the direction of the screw rod, and a linear bearing is provided inside the mounting seat, and the linear bearing is slidably nested on the guide rod.

[0020] The present invention also provides a container intelligent crossing inspection method, which is applied to the above-mentioned container intelligent crossing inspection system. The container intelligent crossing inspection method comprises the following steps:

[0021] S1. Obtain an image of a container, identify the image identifier on the container, and obtain the container type;

[0022] S2. Obtain the x-axis displacement of the container and the y and z coordinates of the container;

[0023] S3. stitching the container images to obtain an overall container surface map;

[0024] S4. Standardize the entire container surface image and combine them to form a three-dimensional container image and three-dimensional coordinates of the container;

[0025] S5. Analyze the three-dimensional container image and the three-dimensional coordinates of the container to obtain component features and a component image;

[0026] S6. Search a pre-set feature-component information table to obtain the component name;

[0027] S7, searching a preset name and type-component model library to obtain an exposed model;

[0028] S8, determine whether the component drawing is the same as the exposed model, if not, execute S9;

[0029] S9, searching a pre-set component diagram-defect library to obtain the defect type;

[0030] S10, magnifying the abnormal three-dimensional coordinates and obtaining a magnified image of the defect, and measuring the magnified image of the defect to obtain the defect size and defect characteristics;

[0031] S11. Search a preset size, feature-measure information table to obtain repair measures.

[0032] Preferably, in S3, the container image is spliced ​​by segmenting the container image, segmenting the container image by the acquisition focus (C i 、X i ) as the center, and obtain a rectangular tissue image with a length of c0, C i is the installation coordinate of the image acquisition module; when the image acquisition module is installed vertically, C i Z i ; When the image acquisition module is installed horizontally, C i Y i ;X i is the travel distance of the container. When the travel sensor senses that the container's △X is c0, the travel sensor sends a shooting signal, and the image acquisition module receives the shooting signal and takes a picture. Where c0 is the side length of the acquisition range of the image acquisition module, and C i+1 =C i+ c0,X i+1 =X i+ c0.

[0033] Preferably, in S4, the specific steps of the method for standardizing the entire container surface image include:

[0034] S41. Confirm the original drawing of the overall container surface;

[0035] S42, performing background separation on the image to obtain a container image;

[0036] S43. Search a preset type-shape standard drawing library to obtain a standard container drawing;

[0037] S44. Compare the container drawing with the standard container drawing, and modify the container drawing to obtain a container body drawing;

[0038] S45. Smooth the box body marking.

[0039] The present invention also provides a computer terminal, which is used to implement the steps of the intelligent container crossing damage inspection method when executing a computer program stored in a memory.

[0040] The present invention also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of the above-mentioned container intelligent crossing inspection method are implemented.

[0041] The technical effects and advantages of this invention include: By constructing a coordinate system to locate each container component and comparing it with the exposure model, defects can be accurately captured and determined with high accuracy. This facilitates simple analysis of abnormal locations, avoiding interference from the organization, and ensuring clear observation. To facilitate search and comparison, component drawings can be processed to make defects more visible, further facilitating comparison. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 This is a structural block diagram of a container intelligent crossing damage inspection system proposed by the present invention.

[0043] Figure 2 This is a schematic diagram of the three-dimensional structure of a container intelligent crossing damage inspection system proposed by the present invention.

[0044] Figure 3 This is a schematic diagram of the top view of the structure of a container intelligent crossing damage inspection system proposed by the present invention.

[0045] Figure 4 for Figure 3 Schematic diagram of the partial cross-sectional structure of section A.

[0046] Figure 5 This is a flow chart of a container intelligent crossing damage inspection method proposed by the present invention.

[0047] Figure 6 This is a flow chart of a standardized processing method in a container intelligent crossing inspection method proposed by the present invention.

[0048] Explanation of the accompanying reference numerals: monocular camera 1, travel sensor 2, microscope camera 3, sliding drive 4, first support frame 5, second support frame 6, sliding groove 7, screw rod 8, motor 9, guide rod 10, distance sensor 11. DETAILED DESCRIPTION

[0049] The present invention will be described in further detail below with reference to the accompanying drawings and specific embodiments. The embodiments of the present invention are provided for purposes of illustration and description and are not intended to be exhaustive or to limit the invention to the disclosed forms. Many modifications and variations will be apparent to those skilled in the art. The embodiments are chosen and described to better illustrate the principles of the invention and its practical application, and to enable those skilled in the art to understand the invention and design various embodiments with various modifications suitable for specific applications. Example

[0050] refer to Figure 1~Figure 2 In this embodiment, a container intelligent crossing inspection system is proposed for inspecting containers for damage. The container intelligent crossing inspection system includes:

[0051] There are several image acquisition modules, which are located on both sides and above the container channel, and are used to shoot the two side faces and top face of the container passing through the channel, thereby obtaining three images of the container. The image acquisition module can be a monocular camera 1, a binocular camera, etc., and is installed on the first support frame 5. The first support frame 5 is an inverted concave structure, forming three mounting rods, which can meet the installation requirements of the three faces of the image acquisition module. Multiple image acquisition modules can be installed on each mounting rod, so that comprehensive acquisition can be performed. The image acquisition module can be tilted, and its direction is the same as and opposite to the flow direction of the container. When the container passes through the image acquisition module according to the flow direction, the front face, top, two side faces, and rear end face of the container can be photographed, so that the container image can be fully acquired.

[0052] The type acquisition module is used to receive the image of the container and identify the image mark on the container to obtain the container type. The container mark can be a type mark written on the surface of the container, a container identification plate, etc. The details are not detailed here.

[0053] The travel sensor 2 can be mounted on the first support frame 5 or other supporting structure to sense the travel of the container in the flow direction. The container flow direction can be used as the positive x-axis direction, and the x-axis coordinate of the container can be obtained by sensing through the travel sensor 2. The x-axis coordinate origin can be the front face of the container, but this is not a limitation.

[0054] Time acquisition module, which counts the current time point to obtain the time point T i .

[0055] Distance sensor 11 can be mounted on first support frame 5 to sense the distance between the container outline and the first support frame 5. Using an axis along the container's flow path as a reference, with the distance L1 between the first support frame 5 and the axis, and the distance a detected by distance sensor 11, the distance from the container outline to the axis can be determined, thereby obtaining the container's y and z coordinates. While image processing can also be used to obtain container coordinates, this method can result in significant errors and requires a large amount of processing, making it inconvenient.

[0056] The image stitching module is used to receive images of the container and stitch the images together to obtain an overall container surface image. The image acquisition module captures images of the container. Because it is necessary to improve clarity and reduce the volume of the first support frame 5, the area captured by the image acquisition module is reduced to improve resolution. The overall container surface image is stitched together to increase its completeness. The stitching method of the image stitching module is to segment the container image, segmenting it based on the acquisition focus (C i 、X i ) as the center, and obtain a rectangular tissue image with a length of c0, C i is the installation coordinate of the image acquisition module. When the image acquisition module is installed vertically, C i Z i When the image acquisition module is installed horizontally, C i Y i Since the image acquisition modules are of the same type and have the same installation method, the details will not be described here. i is the travel distance of the container. When the travel sensor 2 senses that the container's △X is c0, the travel sensor 2 sends a shooting signal, and the image acquisition module receives the shooting signal and takes a picture. Where c0 is the side length of the acquisition range of the image acquisition module, and C i+1 =C i+ c0,X i+1 =X i+ The container surface image is spliced ​​according to the focal position and acquired at a fixed interval to achieve omnidirectional container image acquisition. The details are not detailed here.

[0057] The spatial construction module receives the overall container surface map and the container surface contour coordinates, and standardizes the overall container surface map and combines them to form a three-dimensional container map and the three-dimensional coordinates of the container. The specific steps of the method for standardizing the overall container surface map include: 1. Confirming the original image of the overall container surface map, 2. Performing background separation on the image to obtain the container map. 3. Searching for a pre-set type-shape standard library by container type to obtain the container standard map. The type-shape standard library can be manually established, and the containers can be manually classified according to the shape of the container. The details will not be repeated here. 4. Compare the container map with the container standard map, and correct the container map to obtain the box body map. For example, the container image may have wrinkles, distortions, deviations, tilts, etc., and compared with the container standard map, the container image is corrected to ensure that it is consistent with the container standard. Figure 1 5. Smoothing: removes any spots or colors on the container body drawing that are not outside the standard container drawing to create a plan view. The details are not detailed here.

[0058] The component classification module is used to receive the three-dimensional coordinates and three-dimensional container diagram of the container, analyze them to obtain component features and component diagrams, search a pre-set feature-component information table based on the component features to obtain the component name, search a pre-set name, type-component model library based on the component name and container type to obtain the exposed model, compare the component diagram with the exposed model, and determine whether the component diagram and the exposed model are the same. If not, obtain the component name, three-dimensional coordinates, and send an abnormal signal. Component features may include shape, coordinate position, etc. For example, based on the different coordinates and shapes, it can be determined to be a rear pillar, rear lintel, J-shaped column, upper beam, corner piece, side panel, bottom beam, box door, nameplate, lock, door handle bracket, hinge, etc. The details are not repeated here. The exposed model can be obtained based on the installation method and component model, the details are not repeated here.

[0059] The outlier determination module receives the abnormal signal and the component diagram, and searches a pre-set component diagram-defect library through the component diagram to obtain the defect type. The component diagram-defect library can be manually created and can be updated in real time. The defect type can be dents, light leakage, rust, bending, deformation, tearing, etc. It can also be stickers, debris on the bottom of the box, oil stains, scratches, dust, etc., which will not be described in detail here. By extracting the component diagram and comparing it with the component model and the exposed model, the comparison is clear. This facilitates a simple analysis of the abnormal location points, avoids the interference of the organization on the abnormal location points, and makes the observation clear. In order to facilitate the search and comparison, the component diagram can be processed to make the defects visible, making it easier to compare. The processing can be to increase the brightness, contrast, etc., which will not be described in detail here.

[0060] Microscopic camera 3, which can be mounted on second support frame 6 downstream of the container flow, receives abnormality signals, amplifies the abnormal three-dimensional coordinates, and obtains a magnified image of the defect. The magnified image is then measured to obtain defect dimensions and characteristics. Defect dimensions can include shape, length, and width. Defect characteristics include edge condition, tear curling direction, presence of rust on the light-leaking edge, and rust depth.

[0061] refer to Figure 3~Figure 4 The sliding drive member 4 is mounted on the second support frame 6 and connected to the micro camera 3, and is used to drive the micro camera 3 to slide to the abnormal three-dimensional coordinates of the container. The sliding drive member 4 can be a hydraulic rod, an electric telescopic rod, a pneumatic rod, etc., and can also include a sliding groove 7, a motor 9 and a screw 8. The sliding groove 7 is opened on the inner side of the second support frame 6, and the screw 8 is rotatably set inside the sliding groove 7. The interior of the sliding groove 7 is provided with a mounting seat that slides, and the mounting seat is nested in the screw 8. The motor 9 is mounted on the second support frame 6, and the output shaft of the motor 9 is coaxially fixedly connected to the screw 8. Under the drive of the motor 9, the screw 8 rotates inside the sliding groove 7, thereby adjusting the position of the micro camera 3. A guide rod 10 is fixedly provided on the second support frame 6. The direction of the guide rod 10 is consistent with the direction of the screw 8. A linear bearing is provided inside the mounting seat, and the linear bearing is slidably nested in the guide rod 10 for guiding.

[0062] Time determination module, receiving time point T i and time point T i-1 , where time point T i-1 Get the time obtained by the last time acquisition module and get the time point T i-1 The corresponding image of the coordinate and bound to the time period T i-1 、T i A picture of the coordinates. This makes it easier to find.

[0063] The repair determination module receives defect size and characteristics and searches a pre-set size, characteristic, and action information table based on these dimensions and characteristics to determine the repair action. These repair actions may include repair methods, materials, time required, and budget, though details are omitted here. Example

[0064] refer to Figure 5 In this embodiment, a container intelligent crossing inspection method is proposed, including the following steps:

[0065] S1. Obtain an image of a container, identify the image logo on the container, and obtain the container type.

[0066] S2. Obtain the x-axis displacement of the container and the y and z coordinates of the container.

[0067] S3. Stitching the container images to obtain an overall container surface map.

[0068] S4. Standardize the entire container surface map and combine them to form a three-dimensional container map and three-dimensional coordinates of the container.

[0069] S5. Analyze the three-dimensional container diagram and the three-dimensional coordinates of the container to obtain component features and a component diagram.

[0070] S6. Search a preset feature-component information table to obtain the component name.

[0071] S7. Search a preset name and type-component model library to obtain the exposed model.

[0072] S8. Determine whether the component drawing is the same as the exposed model. If not, execute S9.

[0073] S9. Search a preset component diagram-defect library to obtain the defect type.

[0074] S10, magnifying the abnormal three-dimensional coordinates and obtaining a magnified image of the defect, and measuring the magnified image of the defect to obtain the defect size and defect characteristics.

[0075] S11. Search a preset size, feature-measure information table to obtain repair measures.

[0076] In S3, the image stitching module stitching method is:

[0077] The container image is segmented, and the segmentation is based on the acquisition focus of the image acquisition module (C i 、X i ) as the center, and obtain a rectangular tissue image with a length of c0, C i is the installation coordinate of the image acquisition module; when the image acquisition module is installed vertically, C i Z i When the image acquisition module is installed horizontally, C i Y i ;X i is the travel distance of the container. When the travel sensor 2 senses that the container's △X is c0, the travel sensor 2 sends a shooting signal, and the image acquisition module receives the shooting signal and takes a picture. Where c0 is the side length of the acquisition range of the image acquisition module, and C i+1 =C i+ c0,X i+1 =X i+ c0.

[0078] refer to Figure 6,In S4, the specific steps of the method for standardizing the overall container surface map include:

[0079] S41. Confirm the original drawing of the overall container surface;

[0080] S42, performing background separation on the image to obtain a container image;

[0081] S43. Search a preset type-shape standard drawing library to obtain a standard container drawing;

[0082] S44. Compare the container drawing with the standard container drawing, and modify the container drawing to obtain a container body drawing;

[0083] S45. Smooth the box body marking. Example

[0084] As a preferred embodiment of the present invention, a computer terminal is provided, configured to implement the steps of the aforementioned intelligent container crossing inspection method when executing a computer program stored in a memory. The computer terminal may be a computer, a smartphone, a control system, or other IoT device. The intelligent container crossing inspection method may also be designed as an embedded program and installed on a computer terminal, such as a single-chip microcomputer.

[0085] As a preferred embodiment of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program. When executed by a processor, the program implements the steps of the method for verifying the integrity of containers at intelligent level crossings. The method for verifying the integrity of containers at intelligent level crossings can be implemented in the form of software, such as a program designed to run independently on a computer-readable storage medium. The computer-readable storage medium can be a USB flash drive, designed as a USB shield, and the USB flash drive can be configured to initiate the entire method program via an external trigger.

[0086] When the computer program is executed by the processor, the steps of the above-mentioned container intelligent crossing inspection method are implemented.

[0087] Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field and related fields without making creative efforts should fall within the scope of protection of the present invention. Structures, devices, and operating methods not specifically described and explained in the present invention shall be implemented in accordance with conventional means in the field unless otherwise specified or limited.

Claims

1. A container intelligent crossing inspection system, characterized by: The container intelligent crossing inspection system includes: There are several image acquisition modules, which are used to capture the two sides and the top of the container to obtain images of the three sides of the container; A type acquisition module is used to receive an image of a container and identify the image identifier on the container to obtain the container type; A travel sensor is used to sense the x-axis travel of the container and obtain the x-axis coordinate of the container; A distance sensor, which is used to sense the outline of the container and obtain the y and z coordinates of the container; An image stitching module, which is used to receive images of the container and stitch the images of the container to obtain an overall container surface image; A spatial construction module is used to receive the overall container surface map and the container surface contour coordinates, perform standardization on the overall container surface map, and combine and construct a three-dimensional container map and the three-dimensional coordinates of the container; The component classification module is used to receive the 3D coordinates and 3D container map of the container, analyze them to obtain component features and component maps; search a pre-set feature-component information table to obtain the component name; search a pre-set name, type, and component model library to obtain the exposed model; determine whether the component map and the exposed model are the same; if not, obtain the component name and 3D coordinates and send an abnormal signal; The outlier determination module is used to receive the abnormal signal and the component diagram, search a pre-set component diagram-defect library, and obtain the defect type; A microscopic camera is used to receive abnormal signals, amplify the abnormal three-dimensional coordinates, and obtain a magnified image of the defect; and measure the magnified image of the defect to obtain the defect size and defect characteristics; A repair determination module is configured to receive defect size and defect characteristics, search a preset size, characteristic-measure information table, and obtain a repair measure; Time acquisition module, which counts the current time point to obtain the time point T i ; Time determination module, receiving time point T i and time point T i-1 , time point T i-1 Get the time obtained by the last time acquisition module and get the time point T i-1 The corresponding image of the coordinate and bound to the time period T i-1 、T i Picture of coordinates; The micro camera is mounted on the second support frame, the sliding drive member is mounted on the second support frame and connected to the micro camera, and the sliding drive member is used to drive the micro camera to slide to the abnormal three-dimensional coordinates of the container; The sliding drive member includes a sliding groove, a motor and a screw rod. The sliding groove is opened on the inner side of the second support frame. The screw rod is rotatably arranged inside the sliding groove. A mounting seat is slidingly provided inside the sliding groove. The mounting seat is nested in the screw rod. The motor is installed on the second support frame. The output shaft of the motor is coaxially fixedly connected to the screw rod. A guide rod is fixedly provided on the second support frame, and the direction of the guide rod is consistent with the direction of the screw rod. A linear bearing is provided inside the mounting seat, and the linear bearing is slidably nested on the guide rod.

2. A container intelligent crossing inspection method, applied to the container intelligent crossing inspection system according to claim 1, the container intelligent crossing inspection method comprising the following steps: S1. Obtain an image of a container, identify the image identifier on the container, and obtain the container type; S2. Obtain the x-axis displacement of the container and the y and z coordinates of the container; S3. stitching the container images to obtain an overall container surface map; S4. Standardize the entire container surface image and combine them to form a three-dimensional container image and three-dimensional coordinates of the container; S5. Analyze the three-dimensional container image and the three-dimensional coordinates of the container to obtain component features and a component image; S6. Search a pre-set feature-component information table to obtain the component name; S7, searching a preset name and type-component model library to obtain an exposed model; S8, determine whether the component drawing is the same as the exposed model, if not, execute S9; S9, searching a pre-set component diagram-defect library to obtain the defect type; S10, magnifying the abnormal three-dimensional coordinates and obtaining a magnified image of the defect, and measuring the magnified image of the defect to obtain the defect size and defect characteristics; S11. Search a preset size, feature-measure information table to obtain repair measures.

3. A container intelligent crossing inspection method according to claim 2, characterized in that: In S3, the container image is spliced ​​by segmenting the container image, segmenting the container image by the acquisition focus (C i 、X i ) as the center, and obtain a rectangular tissue image with a length of c0, C i is the installation coordinate of the image acquisition module; when the image acquisition module is installed vertically, C i Z i ; When the image acquisition module is installed horizontally, C i Y i ;X i is the travel distance of the container. When the travel sensor senses that the container's △X is c0, the travel sensor sends a shooting signal, and the image acquisition module receives the shooting signal and takes a picture. Where c0 is the side length of the acquisition range of the image acquisition module, and C i+1 =C i+ c0,X i+1 =X i+ c0.

4. A container intelligent crossing inspection method according to claim 2, characterized in that: In S4, the specific steps of the method for standardizing the entire container surface map include: S41. Confirm the original drawing of the overall container surface; S42, performing background separation on the image to obtain a container image; S43. Search a preset type-shape standard drawing library to obtain a standard container drawing; S44. Compare the container drawing with the standard container drawing, and modify the container drawing to obtain a container body drawing; S45. Smooth the box body marking.

5. A computer terminal, characterized in that: The computer terminal is used to implement the steps of a container intelligent crossing inspection method as described in any one of claims 2 to 4 when executing the computer program stored in the memory.

6. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the program is executed by the processor, the steps of the container intelligent crossing inspection method according to any one of claims 2 to 4 are implemented.

Images