Truck number recognition camera distributed installation assembly for a reach stacker

Abstract

This utility model discloses a distributed installation assembly for truck license plate recognition cameras used in reach stackers, comprising at least two flatcar license plate recognition cameras, one open wagon license plate recognition camera, and a distributed collaborative control module. The flatcar license plate recognition cameras are installed at different positions on the front of the reach stacker, while the open wagon license plate recognition cameras are installed on the front of the reach stacker and at a higher height than the flatcar license plate recognition cameras. The distributed collaborative control module is communicatively connected to each camera, controlling the cameras to acquire and process license plate images. This distributed installation assembly can adapt to the license plate recognition needs of different types of trucks, eliminate blind spots, improve the accuracy of license plate recognition, and enhance the overall efficiency and management level of container loading and unloading operations.

Description

[0001] This application claims priority to Chinese patent application No. 202521331598.2, filed on June 26, 2025, entitled “Distributed Mounting Assembly for Truck License Plate Recognition Cameras for Front-end Cranes”, the entire contents of which are incorporated herein by reference. Technical Field

[0002] This utility model relates to the field of container terminal equipment, specifically to a distributed installation component for truck number recognition cameras used for reach stackers. Background Technology

[0003] With the rapid development of container shipping, railway freight is playing an increasingly important role in the modern logistics system. During container loading and unloading operations, reach stackers, as the primary loading and unloading equipment, directly impact the operational quality of the entire logistics chain through their efficiency and accuracy. Accurate identification of freight car numbers is one of the key links in ensuring the smooth operation of container loading and unloading.

[0004] Currently, railway freight car number identification mainly employs two methods: manual visual identification and fixed camera identification. Manual visual identification suffers from low efficiency, error-proneness, and high labor intensity; while fixed camera identification, although improving efficiency to some extent, still faces numerous technical challenges. First, traditional fixed camera identification systems typically use a single camera setup. Due to the significant differences in the height and angle of car number positions for flatcars and open wagons, it is difficult to simultaneously cover the car number positions of different types of vehicles, resulting in blind spots. Second, in reach stacking operations, due to the diverse types of vehicles, varying car number positions, and complex operating environment, a single camera is insufficient to adapt to various identification scenarios. Utility Model Content

[0005] In view of this, the present invention provides a distributed installation component for truck license plate recognition cameras for front-end cranes, which can adapt to various recognition scenarios, eliminate blind spots, improve the accuracy of license plate recognition, and enhance the overall efficiency and management level of container loading and unloading operations.

[0006] This utility model embodiment provides a distributed installation component for truck license plate recognition cameras used in front-mounted cranes, the distributed installation component comprising:

[0007] At least two flatcar number recognition cameras are installed at different positions on the front of the front-end crane;

[0008] A wagon license plate recognition camera is installed at the front of the front-end crane, and the height of the wagon license plate recognition camera is higher than the height of the flatcar license plate recognition camera.

[0009] The distributed collaborative control module is communicatively connected to the flatbed car license plate recognition camera and the open car license plate recognition camera. The distributed collaborative control module controls the flatbed car license plate recognition camera or the open car license plate recognition camera to acquire and process license plate images.

[0010] Optionally, the distributed installation component includes two flatcar number recognition cameras, symmetrically arranged on both sides of the cab of the front-mounted crane, with the horizontal field of view coverage of the two flatcar number recognition cameras overlapping by ≥30%.

[0011] Optionally, the horizontal field of view of the two flatbed vehicle number recognition cameras is ±45°.

[0012] Optionally, the vertical field of view of the flatbed vehicle number recognition camera covers the top of the flatbed vehicle by ±30°.

[0013] Optionally, the distributed installation assembly further includes a support device fixed to the left side of the cab of the front-mounted crane, and the open wagon license plate recognition camera is fixed to the support device.

[0014] Optionally, the support device includes a first support tube, a second support tube, and an adjusting member. The first support tube is slidably connected inside the second support tube, and the adjusting member passes through the second support tube and abuts against or is separated from the outer wall of the first support tube to fix or adjust the height of the support device.

[0015] Optionally, the support device further includes a universal joint bracket fixed to the top of the first support tube, and the open vehicle license plate recognition camera is fixed on the universal joint bracket.

[0016] Optionally, the distributed collaborative control module includes:

[0017] The vehicle model recognition submodule is used to determine the vehicle type based on the vehicle number image;

[0018] The parameter optimization submodule is used to adjust the parameters of the flatbed vehicle number recognition camera or the open vehicle number recognition camera according to the vehicle type.

[0019] The data fusion submodule is used to process vehicle number images captured by at least two flatbed vehicle number recognition cameras to generate a complete vehicle number image;

[0020] Edge computing unit, used to identify vehicle numbers in vehicle number images;

[0021] The cloud-based collaborative unit is used to compare the identified vehicle number with the preset vehicle number database and send back the verification results.

[0022] This utility model embodiment provides a distributed installation component for truck license plate recognition cameras used in reach stackers, including at least two flatcar license plate recognition cameras, one open wagon license plate recognition camera, and a distributed collaborative control module. The flatcar license plate recognition cameras are installed at different positions on the front of the reach stacker, while the open wagon license plate recognition cameras are installed on the front of the reach stacker and at a higher height than the flatcar license plate recognition cameras. The distributed collaborative control module is communicatively connected to each camera, controlling the cameras to acquire and process license plate images. This distributed installation component can adapt to the license plate recognition needs of different types of trucks, eliminate blind spots, improve the accuracy of license plate recognition, and enhance the overall efficiency and management level of container loading and unloading operations. Attached Figure Description

[0023] The above and other objects, features and advantages of the present invention will become clearer from the following description of embodiments of the present invention with reference to the accompanying drawings, in which:

[0024] Figure 1 This is an overall installation diagram of the distributed installation component according to an embodiment of the present utility model;

[0025] Figure 2 This is a partial installation diagram of the distributed installation component according to an embodiment of the present invention.

[0026] Figure label:

[0027] 1- Flat wagon number recognition camera; 2- Open wagon number recognition camera; 3- Front-end lifting equipment; 4- Driver's cab; 5- Support device. Detailed Implementation

[0028] The present application is described below based on embodiments, but it is not limited to these embodiments. In the detailed description of the present application below, certain specific details are described in detail. Those skilled in the art can fully understand the present application without these details. To avoid obscuring the substance of the present application, well-known methods, processes, flows, elements, and circuits are not described in detail.

[0029] Furthermore, those skilled in the art should understand that the accompanying drawings provided herein are for illustrative purposes only and are not necessarily drawn to scale.

[0030] Unless otherwise expressly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise expressly limited. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0031] Unless the context explicitly requires it, words such as "including" or "contains" throughout the application should be interpreted as including rather than exclusive or exhaustive; that is, meaning "including but not limited to".

[0032] In the description of this application, it should be understood that the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Furthermore, in the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0033] This embodiment provides a distributed installation component for truck license plate recognition cameras used in front-end cranes, such as... Figure 1 and Figure 2 As shown, the distributed installation component includes at least two flatcar license plate recognition cameras 1, one open wagon license plate recognition camera 2, and a distributed collaborative control module. Flatcar license plate recognition camera 1 is used to capture images of the flatcar license plate numbers, and open wagon license plate recognition camera 2 is used to capture images of the open wagon license plate numbers, thus identifying the license plate numbers of both flatcars and open wagons for easy collection and management of logistics information. Since the front of the reach stacker 3 is close to the vehicle during reach stacker operations, placing the flatcar license plate recognition camera 1 and the open wagon license plate recognition camera 2 at the front of the reach stacker 3 allows for clear and complete acquisition of license plate images, avoiding obstruction issues caused by the reach stacker 3.

[0034] Since the vehicle numbers of different flatcars are distributed in different locations, such as the front, middle, or rear, the distributed installation component in this embodiment sets up at least two flatcar number recognition cameras 1, which are installed at different positions on the front of the reach stacker 3. This allows for the capture and recognition of vehicle numbers from multiple angles, avoiding blind spots that may occur with single-angle recognition and thus improving recognition accuracy. Since the vehicle number positions of open wagons are relatively fixed, installing one open wagon number recognition camera 2 on the front of the reach stacker 3 is sufficient to capture a complete image including the vehicle number. The distributed collaborative control module communicates with both the flatcar number recognition camera 1 and the open wagon number recognition camera 2. The distributed collaborative control module controls either the flatcar number recognition camera 1 or the open wagon number recognition camera 2 to acquire and process vehicle number images to identify vehicle numbers of different vehicle types, facilitating management.

[0035] In one specific implementation, the distributed installation component includes two flatcar number recognition cameras 1, symmetrically arranged on both sides of the cab 4 of the front-mounted crane 3. This location offers a wide field of view and a high degree of matching with the flatcar number position, allowing for sufficient capture of the flatcar number information. The two flatcar number recognition cameras 1 work together to enhance the redundancy of flatcar number recognition, ensuring coverage of the flatcar's number marking area and meeting the number capture requirements for flatcars of different widths.

[0036] Preferably, after installation, the two flatbed vehicle number recognition cameras 1 must have a horizontal field-of-view coverage overlap of ≥30%, which can eliminate blind spots in flatbed vehicle number recognition, ensure that vehicle number information can be clearly captured by at least one camera, and improve the accuracy of recognition. The existence of the overlapping area also improves the system's fault tolerance; even if one camera fails to clearly capture the vehicle number due to lighting, angle, or other reasons, the other camera may still obtain a clear image. In addition, the aforementioned overlap value can be adjusted and set according to actual operational needs.

[0037] Preferably, the horizontal field of view of the two flatbed vehicle number recognition cameras 1 is ±45°, which meets the vehicle number capture requirements of flatbed vehicles of different widths. In addition, the size of the horizontal field of view of the flatbed vehicle number recognition cameras 1 can be selected from different types and models of cameras according to requirements.

[0038] Furthermore, the flatbed vehicle license plate recognition camera 1 is adjusted to shoot downwards (i.e., by shooting from a downward angle) to capture the license plate image, which can eliminate obstruction by the vehicle body structure. Preferably, the downward shooting angle of the flatbed vehicle license plate recognition camera 1 can cover a range of ±30° above the top of the flatbed vehicle body, that is, the vertical field of view covers a range of ±30° above the top of the flatbed vehicle. In addition, the downward shooting angle can be adjusted as needed.

[0039] The open wagon license plate recognition camera 2 is installed at the front of the front-end crane 3, and its height is higher than that of the flatcar license plate recognition camera 1. Preferably, the height of the open wagon license plate recognition camera 2 is adjusted according to the height of the open wagon license plate to ensure that the open wagon license plate can fall within the shooting range of the open wagon license plate recognition camera 2.

[0040] The distributed installation assembly also includes a support device 5, fixed to the left side of the cab 4 of the front-end crane 3, with the open wagon license plate recognition camera 2 fixed to the support device 5. This position has been carefully calculated to accurately align with the open wagon license plate area, ensuring the efficiency and accuracy of open wagon license plate recognition.

[0041] In addition, the support device 5 can be installed on the right side of the cab 4 as needed, thereby placing the open wagon license plate recognition camera 2 on the right side of the cab 4. Specifically, the positions of the open wagon license plate recognition camera 2 and the support device 5 can be determined according to the direction in which the vehicle enters the work area.

[0042] The support device 5 is height-adjustable and has adjustable horizontal and vertical rotation angles, so that the open car license plate recognition camera 2 can be adapted to the license plate positions of different open car models and eliminate the obstruction and interference of the front-end crane 3 during operation.

[0043] In one embodiment, the support device 5 includes a first support tube, a second support tube, and an adjusting member. The first support tube is slidably connected inside the second support tube, and the adjusting member passes through the second support tube and abuts against or is separated from the outer wall of the first support tube to fix or adjust the height of the support device 5, thereby adjusting the height of the open car number recognition camera 2 to meet the recognition needs of different models of open cars.

[0044] The support device 5 also includes a universal joint bracket, fixed to the top of the first support tube, on which the open-top vehicle license plate recognition camera 2 is fixed. The universal joint bracket is made of aluminum alloy, is lightweight and high-strength, and can rotate 360° horizontally and tilt ±90° vertically, facilitating adjustment of the camera's shooting angle. The universal joint bracket has graduated lines for easy recording and reproduction of the optimal shooting angle. The open-top vehicle license plate recognition camera 2 is fixed to the universal joint bracket via a quick-release plate, facilitating installation and removal.

[0045] In another embodiment, the support device 5 further includes a spherical universal joint bracket, instead of a conventional universal joint bracket, fixed to the top of the first support tube. The spherical universal joint bracket provides greater adjustment freedom, allowing stepless adjustment of the camera angle in three dimensions to adapt to more complex shooting environments and needs. Made of high-strength engineering plastics and metal composite materials, the spherical universal joint bracket is lightweight, high-strength, and has a load-bearing capacity of 5 kg, sufficient to support the open vehicle license plate recognition camera 2 and its accessories.

[0046] In one embodiment, both the first and second support tubes are made of 304 stainless steel, which has good strength and corrosion resistance. The adjusting component is a bolt; by rotating the bolt, the extension length of the first support tube within the second support tube can be adjusted, thereby adjusting the height of the open vehicle license plate recognition camera 2.

[0047] In another embodiment, the first and second support tubes of the support device 5 are made of aluminum alloy. This design reduces the weight of the support device 5, facilitating installation and maintenance, while still providing sufficient strength and stability. The adjusting mechanism employs a quick-locking device, allowing for tool-free adjustment of the height of the support device 5, thus improving ease of operation.

[0048] In one embodiment, the flatcar number recognition camera 1 is a high-definition industrial camera with 5 megapixels, a resolution of 2592×1944, a frame rate of 30fps, and a 12mm focal length lens. It has an autofocus function and can clearly capture flatcar numbers under different lighting conditions. The flatcar number recognition camera 1 can also be fixed to the front of the front-mounted crane 3 by the aforementioned support device 5 for adjustment.

[0049] In one embodiment, the open-top vehicle license plate recognition camera 2 employs an 8-megapixel high-definition industrial camera with a resolution of 3264×2448 and a frame rate of 25fps. Equipped with a 16mm telephoto lens, it offers greater shooting distance and higher clarity, making it suitable for capturing license plate information from the side of open-top vehicles. The installation height of the open-top vehicle license plate recognition camera 2 is approximately 1.5 meters higher than that of the flatbed vehicle license plate recognition camera 1. This height difference allows the open-top vehicle license plate recognition camera 2 to capture the license plate from the side of the open-top vehicle at a suitable angle, avoiding image distortion or blurring caused by improper angles.

[0050] The distributed collaborative control module communicates with both the flatcar license plate recognition camera 1 and the open wagon license plate recognition camera 2, and can activate the corresponding recognition camera to enter the working state according to the vehicle type. For example, when the vehicle type is a flatcar, the distributed collaborative control module controls the flatcar license plate recognition camera 1 to enter the working state, collects the license plate image of the flatcar and sends it to the distributed collaborative control module for recognition; when the vehicle type is an open wagon, the distributed collaborative control module controls the open wagon license plate recognition camera 2 to enter the working state, collects the license plate image of the open wagon and sends it to the distributed collaborative control module for recognition. This allows for the recognition of license plates of different vehicle types while also reducing power consumption.

[0051] The distributed collaborative control module includes a vehicle model recognition submodule, a parameter optimization submodule, a data fusion submodule, an edge computing unit, and a cloud collaboration unit.

[0052] The vehicle type recognition submodule is used to determine the vehicle type based on the license plate image. It employs a deep learning algorithm, built on a convolutional neural network (CNN), using a ResNet-50 architecture. Through training on a large number of truck images, it can accurately distinguish between different types of trucks, such as flatbed trucks and open wagons. The vehicle type recognition submodule achieves an accuracy rate of over 98% and an average recognition time of less than 200 milliseconds. The submodule first preprocesses the input image, including image scaling, grayscale conversion, and histogram equalization. Then, it inputs the processed image into a pre-trained neural network model for inference, outputting a probability distribution of the vehicle type to ultimately determine the vehicle type.

[0053] The parameter optimization submodule is used to adjust the parameters of the flatbed vehicle license plate recognition camera 1 or the open vehicle license plate recognition camera 2 according to the vehicle type. Based on the vehicle type recognition results, the parameter optimization submodule dynamically adjusts the camera's exposure time, gain, contrast, sharpness, and other parameters to obtain the best image quality. For flatbed vehicles, the parameter optimization submodule sets a shorter exposure time (approximately 1 / 1000 second) and a lower gain (approximately 6dB) to avoid overexposure; for open vehicles, it sets a longer exposure time (approximately 1 / 500 second) and a higher gain (approximately 12dB) to ensure clear images even in low-light conditions. The parameter optimization submodule also automatically adjusts the camera parameters according to changes in ambient light, using different parameter configurations for daytime and nighttime to ensure all-weather recognition performance.

[0054] The data fusion submodule is used to process vehicle license plate images captured by at least two flatbed vehicle license plate recognition cameras 1 to generate a complete vehicle license plate image. The data fusion submodule uses image deduplication, stitching, filtering, and feature matching algorithms to fuse images captured by multiple cameras, generating a more complete and clearer vehicle license plate image.

[0055] The edge computing unit is used to identify vehicle license plate numbers in images. It employs OCR (Optical Character Recognition) technology combined with deep learning algorithms to process and recognize vehicle license plate numbers. The edge computing unit first preprocesses the image, including binarization, denoising, and character segmentation. Then, it uses a trained deep learning model (based on a CRNN architecture) to recognize the segmented characters. Finally, post-processing rules (such as vehicle license plate format verification) are applied to improve recognition accuracy. The edge computing unit achieves a vehicle license plate recognition accuracy of over 95%, with an average recognition time of less than 150 milliseconds. The edge computing unit also possesses self-learning capabilities, continuously optimizing model parameters based on feedback from the recognition results to further improve accuracy.

[0056] The cloud-based collaborative unit compares the identified vehicle license plate number with a preset vehicle license plate number database and sends back the verification result. The cloud-based collaborative unit establishes a secure connection with the cloud server via 4G / 5G network or Wi-Fi, uploading the vehicle license plate number information identified by the edge computing unit to the cloud. The cloud server maintains a database containing all truck license plate number information, using a MySQL architecture that supports high-concurrency queries. The cloud-based collaborative unit compares the identified vehicle license plate number with the records in the database to verify the validity of the vehicle license plate number and sends the verification result back to the front-end crane device 3. The cloud-based collaborative unit also records the result and time of each identification, generating an identification log for subsequent data analysis and system optimization. Data transmission uses an SSL encryption protocol to ensure data transmission security.

[0057] In this embodiment, the distributed collaborative control module uses an industrial-grade embedded computer with an Intel i7-9700 processor, 8GB of memory, 1TB SSD storage, and runs a Linux operating system, supporting gigabit Ethernet communication. The distributed collaborative control module establishes a stable communication connection with each camera via Ethernet and uses the TCP / IP protocol for data transmission, ensuring real-time transmission and processing of image data.

[0058] In practical applications, when the reach stacker 3 approaches a truck, the distributed collaborative control module simultaneously activates both the flatcar license plate recognition camera 1 and the open wagon license plate recognition camera 2 to capture truck images. The vehicle type recognition submodule first analyzes the images to determine the truck type. If identified as a flatcar, the image from flatcar license plate recognition camera 1 is used first; if identified as an open wagon, the image from open wagon license plate recognition camera 2 is used first. The parameter optimization submodule dynamically adjusts camera parameters based on vehicle type and ambient lighting conditions to ensure optimal image quality. For flatcars, the data fusion submodule fuses the images captured by the two flatcar license plate recognition cameras 1 to generate a more complete license plate image. The edge computing unit performs license plate recognition on the processed image, and the recognition result is uploaded to the cloud server for verification via the cloud collaboration unit. The verification result is then sent back to the reach stacker 3 to guide subsequent loading and unloading operations.

[0059] In another implementation, when the reach stacker 3 approaches the truck, the distributed collaborative control module first activates the flatcar number recognition camera 1 to capture an image of the truck. The vehicle type recognition submodule first analyzes the image to determine the truck type. If it is identified as a flatcar, the image from the flatcar number recognition camera 1 is used directly; if it is identified as an open wagon, the open wagon number recognition camera 2 is activated to capture an image of the truck, and the image from the open wagon number recognition camera 2 is used.

[0060] This distributed installation component fully leverages the positional advantages of multiple camera sets to achieve multi-angle, all-around imaging and recognition of vehicle license plate numbers for different vehicle types. For flatcars, the two front camera sets work together to acquire vehicle license plate images from different directions, effectively avoiding blind spots caused by cargo obstruction or poor shooting angles. Even if the containers on the flatcar are irregularly stacked, the vehicle license plate number can be accurately extracted through comprehensive judgment of multi-angle image information. For open wagons, the camera installed on the support device 5 can accurately capture images of the open wagon's license plate position, unaffected by the cargo or structure inside the open wagon, ensuring complete acquisition of the vehicle license plate information. This distributed layout greatly expands the field of view for vehicle license plate recognition, adapting to the vehicle license plate recognition needs of different types of freight cars, improving the accuracy and completeness of recognition, and providing strong support for the automated operation of the reach stacker 3.

[0061] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A distributed mounting assembly for truck license plate recognition cameras used in front-end cranes, characterized in that, The distributed installation component includes: At least two flatcar number recognition cameras (1) are installed at different positions in front of the front-end crane (3); A wagon license plate recognition camera (2) is installed at the front of the front-end crane (3), and the height of the wagon license plate recognition camera (2) is higher than the height of the flatcar license plate recognition camera (1); The distributed collaborative control module is communicatively connected to the flatbed car number recognition camera (1) and the open wagon number recognition camera (2). The distributed collaborative control module controls the flatbed car number recognition camera (1) or the open wagon number recognition camera (2) to acquire and process car number images.

2. The distributed installation component according to claim 1, characterized in that, The distributed installation component includes two flatcar number recognition cameras (1), which are symmetrically arranged on both sides of the cab (4) of the front-mounted crane (3). The horizontal field of view coverage of the two flatcar number recognition cameras (1) overlaps by ≥30%.

3. The distributed installation component according to claim 2, characterized in that, The horizontal field of view of the two flatbed car number recognition cameras (1) is ±45°.

4. The distributed installation component according to claim 2, characterized in that, The vertical field of view of the flatbed vehicle number recognition camera (1) covers the top of the flatbed vehicle by ±30°.

5. The distributed installation component according to claim 1, characterized in that, The distributed installation assembly also includes a support device (5) fixed to the left side of the cab (4) of the front-mounted crane (3), and the open vehicle license plate recognition camera (2) is fixed on the support device (5).

6. The distributed installation component according to claim 5, characterized in that, The support device (5) includes a first support tube, a second support tube, and an adjusting member. The first support tube is slidably connected inside the second support tube, and the adjusting member passes through the second support tube and abuts against or is separated from the outer wall of the first support tube to fix or adjust the height of the support device (5).

7. The distributed installation component according to claim 6, characterized in that, The support device (5) also includes a universal joint bracket, which is fixed to the top of the first support tube, and the open vehicle license plate recognition camera (2) is fixed on the universal joint bracket.

8. The distributed installation component according to claim 1, characterized in that, The distributed collaborative control module includes: The vehicle model recognition submodule is used to determine the vehicle type based on the vehicle number image; The parameter optimization submodule is used to adjust the parameters of the flatbed vehicle number recognition camera (1) or the open vehicle number recognition camera (2) according to the vehicle type. The data fusion submodule is used to process the vehicle number images collected by at least two flatbed vehicle number recognition cameras (1) to generate a complete vehicle number image; Edge computing unit, used to identify vehicle numbers in vehicle number images; The cloud-based collaborative unit is used to compare the identified vehicle number with the preset vehicle number database and send back the verification results.

Images