chrispeng [ad_1]
【introduction】
In recent years, with the continuous development of my country’s economy, the port container business has grown rapidly, and the demand for increasing throughput has become stronger and stronger. At present, the technical methods of domestic terminals are generally backward. Due to manual intervention, many gates have caused long queues of vehicles in front of the gates, and errors caused by manual recording have increased the burden of repeated processing by the system. Many of the above problems have led to a decline in the service level of the terminal, which not only affects the living environment of the terminal itself, but also causes the import and export enterprises in the area covered by the terminal to suffer greater economic losses.
As the ID symbol that uniquely identifies the container, the container number must be recorded in all aspects of the container transportation process. The system uses neural network algorithm-based image recognition technology and advanced CCD image acquisition technology to automatically capture various container numbers entering and leaving the port, and quickly perform automatic identification, eliminating errors caused by manual transcription and reducing entry into the port. Time improves work efficiency.
【System Principle】
When a container truck enters the aisle, the photoelectric sensors fixed on both sides of the aisle will detect the position of the container. The control system determines the length and type of the container (a short container or a short container or One long box or two short boxes), and at certain moments, the four color cameras (all equipped with powerful flashes for fill light with automatic function) fixedly installed at specific locations around the aisle will issue photographing instructions ( There are different shooting modes corresponding to different loading types) to shoot the parts with box numbers on the four boxes of the container; the camera produces digital optical images of the parts with box numbers (short box shooting mode produces 4 images, long box or Two short box shooting modes produce 6 images). The image data is collected in real time and processed by the computer, and the container number and box type are identified by the optical image recognition module (OCR) for back-end use.
The system consists of the following parts:
1. Front-end photoelectric trigger device;
2. Front-end image acquisition equipment;
3. Back-end logic control equipment;
4. Back-end recognition algorithm software;
5. Back-end statistical software and communication software;
【Overall structure of the system】
BS structure can connect multiple lanes at the gate of the terminal at the same time. The 4 cameras installed in each lane are connected to the video capture card through the national standard video cable, and the optoelectronic equipment is managed by the logic control device to ensure the correct image capture logic and prevent the spoiler or other objects from interfering with the front of the car. Perform unified data integration processing through the computer room. The overall structure of the recognition system is shown in Figure 1.
【System functions】
By connecting the identification system with the on-site computer, it can provide the back-end system with the container number information of the vehicle, count the number of containers entering and exiting the crossing and the owner of the container, and generate relevant reports.
This system has the following technical advantages:
1. The system can recognize the container number of GB/T 1836-1997 standard, and can handle any printing method of container number, including one row, two rows, three rows, four rows, one column, two columns, etc.;
2. It can identify three situations: one small box, two small boxes, and one long box;
3. Under various weather conditions, the overall system recognition rate after correct installation is >=95%;
4. FCL identification time <=6 seconds. (Based on two short boxes and 6 valid pictures); the recognition time of each picture is <=1 second. (P4 3.0/512M configuration);
5. The identification system is reliable and can work effectively without being affected by the truck loading container type, double-box gap, and container placement position. When some sides of the container are stained, the characters are peeled off, and the handwriting is illegible, the system can also ensure the correct identification of the container number through image complementation.
6. The gantry bracket is adopted, which occupies a small space and is more suitable for the transformation of the old gate; it can be reused with the box damage detection system to reduce the system cost;
7. Real-time recognition rate statistics and data query functions, recognition data and images can be stored for a long time, and support a variety of data communication functions to facilitate third-party integration;
【Application Effect】
Through the implementation of the container number identification system at several terminal gates, the containers entering and leaving the terminal are effectively managed, reducing the amount of errors caused by manual recording, the speed of vehicles entering and leaving the port has increased significantly, the system is operating stably, and the identification rate meets the requirements of customers need.
If you have any questions or for more detailed plans, please contact
Address: 606, Chuangye Building, Academician Road, High-tech Zone, Ningbo
Contact: Liu Feng
Phone: 0574-87910130
Mobile: 13967835543
Fax: 0574-87908484
MSN:[email protected]
QQ: 38291354
URL:http://www.onepass-tech.com
[ad_2]
