by:Shenzhen TGW Technology Co.,Ltd
How to improve the recognition accuracy of license plate recognition system
From the current market, license plate recognition manufacturers see that, as the license plate recognition system can enable vehicles to enter and exit the parking lot, they can complete the charges without the card, which speeds up vehicle traffic and reduce card management Work, while reducing the cost of management staff, so it is widely used in major parking lots. However, the license plate recognition system sometimes recogniize license plates in a wrong way. How to solve this problem and improve the recognition accuracy ?
From the perspective of external factors, the factors affecting the license plate recognition accuracy mainly include viewpoint transformation, light, occlusion, proportion, deformation, and complex background.
The license plate recognition system used in the parking lot is mainly composed of barriers, vehicle detectors, high-definition license plate capture cameras, electronic displays, and control centers. Capture the license plate of the import and export vehicle through a high-definition license plate capture camera, extract the comparison of the pictures before and after, and determine whether to release the vehicle. It can be seen that the license plate recognition parking lot system has high requirements for the entire system, and at the same time, the external environment will have a great impact on the recognition rate.
Three main reasons affect the recognition rate of parking lots with a license plate recognition parking system.
First, it is necessary to ensure that the ambient light in the parking lot is bright enough. For dark or nighttime, some supplementary lighting equipment should be appropriately added;
Second, ensure that the license plates of vehicles entering and leaving the parking lot are intact so that the license plates of each vehicle entering and leaving the stadium can be captured;
Third, ensure the license plate recognition position, the size of the license plate pixels, not less than the size of the best license plate pixels;
Therefore, optimizing the recognition environment, reducing the complexity of the environment, and slowing down the vehicle speed through auxiliary means can effectively improve the license plate recognition rate.
From the internal cause, the license plate recognition system is based on the license plate recognition algorithm, and the license plate recognition algorithm is composed of space complexity and time complexity and is divided into five different characteristics:
1, the entry is the specific license plate image;
2, the output item is the recognition result of the license plate;
3. Finiteness means that the evolution of the algorithm is limited, and infinite calculation is impossible;
4. Exactness means that each calculation step of the algorithm has a precise definition;
5. Feasibility is the effect of license plate recognition.
By optimizing the above five characteristics, fundamentally optimizing the license plate recognition algorithm is one of the most effective methods to improve the license plate recognition rate.