Unique Advantages of License Plate Recognition System in Parking Lot - Tigerwong Technology

The unique function of the vehicle recognition system is to extract the unique license plate number of the vehicle from the previous passing vehicles for recognition, and the pictures captured by the camera are relatively complex due to the interference of the external environment, so the speed, clarity and accuracy of extracting the license plate number are greatly affected. Because the interference of the environment is generally the factor of road checkpoints and parking lots, in order to improve the recognition ability of resolution, it is necessary to deal with different scenes from the algorithm layer. The HD license plate recognition system specially develops the embedded license plate recognition algorithm for parking lots to improve the effect of large angle and license plate matching rate, This is leading in the same industry. The following are the special application scenarios of license plate recognition system in the parking lot: wide angle recognition recognition under large angle is the biggest difference between the parking lot and other scenes (intersections), and it is also the most difficult point in the license plate recognition system technology in the parking lot environment. Because the license plate recognition system is difficult to capture the rectangular license plate of the vehicle at a large angle in the parking lot, the obtained license plate image is often not a rectangular image, which must be affine deformed and corrected to a rectangular image.

License plate imaging deformation caused by large angle is one of the recognized technical difficulties in license plate recognition. The difficulties lie in (1) difficult license plate location; Due to the angle taken by the license plate recognition camera, the license plate image is often not rectangular, so affine deformation must be made first to correct it into a rectangular image. Therefore, it is not easy to extract license plates from such a complex external environment. 2 Image correction is more difficult; Image correction uses directional high pass filtering (directional differentiation) to determine the direction of motion blur. Then the fuzzy scale is calculated by using the correlation principle. contradiction between recognition speed and large pixels for license plate recognition camera, the farther the license plate is, the smaller the pixel is, and the closer the license plate recognition camera is, the larger the pixel is. Compared with other scenes, the lane of the parking lot is shorter, and the recognition of today's license plate recognition system is real-time.

It is a typical contradiction to support the recognition of license plates with large pixel width under the condition of ensuring fast recognition speed. Generally, if the license plate pixel width is too large, the recognition speed must be slow. If the recognition speed is fast, the license plate pixel width is required to be within a certain range. white list function a major feature of the community parking lot is that the vehicles in and out every day are basically fixed, so the white list function of the license plate recognition system can be applied to this local vehicle. After importing the white list into the license plate recognition system, the vehicle in the white list function list can be stopped and released automatically. Vehicles that do not want to be placed can also be added to the blacklist. multi technology to achieve the best vehicle matching rate at the entrance and exit of the parking lot.

The system structure and internal recognition algorithm of the license plate recognition system are also improved and improved from time to time. It is particularly noteworthy that some technologies extended based on the license plate recognition hardware platform, such as video speed measurement, vehicle type recognition, vehicle partial feature recognition and comparison and equivalence technology, cooperate with license plate recognition to complete the best vehicle matching rate at the entrance and exit of the parking lot, which is also the feature of the parking lot environment. License plate recognition system is characterized by extracting the unique license plate number of the vehicle from the past vehicles for recognition. After the above four features are well used, the license plate recognition system will greatly improve the recognition rate and recognition speed when used in the parking lot. The recognition rate of today's license plate recognition system can not reach 100%, However, it has reached an amazing 99%, and there is still room for improvement.

We are also waiting for the day when the recognition rate is 100%.