A new energy license plate detection method based on video surveillance

Abstract

The invention discloses a new energy license plate detection method based on video monitoring, which includes: collecting a large number of new energy vehicle pictures containing license plate areas as a data set for training models, manually punctuating each picture in the data set; constructing a new energy license plate Detect and train the neural network, input the data set into the neural network for training to obtain a new energy license plate detection model; then obtain new energy vehicle images from the surveillance video, establish a new energy vehicle image set containing license plate position coordinate information, and Name the image of the target vehicle according to the time sequence of the appearance of the vehicle; extract the image features to be detected through the neural network, send the features to the detection model for judgment, and output the confidence of the license plate image candidate area in the vehicle image according to the model; finally, according to the confidence-based degree of the license plate candidate area, and calculate the best position coordinates of the license plate.

Description

technical field [0001] The invention belongs to the technical field of digital security license plate detection, in particular to a new energy license plate detection method based on video monitoring. Background technique [0002] The increasing number of motor vehicles has brought severe challenges to the supervision and management of the transportation department. The intelligent transportation system is particularly important. The license plate in the monitoring scene is used as the identity of the vehicle, making the automatic license plate detection and recognition (ALPR) technology an intelligent transportation system. The key technology in the process is to recognize the characters on the license plate through various image processing and computer vision technologies, so as to identify each vehicle. License plate recognition usually includes license plate detection, character segmentation and recognition; among them, license plate detection is the basic component of l...

AI Technical Summary

Problems solved by technology

[0007] For the vehicle images in the monitoring scene, the two-step detection method of coarse inspection and fine inspection is usually used to locate the license plate, which has problems such as high computational cost, long time consumption, and background interference. At the same time, the emergence of new energy vehicle license plates makes these problems more obvious. There is a big difference between the new energy vehicle license plate and the ordinary vehicle license plate. The plate number of the new energy vehicle is increased by one digit. It caused some inconvenience in traffic management and vehicle management. The license plate number became longer, and the previous license plate detection technology could not fully detect it. The traditional detection method for blue plate vehicles has a detection effect and robustness on new energy license plates. There is still room for improvement. In summary, there is an urgent need for a detection method that can effectively detect new energy license plates

Images