摘要
针对普通场景车牌识别存在的问题,本文提出使用"OpenCV图像处理+自适应卷积神经网络"的方法来处理汽车图像。该方法通过对汽车图像进行定位检测将车牌数据提取出来,再使用卷积神经网络进行识别,整体的准确率达到97%。实验结果表明,该方法具有较高的准确率。
Aiming at the problems of license plate recognition in common scenes, this article proposes the use of "OpenCV image processing + adaptive convolutional neural network" method to process car images. This method extracts the license plate data by positioning and detecting the car image, and then uses the convolutional neural network for recognition. The overall accuracy rate reaches 97%. Experimental results show that this method has a higher accuracy rate.
作者
袁程
曹爱青
YUAN Cheng;CAO Aiqing(Nanning Normal University,Nanning Guangxi 530023,China)
出处
《信息与电脑》
2021年第2期156-158,共3页
Information & Computer
关键词
卷积神经网络
车牌检测
车牌识别
图像处理
convolutional neural network
license plate detection
license plate recognition
image processing