As an important part of license plate recognition system, research for the license plate detection has made great progress in recent years, it is still affected by complex environments such as weather, distance, angle and brightness. Therefore, a MFA-UNet model is proposed in this paper, which is based on the UNet model structure and combines the multi-scale convolution feature fusion module and the spatial attention mechanism. In the last two layers of the up-sampling stage, the multi-scale dilated convolution feature fusion module is used to cancel the pooling operation, which ensures that the receptive field can be increased without losing the image resolution, and the image features can be enhanced. The attention of the license plate area is increased by introducing a spatial attention mechanism; the learning and training process have been optimized by using the focal loss function. Based the experiments results, accuracy of the model algorithm mentioned in this paper is 4.5% higher than the original UNet in IoU (Intersection over union), and average detection accuracy of the MFA-UNet model on the Chinese City Parking Dataset (CCPD) dataset is 97.8%, which is a great improvement compared with the target detection algorithm.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.