Convolutional neural networks and support vector machines for hybrid number plate recognition model
by Peter Muthuri Kibaara; Edna C. Too; David Gitonga Mwathi
International Journal of Hybrid Intelligence (IJHI), Vol. 2, No. 2, 2023

Abstract: Recognition accuracy is a determinant performance metric factor that greatly affects the optimal implementation of ANPRs. The recognition accuracy of the existing ANPRs can be improved by adopting a hybrid approach that leverages on the strengths of two machine learning algorithms, i.e., CNN and SVM that have previously been deployed independently in ANPRs. Two models were developed using a deep cascade framework: a pure CNN with a SoftMax classifier and a hybrid CNN with a SVM classifier. UFPR-ALPR dataset was used to train validate and test the models. The hybrid CNN-SVM model had a recognition accuracy of 91.25% against 89.07% from the pure CNN model. The weighted average precision, recall, and F1-score of the hybrid CNN-SVM was 92%, 91% and 91%, respectively, which was better compared to that of pure CNN. The hybrid model was tested for external validity using the SSIG dataset. The hybrid CNN-SVM model had a recognition accuracy of 91% against 89% from the pure CNN model. The weighted average precision, recall, and F1-score of the hybrid CNN-SVM was 91%, 91% and 91% respectively which was better compared to that of pure CNN.

Online publication date: Mon, 06-Mar-2023

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Hybrid Intelligence (IJHI):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com