Publication: Car plate detection system by using CNN
| datacite.subject.fos | oecd::Engineering and technology::Electrical engineering, Electronic engineering, Information engineering::Electrical and electronic engineering | |
| dc.contributor.author | Tiong, Sieh Hang | |
| dc.date.accessioned | 2025-02-17T07:08:05Z | |
| dc.date.available | 2025-02-17T07:08:05Z | |
| dc.date.issued | 2023-07 | |
| dc.description.abstract | Car license plate detection system is a critical component of many intelligent transportation systems, including traffic management, toll collection, and parking control. In the past, the traditional method to detect car plate require manpower which is inefficient. With the increasing demand for license plate recognition in daily life, high accuracy models are becoming more important to users. Hence, a car plate detection system with higher accuracy is proposed in this thesis. The proposed system employs a Convolutional Neural Network (CNN) for car license plate detection with the Mini-DSI Panel Display as the extension of Ti60 F225 FPGA to display the labelled test images. The system uses two datasets which dataset A include car license plates with different angles from camera and dataset B with top view of the car plate. Model A and Model B are used to compare the accuracy and loss of each model. Model A is using dataset A and Model B is using dataset A and B for training. 9 layers of CNN is built for training these models with 40 epochs and 5 batch sizes. In short, the system is expected to be used in different advanced applications with high accuracy of car plate detection. | |
| dc.identifier.uri | https://erepo.usm.my/handle/123456789/21074 | |
| dc.language.iso | en | |
| dc.title | Car plate detection system by using CNN | |
| dc.type | Resource Types::text::report::technical report | |
| dspace.entity.type | Publication | |
| oairecerif.author.affiliation | Universiti Sains Malaysia |