Publication: Vision-based halal logo verification system
| datacite.subject.fos | oecd::Engineering and technology::Electrical engineering, Electronic engineering, Information engineering::Electrical and electronic engineering | |
| dc.contributor.author | Chew, Wen Joo | |
| dc.date.accessioned | 2025-05-13T04:45:17Z | |
| dc.date.available | 2025-05-13T04:45:17Z | |
| dc.date.issued | 2023-07 | |
| dc.description.abstract | The Muslim community places great importance on Halal certification regarding food intake. Given the rising demand for Halal products, it is crucial to have a reliable mechanism to confirm Halal emblems on food packaging. Thus, this project aims to develop a vison-based halal logo verification system. The research methodology employs VGG-16 convolutional neural network (CNN) architecture. The dataset used in this study consists of 51 unique Halal logos collected from 33 countries. Developing a vision-based Halal logo verification system has two essential stages: developing an image comparison algorithm and creating a Graphical User Interface (GUI) to apply the trained model. In developing the image comparison stage, deep learning methods and feature extraction were chosen to be tested. Based on the results, deep learning is chosen as it performs better and more consistently than the feature extraction method. VGG-16 model used in developing image comparison algorithms indicates an accuracy of 100.00% in the tested sample. In conclusion, the vision-based Halal logo verification system is successfully developed using a deep learning approach. The system can be used as a tool for verifying whether the scanned Halal logo is genuine or not. | |
| dc.identifier.uri | https://erepo.usm.my/handle/123456789/21598 | |
| dc.language.iso | en | |
| dc.title | Vision-based halal logo verification system | |
| dc.type | Resource Types::text::report::technical report | |
| dspace.entity.type | Publication | |
| oairecerif.author.affiliation | Universiti Sains Malaysia |