Hendrick, Chih-Min Wang, Aripriharta, Ciou-Guo Jhe, Ping-Chong Tsu, Gwo-Jia Jong
Deep learning has a rapid development in image processing application such as face detection, face recognition, object detection and also gesture detection. The other application of deep learning is in the identification of the traffic signs, logo and characters. For Muslim, the halal logo is important to identify before buying some products. The Halal logo is not the same for every country. Both Halal logo Indonesia and Taiwan are different. In this research, the deep learning was applied to classify the halal logo. The classification is based on the caffe framework with GoogleLeNet architecture. As datasets, the halal logo and soft drink logo were created. The purpose of this study is to produce a deep learning pre-trained model of the halal logo. The pre-trained model will be implemented in mobile phone application in halal logo identification. The accuracy of the deep learning pre-trained model is 81.7 %. © 2018 IEEE.
National Kaohsiung University of Science and Technology, Kaohsiung City, 80778, Taiwan; State University of Malang, Indonesia
