HALALCheck: A Multi-Faceted Approach for Intelligent Halal Packaged Food Recognition and Analysis

• Published in: IEEE access 2024-01, Vol.12, p.1-1
• Main Authors: Tarannum, Sabrina, Jalal, Md Shah, Huda, Mohammad Nurul
• Format: Article
• Language: English
• Subjects: Algorithms; Certification; Classification algorithms; Consumers; Cultural aspects; Data models; Deep learning; Electronic noses; Feature extraction; Food; Food classification; Food packaging; Food products; Food security; Halal food; Image recognition; Ingredients; Intelligent systems; Islamic law; Islamic Shariah; Machine learning; Mathematical models; Muslim consumer; Muslims; Neural networks; Optical character recognition; Rule-based model; Social factors; Surveys; YOLO
• ISSN: 2169-3536; 2169-3536
• DOI: 10.1109/ACCESS.2024.3367983

Halal cuisine in Islamic religion is crucial because it signifies food that is legal according to Islamic Shariah to guarantee that they adhere to their religious beliefs. This study addresses the critical role of Halal cuisine in Islamic dietary practices, emphasizing its significance in adhering to Islamic Shariah. The research aims to tackle the challenge of identifying and validating Halal-certified foods where Halal certification is absent or unclear, making it challenging for Muslim consumers, especially tourists, to confidently identify permissible foods. Traditional approaches are often not scalable or adaptable to the dynamic nature of packaged food products, leading to inaccuracies in Halal classification. To address this issue, we propose a more robust, automated, and technologically advanced solution, a novel method utilizing deep learning and machine learning approaches for Halal food recognition. The system employs the YOLOv5 detection algorithm to identify ingredients in packaged food products, and subsequent feature extraction through an optical character recognition system. After preprocessing, a combination of trained machine learning models, neural networks, and rule-based systems classifies the ingredients, accurately determining the Halal or Haram status of food products. The research yields promising results, with the proposed Halal packaged food recognition system achieving an accuracy rate of 98%, validated through comparison with the opinions of three Islamic scholars. This innovative approach holds significant potential for aiding Muslim consumers in efficiently recognizing Halal-certified items, especially when navigating unfamiliar environments. However, the proposed intelligent system assists Muslim tourists in identifying Halal food across the world.