Padang Cuisine Classification using Deep Convolutional Neural Networks (CNN) and Transfer Learning

Abstract

Padang cuisine, an integral part of Indonesia's culinary richness, offers diverse flavors and distinctive aromas. Traditional Indonesian cuisines, such as Padang, are characterized by the richness of flavors, aromas, and textures. However, the traditional cuisines of Indonesia are not fully understood. This research aims to develop an automatic classification system capable of distinguishing various Padang dishes using Deep Convolutional Neural Networks (CNN) and Transfer Learning. This innovative approach uses a deep convolutional neural network architecture to extract deep features from food images, allowing for accurate identification. Transfer Learning leverages existing knowledge from pre-trained CNN models on large datasets, adapting it to the specific context of Padang culinary classification. The results of this research indicate that this approach provides excellent results in automatically identifying and classifying Padang recipes. In addition, this model can be easily adapted to accommodate new developments in Padang cooking variations or even applied to other types of cuisines with appropriate adjustments. Therefore, this research opens opportunities for utilizing AI technology to support the culinary industry, especially in understanding and recognizing the culinary diversity of this traditional cuisine.