Deteksi Bahan Makanan untuk Rekomendasi Resep Masakan pada Program Diet Menggunakan Algoritma CNN

Abstract

The increasing need for efficient dietary planning has led to the development of automated systems for identifying food ingredients and generating suitable diet recommendations. This study focuses on implementing a Convolutional Neural Network (CNN) using the VGG16 architecture to classify food ingredients and determine appropriate diet recipes. The problem addressed is the difficulty of manually identifying various food ingredients, which can be time-consuming and error-prone, especially in large-scale dietary planning. The proposed solution integrates deep learning technology with a user-friendly application that automates the classification process and generates diet suggestions. The method involves utilizing the VGG16 model pre-trained on the ImageNet dataset. The dataset underwent preprocessing techniques, including Gaussian Blur for noise reduction, normalization, and data augmentation, to improve model generalization. The model was trained over 50 epochs, achieving a training accuracy of 96.28% and a validation accuracy of 95%. This study contributes to the development of intelligent dietary systems, providing significant benefits in enhancing user convenience, accuracy in food classification, and promoting healthier lifestyles.