AI Powered Nutrition Estimator
Year: 2025
Type: article
Abstract: Maintaining a healthy diet is challenging in today's fast-paced world. Accurate Identifying what you eat and estimating its calories are vital for effectively managing your dietary habits. In this paper, we put forward a proposal. an AI-driven system for Identifying what you eat and estimating its calories using a distributed approach. Our algorithm, Food Recognition and Adaptive Calorie Estimation (FRACE), leverages convolutional neural networks (CNNs) to identify food items and estimate their caloric content with high accuracy. The foundation of our methodology is image based food identification and calorie prediction using advanced learning. Dataset diversity and real-time processing impact the recognition process as well. This work presents the Food Recognition and Adaptive Calorie Estimation (FRACE) algorithm, which enhances accuracy in food recognition by minimizing misclassification errors and optimizing dataset variability, outperforming state-of-the-art techniques. Our approach efficiently estimates calorie content even with limited food datasets. We also investigate the impact of portion size variations and food diversity on estimation accuracy. Empirical analysis shows that misclassification of complex dishes affects the accuracy of the proposed algorithm. To address this problem, we further optimized the FRACE algorithm by improving feature extraction and dataset annotation. The suggested approach outperforms existing food recognition techniques, as demonstrated through an empirical research study conducted using a CNN-based simulation model.
Authors: Moses Abba -, Naman -, E. Vignesh -, Ms. Anusha -
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
SDG: Zero hunger
Open Access status: hybrid
OpenAlex ID: W4408909123