Image-Based Food Classification and Nutritional Estimation Using MobileNetV2

Ninad Chaudhari; Surajsingh Girase; Jayesh Baviskar; Dhirajkumar Upacharya; Raj Baisane; Bhagyashree Jawale

Image-Based Food Classification and Nutritional Estimation Using MobileNetV2

Author(s)	Ninad Chaudhari, Surajsingh Girase, Jayesh Baviskar, Dhirajkumar Upacharya, Raj Baisane, Prof. Bhagyashree Jawale
Country	India
Abstract	The increasing prevalence of diet-related health disorders such as obesity, diabetes, and cardiovascular diseases has created a growing demand for accurate and efficient dietary monitoring systems. Traditional food logging methods depend heavily on manual user input, making them time-consuming, inconsistent, and susceptible to recall bias. This paper presents a deep learning-based automated dietary assessment system that performs food image classification and nutritional estimation using the MobileNetV2 convolutional neural network architecture integrated within a Django-based web application. The proposed system accepts food images through browser upload or live camera capture, preprocesses them using resizing and normalization techniques, and classifies them into 101 food categories derived from the Food-101 dataset. A two-stage transfer learning strategy is employed, consisting of frozen-backbone feature extraction followed by selective fine-tuning of higher network layers to improve domain-specific performance. After classification, the predicted food label is mapped to a structured nutritional database to retrieve calorie values and macronutrient information including proteins, fats, and carbohydrates. The application further incorporates user authentication, prediction history tracking, and dashboard analytics to support long-term dietary monitoring. Experimental evaluation demonstrates strong performance with training, validation, and testing accuracies of 95.2%, 94.1%, and 93.6% respectively, alongside balanced precision, recall, and F1-score metrics. Comparative analysis against traditional CNN, VGG16, ResNet50, and EfficientNetB0 architectures confirms that MobileNetV2 provides an optimal balance between classification accuracy and computational efficiency for real-time deployment. The proposed system establishes a lightweight and extensible framework for automated nutritional assessment and offers strong potential for future integration with mobile healthcare and personalized nutrition platforms.
Keywords	Deep Learning, Food Image Recognition, MobileNetV2, Nutritional Estimation, Dietary Monitoring, Computer Vision, Transfer Learning, CNN-Based Classification
Field	Engineering
Published In	Volume 17, Issue 2, April-June 2026
Published On	2026-06-02

View / Download PDF File

doi

CrossRef DOI is assigned to each research paper published in our journal.

IJSAT DOI prefix is
10.71097/IJSAT

Downloads

Research Paper Format Copyright Permission Form and Undertaking Form Cover Page Vol 17 Isu 2 Cover Page Vol 17 Isu 1 Cover Page Vol 16 Isu 4

All research papers published on this website are licensed under Creative Commons Attribution-ShareAlike 4.0 International License, and all rights belong to their respective authors/researchers.

CC-BY-SA

About IJSAT Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us	Message on WhatsApp	+91-9687-182-185	editor@ijsat.org

International Journal on Science and Technology

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

Image-Based Food Classification and Nutritional Estimation Using MobileNetV2

Share this