Related Experiment Video
Updated: May 23, 2025
13:19
Deep Neural Networks for Image-Based Dietary Assessment
Published on: March 13, 2021
8.9K
Image-based food groups and portion prediction by using deep learning
Hidir Selcuk Nogay1, Nalan Hakime Nogay2, Hojjat Adeli3
1Faculty of Engineering, Department of Electrical and Electronics Engineering, Bursa Uludag University, Bursa, Turkey.
Journal of Food Science
|March 7, 2025
Summary
A new deep learning system automatically classifies Turkish foods and estimates portion sizes. This technology aids in managing diet, preventing malnutrition, and addressing chronic diseases like obesity and hypertension.
Area of Science:
- Nutrition Science
- Computer Science
- Artificial Intelligence
Background:
- Chronic diseases like obesity and hypertension are linked to malnutrition and poor dietary habits.
- Accurate food consumption measurement is crucial for personalized nutrition and preventing malnutrition, especially given diverse food cultures.
- Automated systems can help monitor dietary intake and ensure nutritional needs are met.
Purpose of the Study:
- To develop and implement a deep learning system for automatic food grouping and classification.
- To estimate portion sizes of dishes, specifically focusing on Turkish cuisine.
- To improve nutritional assessment and management through image recognition technology.
Main Methods:
- Utilized deep learning, specifically Convolutional Neural Networks (CNNs), for image-based food recognition.
- Developed a system to group and classify food items.
- Implemented data augmentation techniques to enhance model performance.
Main Results:
- Achieved up to 80% accuracy in classifying food groups.
- Reached 80.47% accuracy in estimating portion sizes.
- Demonstrated the effectiveness of CNNs for analyzing Turkish cuisine dishes.
Conclusions:
- The developed deep learning system accurately classifies food groups and estimates portion sizes.
- This technology offers a promising tool for personalized nutrition and dietary management.
- Automated food analysis can aid in preventing malnutrition and diet-related chronic diseases.
