Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Image-Based Dietary Assessment: A Healthy Eating Plate Estimation System

Assylzhan Izbassar, Pakizar Shamoi

TL;DR

This work tackles automated dietary assessment from food images by grounding the analysis in the Harvard Healthy Eating Plate guidelines. It proposes a four-step pipeline—segmentation, classification, nutritional assessment, and recommendations—implemented via color-based segmentation, region merging, and SVM classification, followed by a nutritional evaluation that yields plate balance $B$ and healthy share $H$ using $B = \min(f+v, 50) + \min(hp, 25) + \min(wg, 25)$ and $H = (f+v+hp+wg)/T$. Data collection combines a Roboflow plate/bowl dataset with additional healthy-plate images, and the model categories foods into fruits, vegetables, proteins, and whole grains. A prototype app demonstrates practical use for dietary feedback and restaurant recommendations, with initial results showing promising segmentation and classification performance and real-world feasibility. Limitations include segmentation accuracy for overlapping items and handling of sauces, guiding future work toward more robust object separation and diverse cooking methods.

Abstract

The nutritional quality of diets has significantly deteriorated over the past two to three decades, a decline often underestimated by the people. This deterioration, coupled with a hectic lifestyle, has contributed to escalating health concerns. Recognizing this issue, researchers at Harvard have advocated for a balanced nutritional plate model to promote health. Inspired by this research, our paper introduces an innovative Image-Based Dietary Assessment system aimed at evaluating the healthiness of meals through image analysis. Our system employs advanced image segmentation and classification techniques to analyze food items on a plate, assess their proportions, and calculate meal adherence to Harvard's healthy eating recommendations. This approach leverages machine learning and nutritional science to empower individuals with actionable insights for healthier eating choices. Our four-step framework involves segmenting the image, classifying the items, conducting a nutritional assessment based on the Harvard Healthy Eating Plate research, and offering tailored recommendations. The prototype system has shown promising results in promoting healthier eating habits by providing an accessible, evidence-based tool for dietary assessment.

Image-Based Dietary Assessment: A Healthy Eating Plate Estimation System

TL;DR

This work tackles automated dietary assessment from food images by grounding the analysis in the Harvard Healthy Eating Plate guidelines. It proposes a four-step pipeline—segmentation, classification, nutritional assessment, and recommendations—implemented via color-based segmentation, region merging, and SVM classification, followed by a nutritional evaluation that yields plate balance and healthy share using and . Data collection combines a Roboflow plate/bowl dataset with additional healthy-plate images, and the model categories foods into fruits, vegetables, proteins, and whole grains. A prototype app demonstrates practical use for dietary feedback and restaurant recommendations, with initial results showing promising segmentation and classification performance and real-world feasibility. Limitations include segmentation accuracy for overlapping items and handling of sauces, guiding future work toward more robust object separation and diverse cooking methods.

Abstract

The nutritional quality of diets has significantly deteriorated over the past two to three decades, a decline often underestimated by the people. This deterioration, coupled with a hectic lifestyle, has contributed to escalating health concerns. Recognizing this issue, researchers at Harvard have advocated for a balanced nutritional plate model to promote health. Inspired by this research, our paper introduces an innovative Image-Based Dietary Assessment system aimed at evaluating the healthiness of meals through image analysis. Our system employs advanced image segmentation and classification techniques to analyze food items on a plate, assess their proportions, and calculate meal adherence to Harvard's healthy eating recommendations. This approach leverages machine learning and nutritional science to empower individuals with actionable insights for healthier eating choices. Our four-step framework involves segmenting the image, classifying the items, conducting a nutritional assessment based on the Harvard Healthy Eating Plate research, and offering tailored recommendations. The prototype system has shown promising results in promoting healthier eating habits by providing an accessible, evidence-based tool for dietary assessment.
Paper Structure (17 sections, 4 equations, 10 figures, 1 table, 1 algorithm)

This paper contains 17 sections, 4 equations, 10 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Work
  3. Image processing for food detection and dietary assessment
  4. Nutritional recommendations: Overview
  5. Methods
  6. Experimental settings
  7. Data Collection
  8. Dataset of bowls
  9. Train data of a classification algorithm
  10. Image Processing Techniques
  11. Image Segmentation
  12. Image Classification
  13. Nutritional Assessment Algorithm
  14. Results
  15. Accuracy Evaluation
  16. ...and 2 more sections

Figures (10)

  • Figure 1: Healthy plate recommendation. For more information, please see The Nutrition Source, Department of Nutrition, Harvard T.H. Chan School of Public Health, www.thenutritionsource.org, and Harvard Health Publications, www.health.harvard.edu.
  • Figure 2: Clustering of the plate
  • Figure 3: Categories of healthy objects
  • Figure 4: Types of image segmentation guide
  • Figure 5: Color histogram of the initial plate image
  • ...and 5 more figures