Related Experiment Video
Updated: Apr 5, 2026
07:47
Design and Evaluation of Smart Glasses for Food Intake and Physical Activity Classification
Published on: February 14, 2018
12.0K
Saliency-aware food image segmentation for personal dietary assessment using a wearable computer
Hsin-Chen Chen1, Wenyan Jia2, Xin Sun3
1Department of Radiation Oncology, Washington University in Saint Louis, Saint Louis, MO, USA ; Department of Neurological Surgery, University of Pittsburgh, Pittsburgh, PA, USA.
Measurement Science & Technology
|August 11, 2015
Summary
This study introduces a novel method for automatic food segmentation in images, crucial for obesity research and dietary assessment. The approach significantly improves the accuracy of identifying food items from pictures.
Area of Science:
- Computer Vision
- Medical Imaging
- Obesity Research
Background:
- Image-based dietary assessment is vital for obesity research.
- Current manual methods are labor-intensive.
- Automatic dietary assessment via image processing offers significant potential.
Purpose of the Study:
- To develop an automatic food segmentation method for dietary assessment.
- To address challenges in food segmentation like varied food types, shapes, colors, container patterns, and occlusions.
- To improve the accuracy of food object identification in images.
Main Methods:
- A saliency-aware active contour model (ACM) was developed for automatic food segmentation.
- Integrated saliency estimation used food location priors and visual attention features.
- A geometric contour primitive was fitted to salient maps using multi-resolution optimization.
Main Results:
- The proposed method achieved significantly higher accuracy in food segmentation compared to conventional techniques.
- Experiments were conducted on 60 food images.
- The method successfully extracts food regions after contour fitting.
Conclusions:
- The novel saliency-aware ACM method enhances automatic food segmentation accuracy.
- This technique shows promise for advancing automated dietary assessment systems.
- Accurate food segmentation is a critical step towards objective dietary intake monitoring.
