Digital photography based food intake prediction using artificial neural network
Introduction Many wearable devices monitoring have been proposed to complement self-reporting of users’ caloric intake and eating behaviours. These devices comprise varying sensing modalities, such as acoustic, visual, inertial, EEG, EMG, capacitive and piezoelectric sensors. In this research, food...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English English |
| Published: |
Malaysian Medical Association
2017
|
| Subjects: | |
| Online Access: | http://irep.iium.edu.my/59505/ http://irep.iium.edu.my/59505/ http://irep.iium.edu.my/59505/1/MJM_v72-Supp-1-2017.pdf http://irep.iium.edu.my/59505/2/Teddy_FoodIntakeNN.pptx |
| Summary: | Introduction
Many wearable devices monitoring have been proposed to complement self-reporting of users’ caloric intake and eating behaviours. These devices comprise varying sensing modalities, such as acoustic, visual, inertial, EEG, EMG, capacitive and piezoelectric sensors. In this research, food intake will be predicted from the input of digital photography using ANN.
Methods
In this study, image of selected food or leftovers are captured using digital camera or smartphone. These two images are later compared with images of averaged portions of food. Area based comparison or trained artificial neural network could then predicted the calorie and nutrient intake.
Results
Preliminary results show the effectiveness of measuring food intake using digital photography. It is found that more images are required to train the artificial neural network for various image capturing position to improve the prediction accuracy.
Discussion
The proposed method is rather simple and easy and provides quick feedback on food intake and dietary recommendations to achieve weight loss goal. It is believe that such findings would allow general public to better achieve and maintain their healthy lifestyle. |
|---|