Results

Our system performance is evaluated based upon the
accuracy of calculation of nutrition and calorific value. Because of the use of
pixel-based color image segmentation we use both support vector machine and
fuzzy C-means as the algorithm. Since a complete food images are stored as a
library file, the use of internet is avoided, which holds as greater advantage.
Depending upon the type of diabetes, the calorific and nutritional value for
the patient is pre-fed and hence the output that the food is recommended or not
is based upon those values. Since deep learning concept is used unsupervised
learning is done.In-order to minimize the complexity, we convert the RGB values
to gray scale, thus increasing the processing power. The use of SIFT featuring at
2 different scales results in  best
performance. Since it is a mobile app it becomes handy for the user and can use
it anywhere and at anytime. It results in good user interface with outputs
aligned perfectly.

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!


order now

Discussion

The results are clear with  modest
number of food, the proposed technique is  effective. The app is not yet released in the
public since there are two reasons. First, by improving more computation at
more scales. As for now we have limited the scales due to computational
reasons. Second, for the app to be more realistic huge amount of different
types of food has to be store. Scaling up to thousands of foods requires the development
of new approaches to food ontology by 
incremental approach in upgrading the ontology anytime and anywhere when
a new food is added.

Conclusions

We have developed a mobile app for recognition of nutritional and
calorific content in the food consumed by a diabetes patient.The application recognizes
the  food item present only on the plate
and estimates their calorific and nutrition content automatically without any
user intervention. To identify the food items, the user simply needs to take a
snapshot of the food plate. The system automatically

detects the salient regions base on the algorithm. Hierarchical segmentation
is used to segment the images into lone region. By using SVM classifiers the
application extracts the features at different scales and locations,
classifying these region. The application crops the part of food alone from the
background and starts to estimate the calorific and nutrition content. We have
implemented this experiment as an Android smartphone application.  our experiments, we have achieved.Over 88% of
accuracy in determining 16 food items. For the future work, we have planned to
make this application for personalizing the food base on  user habits, health issues and other
meta-data.

x

Hi!
I'm Erica!

Would you like to get a custom essay? How about receiving a customized one?

Check it out