Food Image Processing for a Semi-Automatic Nutrient Intake Monitoring System

Title: Food Image Processing for a Semi-Automatic Nutrient Intake Monitoring System
Authors: Villalobos, Gregorio
Date: 2012
Abstract: According to the World Health Organization (WHO) statistics, obesity has reached epidemic proportions, with over 1.5 billion adults suffering from overweight reported in 2008. As obesity and overweight are major risks to human health, obesity treatment has been the focus of a large number of recent studies. Obesity treatment requires constant monitoring of the patient’s diet. The smart technologies of today’s intelligent environment can be used for the development of appropriate monitoring systems for obesity treatment. In this thesis, we propose a smart system that takes advantage of smartphones to build a platform for monitoring the caloric intake of obesity patients. The patient uses the built-in camera of the smartphone to take a picture of any food that he/she wants to eat. The system then processes the images to detect the type of food and portion size, and uses the information to estimate the number of calories in the food choice. The use of gradient calculation and color rasterization to increase the capacity of the system to recognize the edges of the objects is applied inside the application to signalize the contours of each portion and perform the analysis. As part of our application, we use the thumb of the patient inside the picture as a measurement pattern, to perform the thumb recognition, a skin color algorithm is applied to locate the thumb and measure its size in pixels; the size extracted is then used to translate the food portions into real life size over the entire image. Finally with the portions separated and the computed sizes, a set of nutritional facts are applied in order to generate the amount of calories present in the picture.
CollectionThèses, 2011 - // Theses, 2011 -
Villalobos_Gregorio_2012_Thesis.pdfMaster Thesis3.09 MBAdobe PDFOpen