FOOD RECOGNITION AND CALORIE ESTIMATION USING CNN

Abstract

Obesity and overweight have traditionally been linked to intake of high calorie food. Obesity may be a medical condition during which excess body fat has accumulated to the extent that it‘s going to have an adverse effect on health. Changes to diet and exercising are the major treatments. Diet quality are often improved by reducing the consumption of energy dense foods like those high in fat or sugars and by increasing the intake of dietary fiber. There are also many other reasons that lead to monitoring of food calorie estimation. Proposed system will be trained with many images in each category, which then classifies the food image and then estimates nutrition / calorie. Proposed system uses CNN as a classifier for food recognition and based on the food weight in grams the calorie of the food items are determined. This deep convolutional neural network will configure to detect and recognize local food images. Data about most consumed local food items were collected from internet. The main aim of this project is to improve the accuracy of the pre-training model using CNN.