Leftovers Nutrition Prediction for Augmenting Smart Nutrition Box Prototype Feature Using Image Processing Approach and AFLE Algorithm

Abstract

Some people tend to leave their food when eating caused by their changing lifestyle during the time. Leaving food means wasting its nutritional content acquired in people’s bodies. By understanding the number of nutrient loss, the factors that influence of leftovers food is found, so that, it can prevent the number of food waste. In this paper, we present a method of leftovers nutrition estimation from food images in a single tray box employing an image processing approach. This feature is also embedded in our prototype named as Smart Nutrition Box (SNB). We apply the Automatic Food Leftover Estimation (AFLE) algorithm, which is suitable to predict the weight of food images placed in the tray box. The information on food weight is then utilized for calculating nutrition inside food leftovers. By using Root Mean Square Error (RMSE), the experimental result achieves 1.35 of error. It shows that the proposed method is able to project the leftover nutrition of food.