Automatic Fruits Calories Estimation through Convolutional Neural Networks
Ghazanfar Latif, Batool Alsalem, Wejdan Mubarky, Nazeeruddin Mohammad, Jaafar Alghazo
Abstract
With the increased interest in healthy eating and diet control, the importance of keeping track of daily calorie intake is an important topic nowadays. In this paper, we propose a novel technique to automatically calculate calories based on the picture of various foods on a plate. As a first step, this paper concentrate on food plates with fruit and vegetable content only. A new dataset was developed internally for this research consisting of a total of 41,509 images. In this paper, we propose a Deep Convolutional Neural Network (DCNN) for the automatic recognition process. A custom design application was developed for capturing the image, recognition, and automatic calculation of calories. The results showed that over 92% recognition rates were achieved on most fruits and vegetables.