Food Recognition and Food Waste Estimation Using Convolutional Neural Network

Abstract

In this study, an evaluation of food waste generation was conducted, using images taken before and after the daily meals of people aged between 20 and 30 years in Serbia, for the period between January 1st and April 31st in 2022. A convolutional neural network (CNN) was employed for the tasks of recognizing food images before the meal and estimating the percentage of food waste according to the photographs taken. Keeping in mind the vast variates and types of food available, the image recognition and validation of food items present a generally very challenging task. Nevertheless, deep learning has recently been shown to be a very potent image recognition procedure, while CNN presents a state-of-the-art method of deep learning. The CNN technique was implemented to the food detection and food waste estimation tasks throughout the parameter optimization procedure. The images of the most frequently encountered food items were collected from the internet to create an image dataset, covering 157 food categories, which was used to evaluate recognition performance. Each category included between 50 and 200 images, while the total number of images in the database reached 23,552. The CNN model presented good prediction capabilities, showing an accuracy of 0.988 and a loss of 0.102, after the network training cycle. The average food waste per meal, in the frame of the analysis in Serbia, was 21.3%, according to the images collected for food waste evaluation.

Multi-Spectral Food Classification and Caloric Estimation Using Predicted Images

“…Since foods have their own unique shapes, textures, and colors, visual cues have been used to classify food types and estimate portion sizes [10][11][12][13][14][15][16][17][18][19][20][21]. From a classical vision-based pattern recognition perspective, automatic food classification is implemented through a series of processes: segmentation, feature selection, and classification of food images.…”

“…From a classical vision-based pattern recognition perspective, automatic food classification is implemented through a series of processes: segmentation, feature selection, and classification of food images. As neural networks have been applied to various image recognition tasks, attempts have been made to use artificial neural networks (ANN) to categorize food types [14,19,20] and estimate the calorie content of foods [14,18,21]. Caloric estimation using visual cues is based on the following assumptions: (1) The calorie counts per size (weight) of a food are uniquely determined by the food item.…”

Multispectral Food Classification and Caloric Estimation Using Convolutional Neural Networks

“…Since types of food are easily distinguished according to their shape, texture, and color, visual cues have been used for the classification of food types and estimation of the food amount [10,[15][16][17][18][19][20][21][22][23]. In a vision-based approach, the classification of food types can be formulated as pattern recognition problems where segmentation, feature selection and classification are sequentially carried out for food images.…”

“…In a vision-based approach, the classification of food types can be formulated as pattern recognition problems where segmentation, feature selection and classification are sequentially carried out for food images. Due to recent advances in machine learning technology, an artificial neural network has been employed to classify food categories and to predict the caloric content of food [18,22,23]. Convolutional neural networks (CNNs) have been used to established 15 food categories with an average classification accuracy of 82.5% and a correlation between the true and estimated calories of 0.81 [18].…”

Why does food loss and waste matter for food security - from the perspective of cause and magnitude

“…There is still potential to reduce avoidable food waste in Hungary. In a four-month study about food waste generation (images were taken before and after meals every day), between 20-30 years old people in Serbia in 2022 estimated that food waste was about 21.3% (Lubura et al, 2022). The wheat-based food loss and waste were estimated at 36% or 4 million tons per year in Morocco (Bartali et al, 2022).…”