Fine-Grained Food Classification Methods on the UEC FOOD-100 Database

Abstract

The development of an automatic food recognition system has severalinteresting applications ranging from waste food management, to advertisement, to calorie estimation, and daily diet monitoring. Despite the importance of this subject, the number of related studies is still limited. Moreover, the comparison in the literature was currently done over the best-shot performance without considering the most common method of averaging over several trials. This article surveys the most common deep learning methods used for food classification, it presents the publicly available databases of food, it releases benchmark results for the food classification experiment averaged over five-trials, and it beats the current best-shot performance experiment reaching the state-of-the-art accuracy of 90.02% on the UEC Food-100 database. The best results have been achieved by the ensemble method averaging the predictions of ResNeXt and DenseNet models. All the experiments are run on the UEC Food-100 database because it is one of the most used databases, and it is challenging due to the presence of multifood images, which need to be cropped before processing. This article aims to contribute to automatic food recognition by presenting the most common algorithms used for food classification, introducing the main databases of food items currently available, and reaching the state-of-the-art performance in the best-shot classification experiment of the UEC Food-100 database. That is, this article improves the current best-shot performance by 0.44 percentage points, and fixes it to 90.02%. Furthermore, with the best of our knowledge, this is the first article to introduce to the research community comparison of performances of the classification experiment on the UEC Food-100 database averaged over five-trails. As expected, performance averaged is slightly lower thanthe best-shot one.