Chinese Common Food Classification by Convolution Neural Network

Yusong Wang ； Chang Wei ； Sice Huang ； Yuning Zhou

《International Core Journal of Engineering》 7卷11期 (2021/11) Pp. 503-508

https://doi.org/10.6919/ICJE.202111_7(11).0074

摘要

The article reviews the progress and application of food/dish recognition through Convolution Neural Network (CNN) technique. Also, the article explores the CNN architecture in an experiment to classify four types of common Chinese dishes and reaches an accuracy of 90% after tuning hyperparameters and applying different strategies of optimization.

關鍵字

Progress and application ； Convolution Neural Network (CNN) ； Chinese dishes ； optimization

參考文獻

Mezgec, S. and Koroušić Seljak, B. (2017) ‘NutriNet: A Deep Learning Food and Drink Image Recognition System for Dietary Assessment’, Nutrients, 9(7). doi: 10.3390/nu9070657.

Katz, D. L. et al. (2020) ‘Dietary assessment can be based on pattern recognition rather than recall’, Medical Hypotheses, 140. doi: 10.1016/j.mehy.2020.109644.

Wang, Y. et al. (2019) ‘Mixed dish recognition through multi-label learning’, CEA 2019 - Proceedings of the 11th Workshop on Multimedia for Cooking and Eating Activities, pp. 1–8. doi: 10.1145/3326458.3326929.

Aguilar, E., Bolaños, M. and Radeva, P. (2017) ‘Food Recognition using Fusion of Classifiers based on CNNs’. doi: 10.1007/978-3-319-68548-9_20.

Kiourt, C., Pavlidis, G. and Markantonatou, S. (2020) ‘Deep learning approaches in food recognition’.