Application of deep learning for image-based Chinese market food nutrients estimation

Peihua Ma; Chun Pong Lau; Ning Yu; An Li; Jiping Sheng

doi:10.1016/j.foodchem.2021.130994

Application of deep learning for image-based Chinese market food nutrients estimation

Food Chem. 2022 Mar 30;373(Pt B):130994. doi: 10.1016/j.foodchem.2021.130994. Epub 2021 Aug 31.

Authors

Peihua Ma¹, Chun Pong Lau², Ning Yu³, An Li⁴, Jiping Sheng⁵

Affiliations

¹ School of Agricultural Economics and Rural Development, Renmin University of China, Beijing 100872, China; Department of Nutrition and Food Science, College of Agriculture and Natural Resources, University of Maryland, College Park, MD 20740, United States.
² Department of Computer Science, Whiting School of Engineering, Johns Hopkins University, Baltimore, MD 21218, United States.
³ Department of Computer Science, College of Computer, Mathematical, and Natural Sciences, University of Maryland, College Park, MD 20742, United States.
⁴ Maryland Applied Graduate Department of Robotics Engineering, Maryland Robotics Center, A. James Clark School College of Engineering, University of Maryland, College Park, MD 20742, United States.
⁵ School of Agricultural Economics and Rural Development, Renmin University of China, Beijing 100872, China. Electronic address: shengjiping@126.com.

PMID: 34731793
DOI: 10.1016/j.foodchem.2021.130994

Abstract

With commercialization of deep learning (DL) models, daily precision dietary record based on images from smartphones becomes possible. This study took advantage of DL techniques on visual recognition tasks and proposed a suite of big-data-driven DL models regressing from food images to their nutrient estimation. We established and publicized the first food image database from the Chinese market, named ChinaMartFood-109. It contained 10,921 images with 23 nutrient contents, covering 18 main food groups. Inception V3 was optimized using other state-of-the-art deep convolutional neural networks, achieving up to 78 % and 94 % for top-1 and top-5 accuracy, respectively. Besides, this research compared three nutrient estimation algorithms and achieved the best regression coefficient (R²) by normalization + AM compared with arithmetic mean and harmonic mean, validating applicability in practice as well as theory. These encouraging results provide further evidence supporting artificial intelligence in the field of food analysis.

Keywords: Chinese market food; Convolutional neural network; Deep learning; Food composition; Food image; Food nutrients; Nutrients.

MeSH terms

Artificial Intelligence*
China
Deep Learning*
Neural Networks, Computer
Nutrients