Academic Journal of Computing & Information Science, 2024, 7(9); doi: 10.25236/AJCIS.2024.070902.
Chen Xieyu1,2, Tang Na1,2
1School of Business, Geely University of China, Chengdu, China
2Key Laboratory of Sichuan Cuisine Artificial Intelligence, Chengdu, China
With the development of deep learning technology and increasing awareness of dietary structures, food recognition has become a popular research field. This article presents an integrated system that combines Sichuan cuisine dish recognition with nutritional assessment, based on the lightweight YOLO v5s model. The system consists of three main modules: data preprocessing module, model training and evaluation module, and nutrition information display and reminder module. In this study, a total of 1,602 images of Sichuan dishes were utilized as the original dataset, which were annotated to generate six categories: Yu Xiang Rou Si, Shui Zhu Rou Pian, Mapo Tofu, Qing Jiao Chao Rou, Gong Bao Ji Ding, and Kuo Shui Huang Gua. The results of the model's performance indicate that the mean Average Precision ([email protected]) value of the model is 0.97, with the prediction, recall, and F1-score for each class exceeding 0.9. The loss function exhibited stable performance, indicating that the model operates effectively. The article concludes by showcasing some recognition results of the dishes along with their nutritional composition information.
YOLO v5s, Sichuan Cuisine Dish, Nutritional Assessment, Image Recognition
Chen Xieyu, Tang Na. Design of a Sichuan Cuisine Dish Image Recognition and Nutritional Information Assessment System Based on YOLO v5s. Academic Journal of Computing & Information Science (2024), Vol. 7, Issue 9: 11-17. https://doi.org/10.25236/AJCIS.2024.070902.
[1] Zhang Zicheng, Xue Yunlian, Xu Jun. Prevalence and influencing factors of suboptimal health among urban middle-aged and elderly residents in China[J]. Chinese Journal of Public Health, 2023, 39(1): 27-31.
[2] Redmon J, Divvala S, Girshick R, et al. You Only Look Once: Unified, Real-Time Object Detection[J]. arXiv e-prints, 2015: arXiv:1506.02640.
[3] Jiang P, Ergu D, Liu F, et al. A Review of Yolo Algorithm Developments[J]. The 8th International Conference on Information Technology and Quantitative Management (ITQM 2020 & 2021): Developing Global Digital Economy after COVID-19, 2022, 199: 1066-1073.
[4] Yin H, Chen M, Fan W, et al. Efficient Smoke Detection Based on YOLO v5s[J]. Mathematics, 2022, 10(19).
[5] Ryu J, Won D, Lee Y. A Study of Split Learning Model[C]//2022 16th International Conference on Ubiquitous Information Management and Communication (IMCOM). 2022: 1-4.
[6] Xu M, Yoon S, Fuentes A, et al. A Comprehensive Survey of Image Augmentation Techniques for Deep Learning[J]. Pattern Recognition, 2023, 137: 109347.
[7] Zhao T, Wei X, Yang X. Improved YOLO v5 for Railway PCCS Tiny Defect Detection[C]//2022 14th International Conference on Advanced Computational Intelligence (ICACI). 2022: 85-90.
[8] Du S, Zhang B, Zhang P, et al. An Improved Bounding Box Regression Loss Function Based on CIOU Loss for Multi-scale Object Detection[C]//2021 IEEE 2nd International Conference on Pattern Recognition and Machine Learning (PRML). 2021: 92-98.
[9] Yu S, Zhu F, Chen D, et al. Multiple domain experts collaborative learning: Multi-source domain generalization for person re-identification[J]. arXiv preprint arXiv:2105.12355, 2021.