Thai food recognition and nutritious calculation on line chatbot with deep learning

Main Article Content

Thanaphon Tangchoopong
Supapong Shakulkhu
Jutamars Phuyawong

Abstract

The majority of people nowadays pay more attention to the nutrients they consume on a daily basis, and most individuals usually take their meal photos. However, the food photo does not utilize to provide the best benefit to building a personal nutritious record. This research presents a system for recording and calculating individual food nutrition from food photos via LINE Chatbot. The proposed system is divided into 2 parts: the first part is the development of a model for image recognition of Thai food by comparing the performance of the model trained with the YoloV5 algorithm and Faster RCNN, then selecting the best model to use in the system. The model was trained on a Thai food dataset including 3,329 Thai food images divided into 20 classes. The experimental result showed that in terms of Average Precision (AP) score at IoU equals 0.5 YoloV5 and Faster RCNN received 0.85 and 0.79 respectively. To find the number of calories received per day, the second part is the gathering of Thai food ingredient nutritious data to calculate food nutrition from recipes, then the system can record the personal daily calories so that it can provide food nutritious data to users via Line Platform.

Article Details

How to Cite
Tangchoopong, T., Shakulkhu, S., & Phuyawong, J. (2022). Thai food recognition and nutritious calculation on line chatbot with deep learning. Rattanakosin Journal of Science and Technology, 4(2), 19–31. Retrieved from https://ph02.tci-thaijo.org/index.php/RJST/article/view/246266
Section
Research Articles

References

Z. Q. Zhao, P. Zheng, S. T. Xu, & X. Wu, “Object detection with deep learning,” A review IEEE transactions on neural networks and learning systems, 30(11), 3212-3232, (2019).

G. Lin, W. Shen, “Research on convolutional neural network based on improved Relupiecewise activation function,” Procedia computer science, 131, 977-984, (2018).

J. Solawetz, YOLOv5 New Version - Improvements And Evaluation. [Online]2020. Available: https:// blog.roboflow.com/yolov5-improvements-and-evaluation

S. Ren, K. He, R. Girshick, & J. Sun, Faster r-cnn: Towards real-time object detection with region proposal networks. [Online](2015). Available: https://arxiv.org/pdf/1506.01497.pdf

A. Bochkovskiy, C. Y. Wang, & H. Y. M. Liao, Yolov4: Optimal speed and accuracy of object detection. [Online]2020. Available:

T. Y. Lin, P. Dollár, R. Girshick, K. He, B. Hariharan, & S. Belongie, “Feature pyramid networks for object detection,” In Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 2117-2125, (2017).

https://arxiv.org/abs/2004.10934

H. Rezatofighi, N. Tsoi, J. Gwak, A. Sadeghian, I. Reid, & S. Savarese, “Generalized intersection over union: A metric and a loss for bounding box regression,” In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 658-666, (2019).

โรงพยาบาลเซนต์หลุยส์. Body Metabolic Rate. [ออนไลน์]2016. [สืบค้นวันที่ 10 ตุลาคม 2564]. จาก https://www.saintlouis.or.th/programs/bmr

iamdmybody. คำนวณแคลอรี่เหมาะสมต่อวัน. [ออนไลน์]2021. [สืบค้นวันที่ 18 กันยายน 2564]. จาก http://iandmybody.com/daily-calorie-calculator/

Saixiii. Line API. [Online]2017. Available: https://saixiii.com/chapter2-line-api-official/

S. Raschka, Model Evaluation, Model Selection, and Algorithm Selection in Machine Learning. [Online]2018. Available: https://arxiv.org/abs/1811.12808

Foodtravel.tv. Foodtravel.tv. [ออนไลน์]2011. [สืบค้นวันที่ 10 ตุลาคม 2564]. จาก http://www.foodtravel.tv/

Agricultural Research Service (ARS). FoodData Central. [ออนไลน์]2007. [สืบค้นวันที่ 15 กันยายน 2564]. จาก https://fdc.nal.usda.gov/