Ingredients estimation and recommendation of Thai-foods


  • Lawankorn Soimart


Food recognition, Deep learning, Ingredients estimation of foods


Thai-foods are one of the most delicious tastes in the world. Many visitors around the world are fascinated in
Thai-foods. In 2011, Thailand had the foods more than other countries from the list of “World’s 50 Most Delicious
Foods” such as Tom-yam-goong (4th ranking), Pad-thai (5th ranking), Som-tam (6th ranking), Mas-sa-man (10th ranking),
Green-curry (19th ranking), Thai-fried-rice (24th ranking) and Moo-nam-tok (36th ranking), respectively. However, some
visitors cannot recognize these foods’ name. What’s more, the ingredients of food are difficult to be considered by
the visitors’ eyes. To that end, this paper firstly introduces a novel ingredients estimation of Thai-foods as a Thaifood guidance of visitors that opens a new gate for Thai-foods with image processing and artificial intelligence. For
the measurement, our system provided the high correctness of estimation in the criterions of accuracy, recall and
precision which were summarized in term of their average values as 0.74, 0.64 and 0.78 respectively.


Download data is not yet available.




How to Cite

Soimart, L. (2017). Ingredients estimation and recommendation of Thai-foods. Creative Science, 9(2), 509–520. Retrieved from