“KaKaloly” a web application for recording, tracking and analyzing calories from Thai food images classification technology using machine learning

This research has created the “Kakaloly” application, which is a food classification application to help users eat the right food and get the right amount of food by applying machine learning to predict the types of Thailand daily food consumption to track and analyze the calorie content of Thai food that users choose to eat. The images of food from the menu were analyzed and the energy value was found to be appropriate for the consumers. This allowed them to control their consumption appropriately. In the design, the developer designed using appropriate UX/UI design and trained the model using 10 types of Thai food and used a total of 4000 image samples to analyze and predict using ML methods. Finally, the quality was measured by experts and satisfaction was obtained from a sample group of 335 working-age personnel working at King Mongkut's University of Technology Thonburi. The results showed that users were satisfied with the ML-based application in various aspects, with an average of 3.0. The topic with the highest score was ML-based work, which users thought was highly accurate. The results of this project were satisfactory in terms of ML-based work efficiency. It is expected that more Thai food content will be added in the future.