Mask R-CNN and GrabCut Algorithm for an Image-based Calorie Estimation System

Background: A calorie estimation system based on food images uses computer vision technology to recognize and count calories. There are two key processes required in the system: detection segmentation. Many algorithms can undertake both processes, each algorithm with different levels of accuracy. Objective: This study aims improve accuracy calculation segmentation using a combination Mask R-CNN GrabCut algorithms. Methods: The mask generated from were combined create new mask, then used calculate calorie. By considering image augmentation technique, observed evaluate method’s performance. Results: proposed method could achieve satisfying result, an average error value less than 10% F1 score above 90% all scenarios. Conclusion: Compared earlier studies, obtain more result calculating calories shapes. Keywords: Augmentation, Calorie Calculation, Detection