MULTI-LABEL FOOD IMAGE DETECTION AND CLASSIFICATION WITH FASTER R-CNN

Food image detection and classification has become an important field due to its applications in health monitoring, food analysis, and automatic food recognition systems.In addition, multi-label food image detection and classification to identify multiple foods contained in a single image has become an important task in the field of computer vision.This paper proposed a multi-label food detection and classification approach by combining Faster R-CNN with the feature extraction capabilities of a DCNN.A newly developed Myanmar food dataset, known as the MF15 dataset, consists of fifteen categories.The proposed method achieved an overall mAP score of 89% for the fifteen-class dataset.