K-Means Food Object Clustering and Feature Detection using MSERF and SURF Region Points

Obesity is a perilous consumer of human lives, and is addressed with growing concerns globally. One of the primary reasons for the origination of surgical studies in Bariatrics is the consumption of unhealthy and indolent practices. Multifaceted literary studies are associated with gremlins of the human body, along with food calorie recognition methods. Most commonly many of the health hazards arise with food regimen that individuals choose to consume. Therefore, identification and anatomization of the food and calorie intake is a cardinal aspect which requires meticulous approaches. The whilom approaches relating to food calorie identification and segmentation have been implemented with K-means clustering and color space segmentation approaches. However, this study focuses on the food image enhancement, feature identification and clustering using MSERF and SURF detection parameters. The proposed work also ensures that the implemented work forms a strong pre-processing method to better accuracy of classification for further stages of study. The indagated study is simulated using MATLAB and the results are successfully acquired.