Image processing based Classification and Segmentation using LVS based Multi-Kernel SVM

People across the world are showing keen interest in maintaining their weight, selecting healthier foods for eating, and also trying to avoid the obesity problems. Systems with dietary assessment help them very much to maintain their eating events. This paper also proposes the dietary assessment system to help the user and the dietitians to measure and maintain a healthy eating habits for day to day life. There exists a large number of dietary system to identify the food items and to calculate the calorie intake. But the accuracy attains by this proposed method overcomes the existing and in an acceptable range. Here the segmentation is done through the Local Variation Segmentation (LVS) and the segmented food items are recognized by the Multi-Kernel Support Vector Machine (Multi-Kernel SVM). It improves the classification rate and helps to calculate the accurate calorie value.