Cloud based virtualization for a calorie measurement e-health mobile application

Smartphones have transformed people approach towards technology. It not only empowers them to use it for communication but also for tracking and maintaining health-fitness via mobile applications. With the increasing number of complex mobile applications, the mobile, with its limited resources (in terms of the computational power and storage capacity) cannot efficiently run these applications autonomously. Similarly our e-health mobile application (Eat Healthy Stay Healthy) requires a platform to run highly computational intensive algorithms like the deep learning (for recognizing food images) and calorie measurement would need higher processing power to perform optimally. We propose a cloud based virtualization model that provides our e-health application with the required computational power that it needs to perform efficiently and at the same time would also give it the flexibility to make use of the various cloud resources. Our model comprises of concepts like virtual swap between various mobile sessions that assist the system for faster processing and intelligent decision mechanism for distributing the task of image processing to cloud servers. By implementing intelligent decision mechanism, the final calorie computation significantly improved by 20.5% while implementing deep learning in cloud.