Modified online signature recognition using software as a service (SaaS) model on public cloud

Signature recognition is one of the most significant research fields in the area of identity recognition based on biometrics. To create a highly Scalable, Pluggable and Faster online signature recognition system, the authors Dr. Vinayak Bharadi and Mr. Godson D'silva had proposed an architecture for implementing online signature recognition system on a public cloud [1]. This proposed architecture was restful in nature, allowing any type of client to communicate with it who can speak HTTP. But the implementation cost of this architecture was very high as it needed web role vm's and worker role vm's running at the same time in the cloud service. Secondly the azure queue which was utilized as a communication medium can't be utilized for communication outside the cloud service with other client devices and application. As a result more number of transactional routes are required to perform even a simple operation such as uploading, thus increasing the time and cost. In this paper a new modified architecture is proposed which will resolve all the issues discussed above and create a highly Scalable, Pluggable and Faster online signature recognition system with more cost effective approch.