People Identification based on Geometric Face Features for Cloud Services

Abstract

Recently, the services provided by cloud computing play an important role in many areas of life. The security component represents the major challenge that hinders the adoption of cloud services in banking, healthcare and other transactions. In this paper, secure identification approach is proposed to use cloud services by recognizing authorized persons based on facial features. In the proposed approach, the face image is sent to the cloud side without any processing. When it is received in the cloud side, the image is processed and the geometric features are extracted using moment features extractor. The features of the face are used during the matching phase by using two different classifiers to make the final decision to grant authorization to access cloud services. The experimental results showed the efficiency and accuracy of the proposed approach. The performance is test using many facial images with online applications. Where an overall accuracy level of approximately 96% was obtained.