Cloud Powered Real Time Facial Recognition

Abstract

The use of video surveillance system is becoming more and more important for investigation and deterrent of crimes due to an increasing number of crimes and suicide bombings. Therefore, identifying a person from a video stream becomes more and more frequent. The task is very difficult especially when suspects have to be caught in a crowded area and in minimal amount of time as the real time matching of an image is still a challenge. Addition to this there is a huge variation in human face image in terms of size, pose and expression. To combat this issue, we are proposing a Real Time Facial Recognition system that is scalable and process all the data using a Cloud powered backend server. The system will be able to detect, identify suspects and also notify on-duty guard via text message using Kannel technology. Meanwhile, using cloud computing technologies at the backend server will help scaling and remotely accessing the system from any location at any desired moment. So far, the system has been tested on four different machines having different processing capabilities. Out of which, the Huawei cloud has proved to give the best performance. Moreover, algorithms such as AdaBoost with Haar cascade, PCA and LDA would provide simple, fast and high accuracy face detection and recognition. Other technologies include Open Stack and Huawei toolkit for Cloud, OpenCV and Kannel.