Abstract:
Keeping our homes and office spaces safe and secure from intruders has always been a priority in our lives and like everything else, adding technology in this aspect can make our existing security measure better and more efficient. For our thesis project, we are aiming to create an intelligent security system that will incorporate Raspberry Pi (RPi), Camera Module and Android operating system to make a real-time, user friendly and reliable form of security and surveillance system. The RPi uses Haar Cascade algorithm to extract facial features in given images to detect faces. With a machine learning approach along with Eigen faces algorithm, the program is trained to recognize the features of a face from images already fed to the system. The RPi uses opencv application to compare the new image with the images already stored in the SD Card for cross-referencing purposes. The images for cross-referencing can be copied to the SD Card with a computer or can also be uploaded via the Android application. If the system finds that the image does not match any image in the directory it uploads that image to the server set up in the RPi given that internet connection is available. On the other hand, if it matches, the RPi does not do anything. Once uploaded, the image can be viewed from the Android Application. The app
will offer the user multiple options on notifications, image receiving/storing and disabling/enabling the service. By using this system, the user will get real time alert when there's activity within a predetermined area, check whether it is an intrusion or not and store photos for further investigations if required.
