Indonesian Journal of Electrical Engineering and Informatics (IJEEI) Vol. 5, No. 4, December 2017, pp. 317~325 ISSN: 2089-3272, DOI: 10.11591/ijeei.v5i4.361  317 Received Aug 1, 2017; Revised Oct 20, 2017; Accepted Nov 3, 2017 Development of Face Recognition on Raspberry Pi for Security Enhancement of Smart Home System Teddy Surya Gunawan* 1 , Muhammad Hamdan Hasan Gani 1 , Farah Diyana Abdul Rahman 1 , Mira Kartiwi 2 1 Department of Electrical and Computer Engineering, International Islamic University Malaysia 2 Department of Information Systems, International Islamic University Malaysia Jalan Gombak, 53100 Kuala Lumpur, Malaysia e-mail: tsgunawan@iium.edu.my Abstract Nowadays, there is a growing interest in the smart home system using Internet of Things. One of the important aspect in the smart home system is the security capability which can simply lock and unlock the door or the gate. In this paper, we proposed a face recognition security system using Raspberry Pi which can be connected to the smart home system. Eigenface was used the feature extraction, while Principal Component Analysis (PCA) was used as the classifier. The output of face recognition algorithm is then connected to the relay circuit, in which it will lock or unlock the magnetic lock placed at the door. Results showed the effectiveness of our proposed system, in which we obtain around 90% face recognition accuracy. We also proposed a hierarchical image processing approach to reduce the training or testing time while improving the recognition accuracy. Keywords: eigenface; Principal Component Analysis; face recognition; Smart Home System. 1. Introduction Nowadays, there is a growing interest in the smart home system using Internet of Things. According to the report in [1], 72% respondents said self-adjusting thermostat and 71% said that doors that can be locked from a remote location, were the most important features when it comes to the most desired smart home devices. Face recognition is one of the most used biometric identidication system beside fingerprint and iris [2]. A typical face recognition system is shown in Figure 1. Figure 1. Typical Face Recognition System There are many algorithms have been developed for face recognition algorithm, including appearance based [3], active appearance [4], support vector machines (SVM) [5], Bayesian model [6], deep learning neural network [7], and texture based [8]. In this research, we will focus on appearance based face recognition which includes direct correlation, eigenface, and fisherface. Unlike direct correlation algorithm that uses face image in their original image size, eigenface and fisherface algorithms reduce the image to the most discriminating factor and find similarity between images in a reduced dimension image size [9]. The fisherface algorithm uses inner class information for face classification and it can use multiple faces of a person to establish in-class variation to maximize class separation. In contrast, the eigenface algorithm uses one image per person applying the variation in one image to the overall recognition process. The eigenface algorithm is susceptible to the variation in illumination or facial expression. However, the computational requirement is lesser compared to the fisherface