Embedded system for face identification based on Iris Detection

Abstract

This research describes the design and implementation of an Embedded System for Biometric Identification based on Iris Detection (BIOI2D). It is based on single board computer (SBC) and utilizing GNU/Linux operating system (OS) which allows the use of open source resources such as libraries, kernels drivers and GNU C compiler in developing and implementing this system. The BIOI2D is designed to operate in real-time mode to execute these following tasks: face (image) capture, preprocessing and matching with database using predetermine user identification number to reduce the processing tasks. The main component for Face Reader is a x86 Single Board Computer (SBC) TS-5500. Other components connected to the SBC consist of LCD Panel, USB Web Camera, Compact Flash Card, PCMCIA Wireless Network Card and keypad. BIOI2D software design is structured in five modules namely as User Interface, Image Acquisition, Image Preprocessing, Network and Biometric Identification. The development of user interface module involves the integration of LCD panel and matrix keypad with the Face Reader system. Image Acquisition Module is developed by utilizing the video4Linux API. BIOI2D is designed to operate in real-time mode, which requires the face identification and recognition software to identify the person face from the captured image. The image preprocessing processes are used to perform initial processing to the captured image by removing background and increased the local contrast. The image preprocessing technique performs are the colour space conversion, motion analysis technique, histogram equalization and image scaling. Biometric identification module is a face recognition system based on iris detection. The recognition system is based on template matching. Output image from the image preprocessing module is used as an input face image for biometric identification module. The recognition process is done one-to-one matching (direct matching) by using user ID to reduced processing time, thus increase the efficiency of the system. The system evaluations are focusing on hardware performance, image preprocessing process and face identification process. Experiment with set of test images consist of 100 images of 10 persons shows the successful rate for the face identification system is 73% and percentage of matching that is reliable and should be used as a threshold for system implementation is 98%. This thesis demonstrate that embedded processing technology, in particular the x86 processor TS-5500 SBC, has been developed well enough to make it useable for implementing a computer vision system adequate for embedded system for biometric identification based on iris detection.