Research And Design For The Access Control System With Palm Shape Recognition Based On ARM

Since the event of 911, an urge need for more security equipments arose worldwide. With the increasingly wide use of Access Control System (ACS), improving the security of this system becomes a important subject. The fourth generation of ACS outstands with its technology based on biological features recognizing. It exploits physical features in human body such as fingerprint, iris, face shape and palmprint, or behavioral features such as gait and signature, to identify a person so as to enhance or replace traditional personal identification methods.In this paper we propose a new scheme based on palmshape recognizing. The CPU of ACS is a EP9315 chip with a core of ARM920T. The hardware platform has been designed and tested.Our work includes the following three parts.First, we investigate the skeleton shape of human hand and introduce a new algorithm based on recognizing it. In this way, errors caused by circumrotation of fingers have been overcome.Second, we investigate the palm features. After an image-processing, three critical lines of human palm is obtained. We pick out the ends spots of those three lines and choose isometry spots in those lines, then we match the information carried by these two kinds of spots. Our method has a good recognizing rate.Third, combing the skeleton shape and palm features, we realize palm recognizing. The skeleton serves as a rough classifying parameter, while the palm features serves as the further classifying parameter. Our algorithm features with a fast response and a stable and high recognizing rate, which may provoke new thoughts and ideas on classifying rules.An access control system based on ARM has also been designed in this paper. Moreover, an mini system with EP9315 has been successfully structured, the circuit of which is tested and debugged.The research and design work of this paper improves the security of the access control system by increasing the rate of recognition of the palm shape recognition method. The design and test for the hardware platform of ARM has great value in practicing engineering.