Introduction
Biometric systems are automated recognition systems that allow the identification of persons based on biological or behavioral characters. Depending on the application of the system and its context, the system can be identified as a verification or identification system. A verification system works with a pre-stored version of user data within a system to provide access while the identification system searches a database for the template match (Lien, Yang, Hsiao, & Kao, 2013). A verification system will normally accept or reject the claim of identity where the one-to-one comparison fails to yield a positive result. In the case of an identification system, the user need not claim an identity since the system can establish the identity. In such a system, any human behavior or characteristic can be used as an identifying factor so long as they can establish the basic requirements for the system: universality, distinctiveness, permanence, and collectability (Wang, et al., 2013) .
Each biometric application has its strengths and weaknesses. The choice of which biometric measure to use will normally depend on the application of the system. Since no single application can meet the needs of all systems, it is necessary to determine the nature of the operations in the water project to determine the possible system that could be used (Chen, et al., 2008). This paper discusses the implementation of a biometric security system for a large water project.
Delegate your assignment to our experts and they will do the rest.
Background Information
Seattle Water Company is a service provider that services approximately 20 million people in the city of Seattle and its environs. As a result, the water project for the city estimates to about 200 million gallons each day being distributed between the city and its environs. Following damages from the weather and recent biological weapons attack threats, the company has approved an added security layer on the controls of water supply to the city through the introduction of a biometrics security system. The intention of the system is to provide an identification method for persons going into the facility for the purpose of restricting outsiders from accessing the facility. The company has contracted Elite Solutions for this security implementation, where it is tasked with describing the most suitable system that could be used at the Seattle Water Company water management plant. The purpose of this report therefore, is to outline the needs of the Water Company with regards to security and determine the most effective system. A description of the system’s functions will be given alongside product specifications.
Since only limited staff is allowed into the water management facility, there is an increased need to improve security within this section. Limited access to this area ensures that the water supply to over 20 million people remains constant, and ensures that equipment stays out of reach of ordinary company workers. This ensures that the security measures are in place for the adequate protection of the water supply for the city. This therefore calls for the fingerprint identification system. This will involve the collection of minutiae from fingerprint data which will be used for the purpose of identification (Cappelli, Ferrara, & Maio, 2012).
System Description
The minutiae pattern is the most common method of representing fingerprint security data. This approach matches corresponding minutiae through pairing and alignment in order to verify the identity of the person (Hong, Wan, & Jain, 1998). Obtaining and extracting minutiae data is not easy since minutiae data depends on the accuracy of fingerprint images as well as clarity of the imaging. Furthermore, extraction algorithms will normally depend on the quality of image presented in order to interpret the minutiae data. Therefore, the first step of this process will be to obtain fingerprint enhancement tools for the accurate extraction of minutiae data. This will be done to improve the efficiency of the identification algorithm through fingerprint image quality improvement (Hsieh, Lai, & Wang, 2003). Furthermore, this process reduces general fingerprint “noise” and gives accurate distinction between ridges and valleys for proper fingerprint identification.
Having considered that the fingerprint enhancement is a very necessary part of the process, we consider the Gabor model of filtering, which is the most common method available. This method has been used since it provides the best quality images necessary for restricted area access such as control rooms for nuclear weapons, airport controls among other areas. Therefore, the Gabor filter allows the individual to have the clearest form of control access to the restricted area (Jain & Farrokhnia, 1991). The Gabor filter involves the identification of edges in the print and identifying those edges based on a discrimination method.
Thereafter, the critical step of extracting minutiae data begins. Here, the quality of input images is critical in obtaining the correct data. Fingerprint enhancement algorithm is therefore involved in ensuring that the system can have the clearest fingerprint data. This process involves a series of techniques having different stages. After the process is complete the image obtained is set into the minutiae extraction module where minutiae are extracted for identification (Wang, Li, Huang, & Feng, 2008). The fingerprint extraction process generally follows four steps: processing, minutiae extraction, registration and patterning. Consider the figure below:
Figure 1 : Image Processing Steps
The system involves a system that can collect minutiae samples. Once a person needing access to the area presses their finger against the smooth surface, the two minutia samples are compared for similarity to determine whether they are from the same finger. The alignment-based matching system is engaged to determine the verification of the person’s identity based on two stages, namely: alignment and matching (Zhu, Yin, & Zhang, 2005). The matching algorithm considers the two prints based on a number of standards, including ridges and determines the percentage match. Where the alignment calculates the match disregarding direction of the print, a value is given to the system. If the matches are beyond the threshold point of match, the individual gains access to the restricted area. As a result, the system becomes a verification system that identifies persons already given access to the facility and allows them to move in.
Testing and Implementation
Based on analyses processes described, the registration system will be built where a fingerprint database will be. MATLAB coded instructions will be provided for image processing, while matching and database specifications will be coded in C#. The necessary testing was performed using a Pentium Core i3 – 1GB RAM and 2.3GHz machine. Here, the registration mechanism involved three major options: automatic, manual and update. Verification function is implemented. Here, the user is asked to identify his ID from the system for which a fingerprint in the database will be brought up for comparison with his. The match function in the system matches the outside fingerprint and produces a match rate against the fingerprint in the LOD system. The system provides a threshold percentage match for giving access to individuals to the restricted area. If the print matches the LOD print past the threshold percentage, the user is granted access. Nonetheless, if the threshold percentage is not reached, the system prints an error message and the user is denied access to the restricted area.
Summary
This identity verification system ensures that restricted access is maintained to Seattle Water Company premises, especially where water control rooms are located as these areas are quite sensitive. Therefore, the Seattle Water Company will install identity verification methods that will involve a series of processes to ensure that only classified personnel are allowed into the control rooms. The system will have already-input personal fingerprint data refined by the Gabor filter and other forms of image processing. This ensures that best quality images are used for minutiae extraction used in fingerprint identification. Thereafter, the system can be used to determine the identity of persons seeking entry into restricted areas by running an alignment and match code for all fingerprints pressed on the smooth service at the entrance of the restricted area. Personnel matching the existing fingerprint database beyond a certain threshold percentage will gain entry while unauthorized persons will receive an error message.
References
Cappelli, R., Ferrara, M., & Maio, D. (2012). A fast and accurate palmprint recognition system based on minutiae. IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), 42(3) , 956-962.
Chen, Y. C., Chen, L. K., Tsai, M. D., Chiu, H. C., Chiu, J. S., & Chong, C. F. (2008). Fingerprint verification on medical image reporting system. Computer methods and programs in biomedicine, 89(3) , 282-288.
Hong, L., Wan, Y., & Jain, A. (1998). Fingerprint image enhancement: algorithm and performance evaluation. IEEE transactions on pattern analysis and machine intelligence, 20(8) , 777-789.
Hsieh, C. T., Lai, E., & Wang, Y. C. (2003). An effective algorithm for fingerprint image enhancement based on wavelet transform. Pattern Recognition, 36(2) , 303-312.
Jain, A. K., & Farrokhnia, F. (1991). Unsupervised texture segmentation using Gabor filters. Pattern recognition, 24(12) , 1167-1186.
Lien, C. Y., Yang, T. L., Hsiao, C. H., & Kao, T. (2013). Realizing digital signatures for medical imaging and reporting in a PACS environment. Journal of medical systems, 37(1) , 1-11.
Wang, F., Ge, B., Zhang, L., Chen, Y., Xin, Y., & Li, X. (2013). A system framework of security management in enterprise systems. Systems Research and Behavioral Science, 30(3) , 287-299.
Wang, W., Li, J., Huang, F., & Feng, H. (2008). Design and implementation of Log-Gabor filter in fingerprint image enhancement. Pattern Recognition Letters, 29(3) , 301-308.
Zhu, E., Yin, J., & Zhang, G. (2005). Fingerprint matching based on global alignment of multiple reference minutiae. Pattern Recognition, 38(10) , 1685-1694.
