13.07.2015 Views

FACIAL SOFT BIOMETRICS - Library of Ph.D. Theses | EURASIP

FACIAL SOFT BIOMETRICS - Library of Ph.D. Theses | EURASIP

FACIAL SOFT BIOMETRICS - Library of Ph.D. Theses | EURASIP

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

98 8. USER ACCEPTANCE STUDY RELATING TO <strong>SOFT</strong> <strong>BIOMETRICS</strong>Figure 8.6. The majority <strong>of</strong> users, although not strongly familiar with biometrics, emphasized thatthe ranking as it is, holds only in the case <strong>of</strong> equal access verification accuracies for all methods.Figure 8.6: Overall user preference <strong>of</strong> access methods. Small bars represent high rankings.However, to verify the ranked preference users were also given the final task to freely chooseone <strong>of</strong> the four systems with the goal <strong>of</strong> reading a file. Figure 8.7 illustrates the ranking orderedby number <strong>of</strong> selections. The S<strong>of</strong>t biometrics module was most <strong>of</strong>ten chosen and thus the favorite,followed by PIN and fingerprint and finally face based verification (due to similarity with s<strong>of</strong>tbiometrics, but longer acquisition due to the blue mark).Figure 8.7: Freely selected access methods for performing a task.To recapitulate the testing, generally users were eager to explore both the known and newaccess methods. The primary concern <strong>of</strong> all users was related to reliability and accuracy <strong>of</strong> thepresented methods. Furthermore about half <strong>of</strong> the users asked about spo<strong>of</strong>ing methods (e.g. holdinga photo in front <strong>of</strong> the camera) and related countermeasures. These reactions are evidence thatusers are aware <strong>of</strong> novel techniques and have a need to be illuminated on system characteristic inorder to gain the trust <strong>of</strong> the users for the access methods.8.3 Comparison <strong>of</strong> access control systemsUsing the above usability study we proceed to perform a broader comparison by includingcharacteristics pertinent to access control systems, such as cost efficiency, accuracy, processingspeed and maximum amount <strong>of</strong> enrolled users. We identified existing commercially available accesscontrol systems based on fingerprint, face and PIN based methods, see Table 8.2. We selectedcost efficient appliances representing each method and set price and related specifications in comparison.We note that the PIN based system is the most widely available access control system,followed by fingerprint based systems, and only few face recognition based systems. There are nocommercially available s<strong>of</strong>t biometrics access control systems yet.

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!