Sulong, Siti Masrina (2002) Pengesanan minutiae imej cap jari berskala kelabu menggunakan algoritma susuran batas. Masters thesis, Universiti Teknologi Malaysia, Faculty of Computer Science and Information System.
Restricted to Repository staff only
PDF (Table of Contents)
PDF (1st Chapter)
Fingerprint-based identification has been known and used for a very long time. The fingerprint image contains narrow ridges separated by narrow background valleys. Most Automatic Fingerprint Identification Systems (AFIS) are based on minutiae matching due to its uniqueness. Minutiae are essentially terminations and bifurcations of the ridge lines that constitute a fingerprint pattern. According to conventional technique, minutiae detection is applied to a binarized and thmned image. This requires time consuming and lots of information may be lost during the processes. Thus, Maio and Maltoni (1997) proposed a method to detect minutiae in gray scale fingerprint images based on ridge line following algorithm, which is to trail the ridge line according to the fingerprint directional image. Even though, Wse minutiae still detected during the process. Based on these errors, an improved Ridge Line Following Algorithm has developed which consist of five processes; tangent direction computation, sectioning and maximum determination, end point detection, bifurcation point detection, and post-processing. A precise tangent value will contribute to a detection of true minutiae. Sobel operator is applied as other alternative of tangent direction computation. A reversing technique is proposed in order to detect true end point in the fingerprint image. A post-processing technique is also added to minutiae extraction process to determine the true minutiae. Raw fingerprint images and filtered images (noise removal and enhancement) are tested to determine the efficiency of the algorithm developed. The improved algorithm has applied to 350 image samples obtained from National Institute of Standards and Technology (NIST). As a result, techniques proposed in this research gives a higher percentage of 60 percent in true minutiae detection compared to previous techniques, although it still could not detect all the true minutiae correctly.
|Item Type:||Thesis (Masters)|
|Additional Information:||Tesis (Sarjana Sains (Sains Komputer)) - Universiti Teknologi Malaysia, 2003; Supervisor : Prof. Dr. Ghazali Sulong|
|Uncontrolled Keywords:||minutiae detection, gray scale fingerprint images, narrow ridges, Automatic Fingerprint Identification Systems (AFIS), data processing|
|Subjects:||Q Science > QA Mathematics > QA75 Electronic computers. Computer science|
|Divisions:||Computer Science and Information System (Formerly known)|
|Deposited By:||Ms Zalinda Shuratman|
|Deposited On:||01 Oct 2007 03:54|
|Last Modified:||04 Sep 2012 03:29|
Repository Staff Only: item control page