الفهرس 
INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 7 |  |  |
| | | from | 7 | | |
Abstract
The development of accurate and reliable security systems is a matter of wide interest, and in this context Biometrics is seen as a highly effective automatic mechanism for personal identification. A biometric identity verification system tries to verify user identities by comparing some sort of behavioral or physiological trait of the user to a previously stored sample of the trait. The recent developments in the biometrics area have led to smaller, faster and cheaper systems, which in turn has increased the number of possible application areas for biometric identity verification. Palmprint authentication is one of the relatively new physiological biometric technologies which exploit the unique features on the human palmprint, namely principal lines, wrinkles, ridges. The rich texture information of palmprint offers the effective means in person authentication due to its non-intrusive, user friendly, stable, low-resolution imaging and low cost requirements.Face is one of the commonly acceptable biometrics used by humans in their visual interaction. The challenges in face recognition stem from various issues such as aging, facial expressions, variations in the imaging environment, illumination and pose of the face.Among biometric methods, Local Binary Pattern (LBP) operator is widely used as a texture descriptor to extract representative features, ”uniform” LBP was proposed and its effectiveness has been validated. However, all ”non-uniform” patterns are clustered into one pattern .By this way many discriminative but non-uniform patterns fails to provide useful information. In this thesis, we propose a novel face and palmprint recognition technique based on hierarchical multi scale local binary pattern (HMLBP) and adaptive local binary pattern with directional statistical features (ALBPF) to increase the recognition rate by building a hierarchical multi scale ALBPF histogram for an image. The usefu I information of ”non-uniform” patterns at large scale is dug out from its counterpart of small scale. Experiments was demonstrated on two palm print databases and two face databases, our experimental results indicate that the proposed methods have better recognition rates than the original LBP, ALBPF and HMLBP methods. Recently many variants of LBP have been proposed for texture analysis. In this thesis we present the effect of the modified LBP operators, Sobel-LBP and Center Symmetric LBP (CS-LBP) on palmprint recognition rate. CS-LBP reduces the LBP histogram dimension while Sobel-LBP enhances the edge information by filtering palmprint image with Sobel operator. The proposed methods are compared with the original LBP on gray¬level images for palmprint recognition. The experimental results indicate that the proposed methods have higher identification rates than the original LBP.