Dissertation > Excellent graduate degree dissertation topics show
Palmprint Identification Using DCT-based Multi-color Component Images Fusion
Author: ZhouJiaJia
Tutor: SunDongMei
School: Beijing Jiaotong University
Course: Pattern Recognition and Intelligent Systems
Keywords: Color channels Feature-level fusion Palmprint DCT Color space
CLC: TP391.41
Type: Master's thesis
Year: 2009
Downloads: 93
Quote: 1
Read: Download Dissertation
Abstract
|
With the rapid development of social information , digitized, networked , the traditional password , the password has been unable to meet the demand of the people of the safety certification . Should be born biometric technology has become the development trend of authentication . Palmprint identification system many advantages because of its own , and become a hot topic of academic research , and the technology aspects of information security , as well as evidence of a crime investigation has a wide range of applications and good prospects . The palmprint recognition of the important one is the extract better discriminative information . Analysis of the major existing palmprint feature extraction algorithm , this paper, palmprint identification using color information , including the two parts of the single color channel and multi- color channel feature fusion palmprint identification algorithms . And achieved better results on low-resolution color palmprint image library , the main work of this paper are as follows : (1) This paper presents a palmprint recognition algorithm Improved -DCT DCT-based gray space with PCA The / WO3 algorithm and the 2D-DCT algorithm comparison , the experiment proved the Improved-DCT algorithm can effectively reduce the feature vector dimension , to improve the recognition rate , and to reduce the recognition time . ( 2 ) summarize and analyze the existing palmprint feature feature extraction algorithm, using the color information is ignored palmprint identification . The experiments show that color channel , R -channel ( RGB color model ) , I channel (YIQ color model ) as well as the S channel (HSI color model ) , better than gray space identification . ( 3) In this paper, in the low-resolution color palmprint identification using feature fusion method based on multi- color channels . The experiments show that the recognition rate and EER value of the multi - color channel feature fusion method is better than the results of the single color channels .
|
Related Dissertations
- Research and Embedded Implementation of Image Watermark Algorithm,TP309.7
- Digital image forensics technology research,TP391.41
- Nonlinear color image watermarking detection program of study,TP309.7
- The Study of Palmprint Image Identification Based on Wavelet Analysis,TP391.41
- Non-equilibrium Encoding Scheme of Distributed Video Coding Based on LDPCA,TN919.81
- Research on Novel Steganography Algorithm with High Capacity,TN918
- Based on the robust non-contact acquisition palmprint recognition,TP391.41
- Re-compression JPEG image tamper detection and localization,TP391.41
- Color medical image segmentation technology research,TP391.41
- Video sequences based image stabilization and target tracking technology,TP391.41
- Model based on visual characteristics and image enhancement algorithms and performance analysis,TP391.41
- Palmprint image feature extraction method,TP391.41
- Based on the H.264 standard video watermarking technology in the network learning platform in the Application and Implementation,TP309.7
- Color Face Detection and Recognition,TP391.41
- HD image license plate location based vehicle tracking and violation detection,TP391.41
- Using wavelet transform and optimize embedded block truncation coding for image compression,TP391.41
- Fuzzy C-means clustering algorithm for color image segmentation,TP391.41
- Road traffic sign detection and recognition technology,U495
- The Study on Face Recognition Technology Based on Subspace Methods,TP391.41
- Research and implementation of multi-resolution digital image watermarking algorithm based,TP309.7
- Static image watermarking processing in the FPGA - based implementation,TP309.7
CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Information processing (information processing) > Pattern Recognition and devices > Image recognition device
© 2012 www.DissertationTopic.Net Mobile
|