The Study of Palmprint Image Identification Based on Wavelet Analysis
|School||Zhengzhou Institute of Light Industry|
|Course||Control Theory and Control Engineering|
|Keywords||Palmprint recognition Preprocessing Feature extraction Wavelet analysis Wavelet packet analysis|
The rapid development of information technology promotes the progress of the society, and then the modern society of information technology proposes updating and higher requirements. Computers have made the whole social information and networking, and the information and networking society put forward higher requisitions for all kinds of information and system security. Identity authentication is one of the basic methods for people to strengthen the security of information and system. The technique of making use of the biological features of human body itself to automatic identification called human biometrics, known as biometrics. The palmprint recognition technique is an important component of biometrics. Because the palm has good uniqueness and stability, and contains rich ridge information so palmprint recognition technology has seen rapid development and it been widely used in the information security fields.This paper did a detailed research and analysis for the palmprint recognition technology development course and the present situation. Based on the study of the main palmprint recognition technology, studied the palmprint image preprocessing and feature extraction technology. In the hand of palmprint image preprocessing, improved the algorithm that using the palm geometric profile feature points to establish coordinate system and extracting the region of interested(ROI) on previous, added the ROI orientation method; put forward the ROI extraction algorithm that based on the location of triangle points in high resolution and high quality palmprint images. In the hand of feature extraction, by using wavelet analysis to decompose the ROI for extracting wavelet energy features, and experiments analyzed and compared the recognition rate under the different wavelet basic functions and different wavelet decomposition layer, laid a foundation for follow-up study of palmprint recognition basing on wavelet decomposition; proposed a new method of palmprint recognition based on wavelet packet analysis, extracted wavelet packet decomposition coefficient characteristics and energy characteristics respectively, used hierarchical recognition method for matching recognition, improved the accuracy and efficiency of the algorithm, and compared with other kinds of main palmprint identification methods, the results show that the algorithm has good feasibility and accuracy.