Font Size: a A A

Research On Palm-Print Principal Lines Extraction On The Whole Palm

Posted on:2014-01-11Degree:MasterType:Thesis
Country:ChinaCandidate:M Y WeiFull Text:PDF
GTID:2248330395489587Subject:Measuring and Testing Technology and Instruments
Abstract/Summary:PDF Full Text Request
Compared with the password, identity card or other traditional identification methods, the biological characteristics identification based on the palm-print, iris and hand shape is more popular because of its stability, conveniently, and safety. The palm-print image can be obtained by non-contact acquisition, which is very suitable for the equipment users who are worried about disease infection. As an obvious characteristic of the palm-print recognition technology, compared with other line features principal lines have distinct difference and stability on quantity, direction and distribution and small light interference.Because of the non-contact acquisition, the size of palm-print feature area, the illumination uniformity, and the palm angle have uncertain factors. These make the existing palm-print principal lines extraction method have limitations. A novel approach of principal lines extraction on the whole palm is proposed to dissolve the problems of light interference, the complexity of algorithm and the inaccurate ROI positioning in traditional extraction process of principal lines.This algorithm is based on gray level image, and uses the priori knowledge of the direction certainty, gray-scale similarity and the regularity of start points and end points of principal lines to extract the principal lines. According to the characteristics of the principal lines emotion line, intelligence line, life line, career line and success line are divided into three categories to extraction by tracking method. This method effectively avoids the complex computing on images enhancing, ROI positioning and palm-print image binaryzation in traditional methods and tracks principal lines on gray level image directly. This method retains the principal lines information of direction, relative position and width. Experiments on HKUST show that this method can reduce the affect of light, translation and rotation, and the principal lines extraction rate is95.09%. By the identification method based on principal lines’block energy, the identification rate is81.33. Compared with the existing algorithms, this method has obvious advantages on low complexity, high extraction rate and integrated information and provides supports for the research of multi-model hand recognition method.
Keywords/Search Tags:palm-print principal lines, whole palm, principal lines extraction, extraction rate
PDF Full Text Request
Related items