Font Size: a A A

Finger Vein Feature Extraction Algorithm Based On Adaptive Multi-threshold

Posted on:2020-03-29Degree:MasterType:Thesis
Country:ChinaCandidate:K ZhouFull Text:PDF
GTID:2428330590495705Subject:Software engineering
Abstract/Summary:PDF Full Text Request
With the rapid development of information technology and the continuous improvement of people's living standards,the protection and identification of personal information has received more and more attention.In era of network information,there are more and more application scenarios that need to identify individuals,and they are required to be digitized and personalized.Traditional biometrics such as fingerprint recognition and iris recognition can only use the representation feature of the organism to identify individual identities,and the safety factor is low.Therefore,a more advanced and unique biometric technology,the vein recognition technology has emerged.The finger vein recognition technology not only improves the pain point of the traditional biometrics' technology in practicality and universality,but also greatly improves the safety.Because of the unique superiority,the finger vein recognition technology has become a research hotspot in this field.By analyzing multiple finger vein images,this thesis proposes an adaptive threshold-based finger vein feature extraction algorithm.Firstly,the valley depth and curvature of the finger vein cross section in the image can be detected throught the existing texture characteristics.Secondly,the finger vein image is divided into multiple equal-sized,disjoint sub-images,and these sub-images are divided into a finger vein feature area,a background area,and a noise area according to a given threshold.Finally,the threshold values set for each sub-image are adaptively determined according to different gray mean values,and the feature information of the finger vein is extracted from the image to the utmost.The results of multiple experiments show that the finger vein pattern extracted by the algorithm is clearer and smoother than the traditional algorithm,which restores the actual topological structure of the finger vein as much as possible,and it can better fit the usage scene with low image quality.Furthermore,as a research platform for the core algorithm of finger vein recognition technology,the finger vein recognition simulation system needs to meet the functional requirements necessary for the algorithm research.So,starting from the registration and recognition of the two basic functional modules of the whole identification system,the system covers several corresponding subfunction modules,including the pre-processing module of the finger vein image,the finger vein feature extraction module and the finger vein feature matching module.Meanwhile,combined with the.Net framework and OpenCV library,this system not only realizes the basic functions of the finger vein recognition simulation system,but also provides background management and data storage functions.
Keywords/Search Tags:Biometric identification, Finger vein, Image enhancement, Feature extraction, Pattern recognition
PDF Full Text Request
Related items