Fuzzy Logic Classification of Handwritten Signature Based Computer Access and File Encryption

fuzzy logic;security;handwriting;signature;Fourier transformation;k-Nearest Neighbor;College of Arts and Sciences: Computer Science

Kwarteng, EmmanuelFlint ;

University of Michigan

关键词: fuzzy logic;    security;    handwriting;    signature;    Fourier transformation;    k-Nearest Neighbor;    College of Arts and Sciences: Computer Science;   

Others  :  https://deepblue.lib.umich.edu/bitstream/handle/2027.42/117719/Kwarteng.pdf?sequence=1&isAllowed=y

瑞士|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

Often times computer access and file encryption is successful based on how complex a password will be, how often users could change their complex password, the length of the complex password and how creative users are in creating a complex passsword to stand against unauthorized access to computer resources or files. This research proposes a new way of computer access and file encryption based on the fuzzy logic classification of handwritten signatures. Feature extraction of the handwritten signatures, the Fourier transformation algorithm and the k-Nearest Algorithm could be implemented to determine how close the signature is to the signature on file to grant or deny users access to computer resources and encrypted files. lternatively implementing fuzzy logic algorithms and fuzzy k-Nearest Neighbor algorithm to the captured signature could determine how close a signature is to the one on file to grant or deny access to computer resources and files. This research paper accomplishes the feature recognition firstly by extracting the features as users sign their signatures for storage, and secondly by determining the shortest distance between the signatures. On the other hand this research work accomplish the fuzzy logic recognition firstly by classifying the signature into a membership groups based on their degree of membership and secondly by determining what level of closeness the signatures are from each other. The signatures were collected from three selected input devices- the mouse, I-Pen and the IOGear. This research demonstrates which input device users found efficient and flexible to sign their respective names. The research work also demonstrates the security levels of implementing the fuzzy logic, fuzzy k-Nearest Neighbor, Fourier Transform.