Online signature verification based on dynamic features from gene expression programming

Hua, Tan; He, Lei; Huang, Zhijian; Zhan, Hang

doi:10.1007/s11042-021-11063-z

Cited by 11 publications

(1 citation statement)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…m. After crossover between signature chromosomes, the chromosomes (along with their fitness vector, fitness 2) are used in the current generation detection to avoid duplication and revert the process to the rest of the signature chromosomes. n. Steps (8) through (13) reapplied to the remaining signature chromosomes from the previous generation. o.…”

Section: Genetic Recognition System (Grs)mentioning

confidence: 99%

Off-line handwritten signature recognition based on genetic algorithm and euclidean distance

Mohammed

Hussien

2023

IJ-AI

View full text Add to dashboard Cite

Biometric authentication is a technology that has become significant in the high level of personal identity security. This paper provides a signature recognition system. This paper provides a static signature recognition system (SSRS). We have classified the signature in two ways. The first method uses the genetic algorithm (GA), considering that the signature is the chromosome with 35 genes, and each feature is a gene. With applying the processes of the GA between chromosomes and the formation of generations in sequence until we reach the optimal solution by finding the chromosome closest to the chromosome that enters the system. In the second method, we have classified the signature by calculating the Euclidean Distance between the query signature and the signatures stored in the database. The signature closest to a confirmed threshold is considered the desired goal. The database uses 25 handwritten signatures (15 signatures for training and five original signatures, and five fake signatures written by other people for testing), so we have a database of 500 signatures. With a 94% discrimination rate, the genetic recognition system (GRS) was able to access the solutions, and with a (91% rate) the euclidean recognition system (ERS) was done. The application uses MATLAB.

show abstract

Section: Genetic Recognition System (Grs)mentioning

confidence: 99%