A Novel Fingerprint Compression Based On Accessible Dictionary Learning

DOI

No DOI

Published In

Volume 2, Issue 1 (2022)

Published Online

January-February 2022

Pages

12-13

Article Type

Research Article

Check for updates

No DOI available for cross-reference

Anewfingerprint compressiontechniquebased ontheconcept of sparserepresentationisintroduced.Givenaset of interesting imprint fixes, an over complete word reference is created and the patches are tended to as a deficient linearcombination of word reference particles. The techniques going before this used K-SVD computation for creating word reference and MP(Matching Pursuit) estimation to deal with the l0 minimization issue. The typical push toward the computations is all that adictionaryisfirstconstructedforpredefinedfingerprintimagepatchesandforanewgivenfingerprintimageitspatchesare tended to by the word reference by figuring l0 minimization. The resulting stage is to quantize and encode therepresentation.ThistechniqueusesanmethodcalledODL(OnlineDictionaryLearning)fordictionaryconstructionandOMP (Orthogonal Matching Pursuit) computation to deal with the l0 minimization issue. The outcomes of this methodology areprovided in this paper and differentiated and the delayed consequences of bunch and stochastic slant strategy to choose the efficientcompressionmethod. Keywords: SparseRepresentation,K-SVD,MatchingPursuit,OnlineDictionaryLearning,OrthogonalMatchingPursuit.