Characteristic individuation assisted handwritten signature verification method and device

Abstract

The invention discloses a handwritten signature verification method and device assisted by personality personalization, and belongs to the field of human-computer interaction. According to the method, for collected user handwritten signature samples, different classifiers are selected for classification by using different features, and due to the fact that information provided by image features, track features and user character features is different, the discrimination precision can be improved by using diversity of different features, and the discrimination efficiency is improved. Meanwhile, the character of the user can be predicted according to the handwritten signature track information of the user, and the signature of the user is personalized. A support vector machine classifier is trained in a supervised manner through data with positive and negative sample labels and character labels, and the verification accuracy can reach 96%.

Description

technical field [0001] The invention belongs to the field of human-computer interaction, and in particular relates to a method and device for verifying handwritten signatures assisted by personality personalization. Background technique [0002] Accurate and fast authentication for each individual is a very important and complex task. The core of this task is to verify the unique identity of the user. Traditional individual user verification methods include token-based verification and knowledge-based verification (Reference: D.Impedovo and G.Pirlo,"Automatic SignatureVerification:The State of the Art,"Systems,Man,and Cybernetics,Part C: Applications and Reviews, IEEE Transactions on, vol. 38, pp. 609-635, 2008.). Token-based authentication includes chips for authorized user identity information, portable password generation devices, etc.; while knowledge-based authentication includes strings or passwords that users need to memorize. However, these verification methods ha...

AI Technical Summary

Problems solved by technology

[0003] However, handwritten signatures may be forged, especially when the forger has seen the name and signature shape of the forged person, it becomes a challenging task to distinguish the real signature from the forged signature (reference: Z. Zhang , K.Wang, and Y.Wang, "A Survey of On-line Signature Verification," in BiometricRecognition.vol.7098, Z.Sun, J.Lai, X.Chen, and T.Tan, Eds., ed: Springer Berlin Heidelberg, 2011, pp.141-149.), while handing over signature verification work to manual processing is inefficient and error-prone, so an efficient signature verification method is necessary

[0004] At present, some methods for signature verification do not comprehensively consider the various characteristic information of the signature, nor analyze the relationship between each characteristic information and the signing user, so they will encounter bottlenecks in the accuracy of verification and user personalization analysis ( References: Z. Zhang, K. Wang, and Y. Wang, "A Survey of On-line Signature Verification," in Biometric Recognition.vol.7098, Z. Sun, J. Lai, X. Chen, and T. Tan , Eds., ed: Springer Berlin Heidelberg, 2011, pp.141-149.)

Images