A novel video-based system for in-air signature verification. (January 2017)

• Record Type: Journal Article
• Authors: Fang, Yuxun; Kang, Wenxiong; Wu, Qiuxia; Tang, Lei
• Is Part Of: Computers & electrical engineering. Volume 57(2017)
• Page Start: 1
• Page End: 14
• Publication Date: 2017-01
• Subjects: In-air signature -- Signature verification -- Video-based tracking -- Dynamic time warping -- Fast Fourier Transform; Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2016.11.010
• Languages: English
• ISSNs: 0045-7906

Abstract

Highlights: Develop a novel video-base in-air signature verification mode. Propose an improved DTW algorithm based on the characteristic of in-air signature. Demonstrate the validity of FFT algorithm in the verification of continuous track signature. Design a DTW-FFT fusion algorithm based on Gaussian probability distribution. Abstract: At present, the common online signature verification systems adopt the contacted model of signing on a tablet or a phone screen, which have a great dependence on the device. Considering that the development of biometrics tends to the long distance and non-contact mode, a novel video-based system for in-air signature verification is proposed. First, the fingertip tracking is used for generating unique signature trajectory from the in-air signing videos. In combination of contrapuntally improved dynamic time warping (DTW) method, the Fast Fourier Transform (FFT) and the analysis on the signature length, a Gaussian distribution-based fusion algorithm is proposed for in-air signature verification. Finally, comprehensive experiments are conducted on a self-built database consisting of 560 signatures, then a false rejection rate (FRR) of 2.86% and a false acceptance rate (FAR) of 1.90% are achieved with an average matching time of only 24 ms, which have demonstrated the effectiveness of our proposed system. Graphical abstract: