Signature recognition

About: Signature recognition is a research topic. Over the lifetime, 2138 publications have been published within this topic receiving 37605 citations.

A Linear-complexity Multi-biometric Forensic Document Analysis System, by Fusing the Stylome and Signature Modalities.

Abstract: Forensic Document Analysis (FDA) addresses the problem of finding the authorship of a given document. Identification of the document writer via a number of its modalities (e.g. handwriting, signature, linguistic writing style (i.e. stylome), etc.) has been studied in the FDA state-of-the-art. But, no research is conducted on the fusion of stylome and signature modalities. In this paper, we propose such a bimodal FDA system (which has vast applications in judicial, police-related, and historical documents analysis) with a focus on time-complexity. The proposed bimodal system can be trained and tested with linear time complexity. For this purpose, we first revisit Multinomial Na\"ive Bayes (MNB), as the best state-of-the-art linear-complexity authorship attribution system and, then, prove its superior accuracy to the well-known linear-complexity classifiers in the state-of-the-art. Then, we propose a fuzzy version of MNB for being fused with a state-of-the-art well-known linear-complexity fuzzy signature recognition system. For the evaluation purposes, we construct a chimeric dataset, composed of signatures and textual contents of different letters. Despite its linear-complexity, the proposed multi-biometric system is proven to meaningfully improve its state-of-the-art unimodal counterparts, regarding the accuracy, F-Score, Detection Error Trade-off (DET), Cumulative Match Characteristics (CMC), and Match Score Histograms (MSH) evaluation metrics.

Research of Online Signature Recognition Based on Energy Feature

Abstract: The paper proposes an on-line signature recognition algorithm with signature energy as feature. The signature energy features at sharp trajectory change points are extracted by means of Daubechies wavelet decomposition of signature signal. Then, 15 points with most dominant energies are chosen. Finally, a new algorithm of classification is put forward, after dynamic time warping matching, with computation amount reduced greatly. Experiment findings show that false acceptance rate is 8 percent while false rejection rate is 0 percent.

Online Signature Verification Using Vector Quantization and Hidden Markov Model

Abstract: In this paper an on-line signature verification system, using vector quantization and Hidden Markov Model (VQ-HMM) is presented. After the signature acquisition, a Chebichef filter is used for noise reduction, and size and phase normalization is performed using Fourier transform. Each signature is segmented and mean velocity, acceleration and pressure are used as extracted features. K-means clustering is used for generation a codebook and VQ generates a code word for each signature. These code words are used as observation vectors in training and recognition phase. HMM models are trained using Baum Welch algorithm. In the verification phase, the forward algorithm is used. The Threshold used in the verification phase is a function of the minimum probability in training phase. Equal Error Rate obtained from this system is 14%.

A P2P Protocol Identification Algorithm Based on PSO-SVM

Abstract: The growing of P2P applications enriched the resources sharing by network, but it also brought many problems such as occupying network bandwidth, security of personal information. Therefore, the monitor of P2P applications is very important, and P2P protocol identification is the key point. So far since the P2P appeared, new P2P applications and data encryption made the traditional port-based and application layer protocol signature recognition useless. To overcome the shortcomings of current methods, a new P2P identification algorithm is proposed which uses DFI (Deep Flow Inspection) by extracting the various properties of P2P data stream characteristics, and then the data flows are divided into TCP and UDP data sets, finally the support vector machine optimized by particle swarm optimization is used to assort the network data streams, Experimental results of the P2P and non-P2P applications show that this algorithm can identify the HTTP applications 100% and for a variety of P2P applications, the accuracy of classification can also reach to 95%.

An Integrated Approach for Automated Authentication Based On Multimodal Biometric System

Abstract: Biometrics refers to the automatic identification of an individual based on his/her physiological or behavioral traits. Unimodal biometric systems perform person recognition based on a single source of biometric information and are affected by problems like noisy sensor data, non-universality and lack of individuality of the chosen biometric trait. Some of the limitations imposed by unimodal biometric systems (that is, biometric systems that rely on the evidence of a single biometric trait) can be overcome by using multiple biometric modalities. Such systems, known as Multimodal biometric systems, are expected to be more reliable due to the presence of multiple, fairly independent pieces of evidence. A multimodal biometric system integrates information from multiple biometric sources to compensate for the limitations in performance of each individual biometric system. An optimal framework for combining the matching scores from multiple modalities using the likelihood ratio statistic computed using the generalized densities estimated from the genuine and impostor matching scores is being proposed in this paper. The motivation for using generalized densities is that some parts of the score distributions can be discrete in nature; thus, estimating the distribution using continuous densities may be inappropriate. The two approaches for combining evidence based on generalized densities: (i) the product rule, which assumes independence between the individual modalities, and (ii) copula models, which consider the dependence between the matching scores of multiple modalities are being presented in this paper.