Online Signature Verification Research Articles

Online handwritten signature verification via network analysis

In this paper, a model for online handwritten signature verification via networks analysis is proposed, in which centrality metrics are used as predictive characteristics. Supervised learning methods were used to classify the veracity of the signatures. The study uses the MCYT-100 and SVC2004 task2 signature databases, which are widely used in signature verification problems. In the framework, signatures were considered as nodes of a complex network in which the existing connection relationship between two signatures is measured through Pearson correlations between time series of signature coordinates that exceed a threshold of comparison. The attributes for the model were topological measures of centralities of the networks and the entropy of the time series signature. Some metrics are used to measure classification errors, including the False Rejection Rate (FRR) which identifies the percentage of true signatures that are rejected by the model, the False Acceptance Rate (FAR) as the percentage of false signatures that are accepted by the learning engine and the Average Error Rate (AER) as the average error. In this framework, we obtained a false positive rate of 6.19%, a false negative rate of 6.39% and an average error of approximately 6.28%. This result is close to ones obtained by state-of-the-art models for this problem, showing that the use of complex networks for online handwritten signature verification is a promising venue of research.

Writer-independent online signature verification based on 2D representation of time series data using triplet supervised network

Online signature verification (OSV) as a critical link of the paperless office still has tremendous challenges. In this paper, we propose a novel Writer-Independent (WI) OSV framework that includes three parts. (1) The two-dimensional (2D) representation method is designed to transform the original time-series signature data into stroke images with dynamic-static hybrid information. (2) The Channel-wise Weight Learning (CWL) mechanism is integrated into the feature extractor to mine the potential relationship between three dynamic attributes (altitude, azimuth, and pressure). (3) A new Triplet Supervised Network (TSN) that contains three weight-shared convolution neural network (CNN) streams was provided to measure the distance of [Anchor, Positive, Test]. The experimental results show that our model has at least 1.29% accuracy improvement when confronted with skilled forgery signatures than that of other classical CNNs and 11.43% than lightweight models. Moreover, the TSN model is superior to the previous OSV algorithms in the WI pattern.

Performance Analysis of Deep Learning Approaches for Offline Signature Verification

Signature verification plays a significant role in many biometric authentication places. Many financial institutes require a robust signature verification process for check clearance, loan sanction, processing, pension relation documents, etc. Expert forgeries make it hard to authenticate an individual's identification based on signatures. Typically, this occurs when the forger understands the user's intricate features of the signature and strives to mimic it. Online signature verification approaches can extract various features such as keystrokes, pressure of the pointer, duration between the strokes and the lettering styles, so that verification becomes effective. However, the lack of these intricate details in offline signature, the authentication process becomes much more difficult. To address these issues, in this paper we propose deep learning-based approaches for offline signature verification. In this regard, we have used ZFNet, LeNet and AlexNet architectures with CEDAR, BHSig20 and UTsig datasets for our extensive. experimentation. We propose a learning model in which the dataset consists of multiple genuine and forged signatures. Further, performance analysis of these techniques has been carried out. It was found that LeNet has provided better training and testing accuracy with above 82% performance.

Multi-scale residual based siamese neural network for writer-independent online signature verification

Multi-scale residual based siamese neural network for writer-independent online signature verification

COMPOSV: compound feature extraction and depthwise separable convolution-based online signature verification

COMPOSV: compound feature extraction and depthwise separable convolution-based online signature verification

SVC-onGoing: Signature verification competition

This article presents SVC-onGoing11https://competitions.codalab.org/competitions/27295., an on-going competition for on-line signature verification where researchers can easily benchmark their systems against the state of the art in an open common platform using large-scale public databases, such as DeepSignDB22https://github.com/BiDAlab/DeepSignDB. and SVC2021_EvalDB33https://github.com/BiDAlab/SVC2021_EvalDB., and standard experimental protocols. SVC-onGoing is based on the ICDAR 2021 Competition on On-Line Signature Verification (SVC 2021), which has been extended to allow participants anytime. The goal of SVC-onGoing is to evaluate the limits of on-line signature verification systems on popular scenarios (office/mobile) and writing inputs (stylus/finger) through large-scale public databases. Three different tasks are considered in the competition, simulating realistic scenarios as both random and skilled forgeries are simultaneously considered on each task. The results obtained in SVC-onGoing prove the high potential of deep learning methods in comparison with traditional methods. In particular, the best signature verification system has obtained Equal Error Rate (EER) values of 3.33% (Task 1), 7.41% (Task 2), and 6.04% (Task 3). Future studies in the field should be oriented to improve the performance of signature verification systems on the challenging mobile scenarios of SVC-onGoing in which several mobile devices and the finger are used during the signature acquisition.

Online Signature Verification Using Locally Weighted Dynamic Time Warping via Multiple Fusion Strategies

Online signature verification is the process of using a dynamic signature verification system to confirm the writer’s identity. It can be used as a security system to confirm entrance applications and password substitutes, as well as a forensic tool to assist an expert’s investigation. This study proposes a novel online signature verification system based on a single-template strategy to improve performance in real-world scenarios. It employs discriminative mean signature template sets as well as fusion strategies of multiple local weighting and warping schemes, for dynamic time warping (DTW). The first step is to generate a set of user-specific mean signature templates for each feature using a recent time-series averaging method, namely, Euclidean barycenter-based DTW barycenter averaging. Then, using multiple and direct matching points between the mean signature templates and references for dependent and independent DTW, we obtain a local weighting estimate considering local stability sequences. Furthermore, we develop fusion strategies for calculating locally weighted DTW sets and concatenating them as a feature vector for each warping, followed by the construction of a support vector machine (SVM) classifier. Finally, in the verification phase, we use the single-template technique to compute a discriminative fused score using SVMs between the mean signature template sets and a query sample. The effectiveness of the proposed method is demonstrated by extensive experimental results obtained using three public online signature datasets: SVC2004 Task1/Task2 and MCYT-100.

Online Signature Verification Systems on a Low-Cost FPGA

This paper describes three different approaches for the implementation of an online signature verification system on a low-cost FPGA. The system is based on an algorithm, which operates on real numbers using the double-precision floating-point IEEE 754 format. The double-precision computations are replaced by simpler formats, without affecting the biometrics performance, in order to permit efficient implementations on low-cost FPGA families. The first approach is an embedded system based on MicroBlaze, a 32-bit soft-core microprocessor designed for Xilinx FPGAs, which can be configured by including a single-precision floating-point unit (FPU). The second implementation attaches a hardware accelerator to the embedded system to reduce the execution time on floating-point vectors. The last approach is a custom computing system, which is built from a large set of arithmetic circuits that replace the floating-point data with a more efficient representation based on fixed-point format. The latter system provides a very high runtime acceleration factor at the expense of using a large number of FPGA resources, a complex development cycle and no flexibility since it cannot be adapted to other biometric algorithms. By contrast, the first system provides just the opposite features, while the second approach is a mixed solution between both of them. The experimental results show that both the hardware accelerator and the custom computing system reduce the execution time by a factor ×7.6 and ×201 but increase the logic FPGA resources by a factor ×2.3 and ×5.2, respectively, in comparison with the MicroBlaze embedded system.

Optimized jk-nearest neighbor based online signature verification and evaluation of the main parameters

In this paper, we propose an enhanced jk-nearest neighbor (jk-NN) classifier for online signature verification. After studying the algorithm's main parameters, we use four separate databases to present and evaluate each algorithm parameter. The results show that the proposed method can increase the verification accuracy by 0.73-10% compared to a traditional one class k-NN classifier. The algorithm has achieved reasonable accuracy for different databases, a 3.93% error rate when using the SVC2004 database, 2.6% for MCYT-100 database, 1.75% for the SigComp'11 database, and 6% for the SigComp'15 database.The proposed algorithm uses specifically chosen parameters and a procedure to pick the optimal value for K using only the signer's reference signatures, to build a practical verification system for real-life scenarios where only these signatures are available. By applying the proposed algorithm, the average error achieved was 8% for SVC2004, 3.26% for MCYT-100, 13% for SigComp'15, and 2.22% for SigComp'11.

Signature barcodes for online verification

As a sub-branch of behavioral biometrics, online signature verification systems deal with unique signing characteristics, which could be better differentiated by extraction of habitual singing styles instead of geometric features in case of perfect forgery. Even if the signatures are geometrically identical, speed and frequency components of the signing process might significantly vary. Therefore, a novel framework is introduced as a new signature verification protocol for touchscreen devices using barcodes containing the dominant frequency component of the speed signals. A special interface is designed as signature tracker to extract the displacement data sampled from the signing process. The speed signals are interpolated from the displacement data and the frequency components of the signals are computed by scalograms analysis governed by continuous wavelet transformations (CWT). The signature barcodes are generated as 4-scale scalograms and classified by support vector machines (SVM). Among several compatible wavelets, Gaussian derivative wavelet is selected for generating scalograms and the results of the process are calculated as 2.25% FAR, 2.75% FRR and 2.81%EER for our dataset. The framework is also tested with SVC2004 data that we achieved 0% FAR, 9.33% FRR and 8%EER, also with SUSIG-Visual, SUSIG-Blind, MOBISIG databases and we reached between 1.22%-3.62% average EERs, which are competitive among the relevant results. Given the promising outcomes, the signature barcoding is very reliable method which could be executed by a simple touchscreen interface collecting the barcodes for storing and benchmarking when needed.

Smartphone-based Handwritten Signature Verification using Acoustic Signals

Handwritten signature verification techniques, which can facilitate user authentication and enable secure information exchange, are still important in property safety. However, on-line automatic handwritten signature verification usually requires dynamic handwritten patterns captured by a special device, such as a sensor-instrumented pen, a tablet or a smartwatch on the dominant hand. This paper presents SonarSign, an on-line handwritten signature verification system based on inaudible acoustic signals. The key insight is to use acoustic signals to capture the dynamic handwritten signature patterns for verification. Particularly, SonarSign exploits the built-in speakers and microphones of smartphones to transmit a specially designed training sequence and record the corresponding echo for channel impulse response (CIR) estimation, respectively. Based on the sensitivity of CIR to the tiny surrounding environment changes including handwritten signature actions, SonarSign designs an attentional multi-modal Siamese network for end-to-end signatures verification. First, multi-modal CIR streams are fused to extract representative signature pattern features from spatio-temporal dimensions. Then an attentional Siamese network is elaborated to verify whether the given two signatures are from the same signatory. Extensive experiments in real-world scenarios show that SonarSign can achieve accurate and robust signatures verification with an AUC (Area Under ROC (Receiver Operating Characteristic) Curve) of 98.02% and an EER (Equal Error Rate) of 5.79% for unseen users.

Online Signature Verification Using Fully Connected Deep Neural Networks

Online Signature Verification Using Fully Connected Deep Neural Networks

Online signature verification using signature down-sampling and signer-dependent sampling frequency

AbstractOnline signature verification considers signatures as time sequences of different measurements of the signing instrument. These signals are captured on digital devices and therefore consist of a discrete number of samples. To enrich or simplify this information, several verifiers employ resampling and interpolation as a preprocessing step to improve their results; however, their design decisions may be difficult to generalize. This study investigates the direct effect of the sampling rate of the input signals on the accuracy of online signature verification systems without using interpolation techniques and proposes a novel online signature verification system based on a signer-dependent sampling frequency. Twenty verifier configurations were created for five different public signature databases and a variety of popular preprocessing approaches and evaluated for 20–40 different sampling rates. Our results show that there is an optimal range for the sampling frequency and the number of sample points that minimizes the error rate of a verifier. A sampling frequency range of 15–50 Hz and a signature point count of 60–240 provided the best accuracies in our experiments. As expected, lower ranges showed inaccurate results; interestingly, however, higher frequencies often decreased the verification accuracy. The results show that one can achieve better or at least the same verification accuracies faster by down-sampling the online signatures before further processing. The proposed system achieved competitive results to state-of-the-art systems for different databases by using the optimal sampling frequency. We also studied the effect of choosing individual sampling frequencies for each signer and proposed a signature verification system based on signer-dependent sampling frequency. The proposed system was tested using 500 different verification methods and improved the accuracy in 92% of the test cases compared to the usage of the original frequency.

SIGNATURE VERIFICATION USING SIAMESE NEURAL NETWORK ONE-SHOT LEARNING

With the acceleration of digitalization in all areas of our lives, the need for biometric verification methods is increasing. The fact that biometric data is unique and biometric verification is stronger against phishing attacks compared to password-based authentication methods, has increased its preference rate. Signature verification, which is one of the biometric verification types, plays an important role in many areas such as banking systems, administrative and judicial applications. There are 2 types of signature verification, online and offline, for identifying the identity of the person and detecting signature forgery. Online signature verification is carried out during signing and temporal dynamic data are available regarding the person's signature. Offline verification is applied by scanning the image after signing, and this verification is limited to spatial data. Therefore, the offline signature verification process is considered a more challenging task. In this study, offline signature verification, independent of the writer, based on One-Shot Learning, was performed using Siamese Neural Network. Due to the fact that the Deep Convolution Neural Network requires a large amount of labeled data for image classification, real and fake signature distinction has been achieved by using the One-Shot Learning method, which can perform a successful classification by using less numbers of signature images. As a result of the experiments conducted on signature datasets, using the Siamese architecture, the proposed approach achieved percentage accuracy of 93.23, 92.11, 89.78, 91.35 verification in 4NSigComp2012, SigComp2011, 4NSigComp2010 and BHsig260 respectively.

Research on online signature verification based on anomaly detection

With the advancement of modern information technology, personal information security has become the focus of social attention. As a biological behavior feature, online signatures have been trained for a long time, have very personal characteristics, and are collected as a real-time sequence. In view of this characteristic, we can know that it is difficult for forgers to imitate the intrinsic and essential characteristics of the signer. In addition, a signature is a rapid writing process driven by the central nervous system and personal writing habits. Even the signer himself may write an unstable and similar forgery signature, so it is necessary to analyze the variability and stability of signatures. Therefore, this paper proposes a method of using the variance and characteristics of local features to find a stable segment of the user’s signature for authentication. In addition, we also adopt a two-stage authentication strategy that combines the above method with the traditional anomaly detection algorithm. An experiment was conducted on the open-access online signature database MCYT, and an EER of 2.63% was obtained. The experiment proved the effectiveness and robustness of the method.

A new wrapper feature selection method for language-invariant offline signature verification

Among various biometric systems, an offline signature verification system has been widely used in all fields such as in banks, educational institutes, legal procedures and, criminal investigation where authentication and verification are utmost required. Despite the popularity of the online signature verification system, its offline counterpart still has great importance in developing countries, especially in rural areas, where easy availability of smart devices along with fast internet connection is not available. In this work, we have developed a language invariant offline signature verification model which is almost equally applicable for both writer dependent and writer independent scenarios. At first, an offline signature is collected as an image, following which a corresponding signal is generated using singular value decomposition. Then four different kinds of features namely, statistical, shape-based, similarity-based, and frequency-based are extracted from the transformed signal of the signature image. Next, to reduce the feature dimension, we have designed a novel wrapper feature selection method based on Red Deer Algorithm, a recently proposed meta-heuristic method, to keep only the relevant features to be used during signature authentication and verification process. Finally, a confidence score from the Naïve Bayes classifier has been used to perform the authentication and verification process. Our model has been evaluated on CEDAR (English), UTSig (Persian), Sigcomp 2011 Dutch, Sigcomp 2011 Chinese, and SigWIcomp 2015 Bengali signature datasets. Obtained results confirm that the proposed model can outperform many of its predecessors.

DeepWriteSYN: On-Line Handwriting Synthesis via Deep Short-Term Representations

This study proposes DeepWriteSYN, a novel on-line handwriting synthesis approach via deep short-term representations. It comprises two modules: i) an optional and interchangeable temporal segmentation, which divides the handwriting into short-time segments consisting of individual or multiple concatenated strokes; and ii) the on-line synthesis of those short-time handwriting segments, which is based on a sequence-to-sequence Variational Autoencoder (VAE). The main advantages of the proposed approach are that the synthesis is carried out in short-time segments (that can run from a character fraction to full characters) and that the VAE can be trained on a configurable handwriting dataset. These two properties give a lot of flexibility to our synthesiser, e.g., as shown in our experiments, DeepWriteSYN can generate realistic handwriting variations of a given handwritten structure corresponding to the natural variation within a given population or a given subject. These two cases are developed experimentally for individual digits and handwriting signatures, respectively, achieving in both cases remarkable results. Also, we provide experimental results for the task of on-line signature verification showing the high potential of DeepWriteSYN to improve significantly one-shot learning scenarios. To the best of our knowledge, this is the first synthesis approach capable of generating realistic on-line handwriting in the short term (including handwritten signatures) via deep learning. This can be very useful as a module toward long-term realistic handwriting generation either completely synthetic or as natural variation of given handwriting samples.

Online signature verification based on dynamic features from gene expression programming

Gene Expression Programming (GEP) is a powerful evolutionary algorithm with simple, linear and compact chromosomes, which has been applied in many fields to solve a large variety of complex problems such as logistic regression, function finding and time series prediction. Since online signature data are composed of discrete points, it is difficult to represent by functional forms, resulting in a limited amount of information used in calculating feature values. Hausdorff distance is utilized as a similarity measure to compute the maximum distance between two point sets, which reduces computational complexity compared with other distance measures. The main contributions of this work are: (1) In preprocessing stage, GEP is used to make signature curve continuous and control each parameter to obtain a fitting curve. Curve fitting is to find a suitable function that is the best fitting for a given set of data; (2) In feature extraction stage, curvature and torsion are utilized to construct eight feature sets for characterizing each user’s signatures, and then Hausdorff distance is proposed to calculate the distances between feature sets of two signatures to form an eight-dimensional feature vector; (3) In verification stage, combined with Feed-Forward BP Neural Network classifier, distance matrices consisting of feature vectors are trained and tested many times. The best performances can be provided with false rejection rate, false acceptance rate, average error rate and standard deviation. The experimental results implemented on three available online signature databases MCYT-100, SVC2004 and SUSIG indicate the effectiveness and robustness of our proposed method.

A Review - Signature Verification System Using Deep Learning: A Challenging Problem

One of the challenging and effective way of identifying person through biometric techniques is Signature verification as compared to traditional handcrafted system, where a forger has access and also attempt to imitate it which is used in commercial scenarios, like bank check payment, business organizations, educational institutions, government sectors, health care industry etc. so the signature verification process is used for human examination of a single known sample. There are mainly two types of signature verification: static and dynamic. i) Static or off-line verification is the process of verifying an electronic or document signature after it has been made, ii) Dynamic or on-line verification takes place as a person creates his/her signature on a digital tablet or a similar device. As compared, Offline signature verification is not efficient and slow for a large number of documents. Therefore although vast and extensive research on signature verification there is need to more focus and review on the online signature verification method to increase efficiency using deep learning.

A two-tier ensemble approach for writer dependent online signature verification

Biometric verification systems are used to recognize people based on their uniqueness or characteristics. Signature is considered as one of the most commonly used biometric that individualizes a human being. It is generally used to keep individual’s privacy in many places such as banking sectors, academic institutes, office premises and trading. But increase of criminal attempts in falsifying an individual’s signature, known as signature forgeries, motivates the researchers to develop computerized systems that can verify the genuineness of a questioned signature. Though many researches have been performed till date, but the issue of identifying skilled forgeries still remains a major concern for the researchers. To this end, in this work, we have designed a two-tier ensemble based writer dependent and language- invariant online signature verification system. In doing so, we have first extracted three different categories of features from each input signature: physical, frequency based and statistical, and then designed a feature-classifier based ensemble (i.e., Ensemble#1) using seven different classifiers. The predictions obtained from the seven classifiers are combined using normalised distribution summation strategy. Decisions obtained from Ensemble#1 are then fed to Ensemble#2, where a majority voting based approach is followed, to identify the input signature as genuine or forged. Our system is evaluated on two standard datasets: SVC 2004 (Task-II) and MCYT-100 in a writer dependent way. The equal error rate (ERR) and accuracy on SVC 2004 dataset are 2.20 and 98.43% respectively, and on MCYT-100 dataset these are 2.84 and 97.87% respectively. The GAR@0.01FAR value obtained for the SVC-2004 dataset is 94.50% while it is 92.90% for MCYT-100 dataset. We have also compared our results with some state-of-the-art methods, and it has been found that our method performs better than most of these methods. The code of this work is available at: https://github.com/prat1999/Online_Signature_Verification .