Offline Signature Verification System Using CNN

CBCapsNet: A novel writer-independent offline signature verification model using a CNN-based architecture and capsule neural networks

Biometric technologies, such as handwritten signature verification, are extremely useful for identifying individuals inside an organization or finance department. The improvement of picture categorization using deep learning (DL) neural networks has offered an opportunity to exhibit computer vision in contemporary research applications by applying image processing approaches. Manual signature verification is inefficient, error-prone, time-consuming, and inconvenient; therefore, it is critical to create an automatic signature verification recognition system. This research offers an automatic recognition method based on DL that makes use of the Grupo de Procesado Digital de Seales. The biggest publicly accessible handwritten signature dataset, the synthetic signature dataset, was used to classify the signatures of 100 people, each of whom possessed 24 genuine signatures and 30 forged signatures. An inception V3 transfer learning (TL) model is proposed by hyper-tuning different layers from the middle of its architecture and this model is fine-tuned by adding layers, such as flatten, dense (1024), dropout (0.5), and dense (1). The suggested model was tested against six well-known pre-trained TL convolutional neural network models: VGG 16, VGG 19, ResNet 50, ResNet 101, MobileNet, and EfficientNet. The suggested model surpasses the pre-trained models. Precision, sensitivity, and F1-score are likewise outperformed by the model, with the values of 88%, 88%, and 87%, respectively. The accuracy of the pre-trained models was evaluated as 80%, 81%, 77%, 73%, 71%, and 74%, respectively. The suggested fine-tuned inception V3 gives the highest accurate classifications, distinguishing between genuine and forged signatures with an accuracy of 88%. This study will aid researchers in developing more effective CNN-based models for offline signature verification with application to computer vision.

Smart phones are advance generation of mobile phones. They enable us to access a large variety of services like data storages, voice communication, wireless connectivity etc. As the number of services increased the number of vulnerabilities and attacks has been increasing as well. There has been a corresponding rise of security solution proposed by researchers for authentication in smart phones like password, face and voice recognition, secret path and signature verification. The most popular authentication mechanism is online biometric signature verification. People adopt this mechanism due to its nature which is most fashionable, secure, trustable and difficult for unauthorized persons to breach privacy. With this work, we aim to provide a structured and broad over view of the research on signature verification process for authentication in smart phones. This paper surveys on signature verification process in smart phones, by focusing on different techniques used in signature verification process. We grouped existing approaches aimed to analyze performance and accuracy rate of different approaches adopted by signature verification process in smart phones. With this categorization, we aim to provide an easy and concise view of different approaches adopted by signature verification process in smart phones.

Signatures are commonly used as a unique way of identifying and verifying a person’s identity. Legal documents like bank cheques and transactions, government documents, and affidavits require signature-based verification to identify the authenticity of the person. The signature identification and verification system help to distinguish whether an input signature is honest or a fake. For very long years, this was considered to be a challenging mission when the offline mode was taken into consideration. The offline mode uses scanned signature images, where the online mode of signing process isn't existing. To overcome the signature verification challenge, an application of deep learning techniques to identify feature representations from signature images can be leveraged. In this paper, we have demonstrated how to identify valid and invalid signatures by building a signature verification and identification system that uses deep convolution neural network (ConvNet).KeywordsDeep learningConvolution neural network (ConvNet)Signature verification

Signature verification task needs relevant signature representations to achieve low error rates. Many signature representations have been proposed so far. In this work we propose a hybrid user-independent/dependent offline signature verification technique with a two-channel convolutional neural network (CNN) both for verification and feature extraction. Signature pairs are input to the CNN as two channels of one image, where the first channel always represents a reference signature and the second channel represents a query signature. We decrease the image size through the network by keeping the convolution stride parameter large enough. Global average pooling is applied to decrease the dimensionality to 200 at the end of locally connected layers. We utilize the CNN as a feature extractor and report 4.13% equal error rate (EER) considering 12 reference signatures with the proposed 200-dimensional representation, compared to 3.66% of a recently proposed technique with 2048-dimensional representation using the same experimental protocol. When the two methods are combined at score level, more than 50% improvement (1.76% EER) is achieved demonstrating the complementarity of them. Sensitivity of the model to gray-level and binary images is investigated in detail. One model is trained using gray-level images and the other is trained using binary images. It is shown that the availability of gray-level information in train and test data decreases the EER e.g. from 11.86% to 4.13%. information about the claimed identity and they can usually be detected with very small error rates. Our focus in this work is offline signature verification where the query is either genuine signature or skilled forgery. Format of the signature image can be binary, gray-level or color. To the best of our knowledge, no public color signature database is available. However the distinction between binary and gray-level image is critical, as a gray-level image carries more information than the binary one such as the pressure of the stroke. Performance of signature verification systems is generally measured by equal error rate (EER) which is the error rate when the false accept (FA) and false reject (FR) rates are equal. If the EER is not reported, distinguishing error rate (DER) can be calculated which is simply the average of FA and FR.

Offline signature verification is undeniably a prominent aspect of the biometric research. It has many applications, including banking and forensics. Signature verification task involves a comparison of questioned signature with a set of one or more reference signatures. The questioned signature may be genuine (written by the authentic writer), a forgery (written by a different person) or a disguise (written by the authentic writer with some modifications with the intent of a later denial). Signature verification generally encompasses the two main steps: feature extraction and classification. The system performance is primarily dependent on the feature extraction step because the characterising features distinguish between genuine, disguised and forged signatures. In this study, we propose several geometric and colour features to characterise the signatures. The features are combined using random forests, logistic regression and generalised linear models. The results are reported on the datasets of several competitions, including ICDAR 2009, ICFHR 2010, ICDAR 2011 and ICFHR 2012 signature verification competitions. The proposed method generally outperforms the other participating methods.

To reduce fraud in financial transactions, signature verification is important for security purposes. In this paper, an attempt has been made to analysis the performance of off-line handwritten signature verification using image-based features. Photocopies and scanned documents are considered as the best possible evidence in the situations when the original documents are either lost or damaged. Although the photocopies are the filtered images of original information and do not reproduce details as in the original documents. In this paper, combinations of four features, i.e., Average object area, mean, Euler number and area of signature image is used to verify the signature. Publically available database BHsig260 is used. In this database, two types of signature are available, i.e.,Bengali and Hindi. Proposed work shows that accuracy of Hindi off-line signature verification is 78.5% with sample size of 15 and accuracy of Bengali off-line signature verification is 69.1 with sample size of 20.

This paper presents the results of the ICDAR 2015 competition on signature verification and writer identification for on- and off-line skilled forgeries jointly organized by PR-researchers and Forensic Handwriting Examiners (FHEs). The aim is to bridge the gap between recent technological developments and forensic casework. Two modalities (signatures and handwritten text) are considered and training and evaluation data are collected and provided by FHEs and PR-researchers. Four tasks are defined for four different languages; Bengali off-line signature verification, Italian off-line signature verification, German on-line signature verification, and English handwritten text based writer identification. In total, 40 systems have participated in this competition. The participants of the signatures modality were motivated to report their results in Likelihood Ratios (LRs). This has made the systems even more interesting for application in forensic casework. For evaluating the performance of the systems, we have used the forensically substantial Cost of Log Likelihood Ratios (Ĉ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">llr</sub> ) in the case of signatures, and the F-measure in the case of handwritten text.

Signature verification and Identification has great importance in authentication systems. Persian signatures are different from other signature types because people usually do not use text in it and they draw a shape as their signature. In this paper, a method for off-line Persian signature identification and verification is proposed that is based on DWT (Discrete Wavelet Transform) and Image Fusion. To extract features, at first DWT is used to access details of signature; then several signatures instances of each person are fused together to generate reference pattern of person's signature. In the classification phase, Euclidean distance between the test image and each pattern is used in each sub band. Experimental results confirmed the effectiveness of proposed method.

This paper addresses issues concerning biometric authentication based on handwritten signatures. Our research aimed to check whether a handwritten signature acquired with a mobile device can effectively verify a user's identity. We present a novel online signature verification method using coordinates of points and pressure values at each point collected with a mobile device. Convolutional neural networks are used for signature verification. In this paper, three neural network models are investigated, i.e., two self-made light SigNet and SigNetExt models and the VGG-16 model commonly used in image processing. The convolutional neural networks aim to determine whether the acquired signature sample matches the class declared by the signer. Thus, the scenario of closed set verification is performed. The effectiveness of our method was tested on signatures acquired with mobile phones. We used the subset of the multimodal database, MobiBits, that was captured using a custom-made application and consists of samples acquired from 53 people of diverse ages. The experimental results on accurate data demonstrate that developed architectures of deep neural networks can be successfully used for online handwritten signature verification. We achieved an equal error rate (EER) of 0.63% for random forgeries and 6.66% for skilled forgeries.

In general, signatures are used to authenticate an individual's authenticity. A solid way of certifying the legitimacy of a signature is still waiting for. The approach offered in this paper will allow people to identify signatures to determine whether a signature is forged or genuine. We have tried to automate the signature verification procedure using Convolutionary Neural Networks in our system. In computer applications such as picture registration and identification of objects, image classification and recovery, feature detectors and descriptors have a key role to play. This article discusses the performance analysis of numerous characteristic and descriptor detectors like SIFT, SURF, ORB, BRIEF, BRISK, FREAK. The number of matching points in areas overlapping between images and subjective stitching accuracy were analysed in terms of the number of features. The removal of varied symmetrical characteristics of the image enhances the possibility that different scene views can reliably satisfy the requirements. One experiment result show that the maximum number of detected key points is AGAST, FAST and BRISK detection, but the lesser number of extracted key points is STAR, AKAZE and MSER. In addition, each algorithm's speed is recorded.

The signature verification is one of the major and general purposes frequently used approach for person’s verification among all the other existing and known biometric-based verification methods. This brought the attention to the development of an automatic signature verification system. In this paper, an off-line signature verification and recognition system based on tree and grid by adopting a feature extraction novel approach such as pixels in tree, eccentricity, and center was used. The problem of using a trained dataset in order to perform the verification was overcome by using only one genuine and test signature at the run time. Decision of the result which was based on authenticity and governed by the maximum correct feature favor, acceptance, and rejection is based on its majority. The usefulness of the proposed approach was acknowledged by the use of experimental results.

Although handwritten signature verification has been extensively researched, it has not achieved an optimal classification accuracy rate. Therefore, efficient and accurate signature verification techniques are required since signatures are still widely used as a means of personal verification. This research work presents efficient distance-based classification techniques as an alternative to supervised learning classification techniques (SLTs). The Local Directional Pattern (LDP) feature extraction technique was used to analyze the effect of using several different distance-based classification techniques. The classification techniques tested, are the Euclidean, Manhattan, Fractional, weighted Euclidean, weighted Manhattan, weighted fractional distances and individually optimized resampling of feature vector sizes. The best accuracy, of 90.8%, was achieved through applying a combination of the weighted fractional distances and locally optimized resampling classification techniques to the Local Directional Pattern feature extraction. These results are compared with results from literature, where the same feature extraction technique was classified with SLTs. The distance-based classification was found to produce greater accuracy than the SLTs.

Online and Offline Signature Verification: A Combined Approach

This paper presents an experimental comparison of different hybrid snake-based algorithms for automatic offline signature verification. Offline signature verification aims to establish the degree of similarity between one given test signature and one reference signature. Our work aims to study how effective introducing fuzziness to tackle this image analysis problem. We developed several hybrid (crisp and fuzzy) snake algorithms adapted to the practical requirements of automatic signature verification. Our work compares different hybrid snake approaches for this verification problem.