A Structural Approach to Offline Signature Verification Using Graph Edit Distance

doi:10.1109/ICDAR.2017.201

Home
/
Papers
/
A Structural Approach to Offline Signature Verification Using Graph Edit Distance

Proceedings Article•DOI•

A Structural Approach to Offline Signature Verification Using Graph Edit Distance

Paul Maergner¹, Kaspar Riesen², Rolf Ingold³, Andreas Fischer⁴•Institutions (4)

Nuance Communications¹, University of Applied Sciences and Arts Northwestern Switzerland FHNW², University of Fribourg³, Swiss National Bank⁴

01 Nov 2017-pp 1216-1222

TL;DR: A novel structural approach to offline signature verification using an efficient cubic-time approximation of graph edit distance is introduced and several ways of creating, normalizing, and comparing signature graphs built from keypoints are put forward.

read less

Abstract: Graphs provide a powerful representation formalism for handwritten signatures, capturing local properties as well as their relations. Yet, although introduced early for signature verification, only a few current systems rely on graph-based representations. A possible reason is the high computational complexity involved for matching two general graphs. In this paper, we introduce a novel structural approach to offline signature verification using an efficient cubic-time approximation of graph edit distance. We put forward several ways of creating, normalizing, and comparing signature graphs built from keypoints and investigate their performance on three benchmark datasets. The experiments demonstrate a promising performance of the proposed structural approach when compared with the state of the art.

...read moreread less

Citations

PDF

Open Access

More filters

Journal Article•DOI•

Combining graph edit distance and triplet networks for offline signature verification

[...]

Paul Maergner¹, Vinaychandran Pondenkandath¹, Michele Alberti¹, Marcus Liwicki², Kaspar Riesen³, Rolf Ingold¹, Andreas Fischer¹ - Show less +3 more•Institutions (3)

University of Fribourg¹, Luleå University of Technology², University of Applied Sciences and Arts Northwestern Switzerland FHNW³

01 Jul 2019-Pattern Recognition Letters

TL;DR: Offline signature verification is a challenging pattern recognition task where a writer model is inferred using only a small number of genuine signatures using a combination of complementary writer mode and reader mode.

...read moreread less

36 citations

Journal Article•DOI•

Offline Handwritten Signature Verification Using Cylindrical Shape Context

[...]

Pradeep N. Narwade, Rajendra R. Sawant, S. V. Bonde

11 Oct 2018-3d Research

TL;DR: Experimental results show that desired pixel matching results are obtained by using cylindrical shape context which automatically increases the accuracy of verification of offline handwritten signatures.

...read moreread less

Abstract: Offline handwritten signatures is a convincing evidence form of biometrics for verification. However, the verification of offline handwritten signatures is challenging task because of the variations in handwritten signatures. To address this difficulty, this paper proposes a new approach to represent the shape. In this newly proposed approach, the signature pixels are represented by: (1) Gaussian Weighting Based Tangent Angle, to represent the curve angle at the reference pixel; (2) a new shape descriptor, i.e. cylindrical shape context is proposed for a detailed and accurate description of the curve at corresponding pixel. Experimental results show that desired pixel matching results are obtained by using cylindrical shape context which automatically increases the accuracy of verification of offline handwritten signatures. The shape dissimilarity measures are computed and given to the Support Vector Machine with Radial Basis Function (RBF) kernel for classification of signature. The results obtained using GPDS synthetic signature database, UTSig persian offline signature database, and MCYT-75 offline signature database shows the effectiveness of proposed cylindrical shape context.

...read moreread less

18 citations

Book Chapter•DOI•

Offline Signature Verification by Combining Graph Edit Distance and Triplet Networks

[...]

Paul Maergner¹, Vinaychandran Pondenkandath¹, Michele Alberti¹, Marcus Liwicki¹, Kaspar Riesen², Rolf Ingold¹, Andreas Fischer¹ - Show less +3 more•Institutions (2)

University of Fribourg¹, University of Applied Sciences and Arts Northwestern Switzerland FHNW²

17 Aug 2018

TL;DR: This work proposes to complement a recent structural approach to offline signature verification based on graph edit distance with a statistical approach based on metric learning with deep neural networks, and demonstrates that combining the structural and statistical models leads to significant improvements in performance.

...read moreread less

Abstract: Biometric authentication by means of handwritten signatures is a challenging pattern recognition task, which aims to infer a writer model from only a handful of genuine signatures. In order to make it more difficult for a forger to attack the verification system, a promising strategy is to combine different writer models. In this work, we propose to complement a recent structural approach to offline signature verification based on graph edit distance with a statistical approach based on metric learning with deep neural networks. On the MCYT and GPDS benchmark datasets, we demonstrate that combining the structural and statistical models leads to significant improvements in performance, profiting from their complementary properties.

...read moreread less

17 citations

Cites methods or result from "A Structural Approach to Offline Si..."

...[16] with a statistical model inspired by recent advances in the field of deep learning, namely metric learning by means of a deep CNN [13] with the triplet loss function [14]....
[...]
...For a more detailed description, see [16]....
[...]
...Therefore, each dissimilarity score is normalized using the average dissimilarity score between the reference signatures of the current user as suggested in [16]....
[...]
...Table 2 shows our results using the same protocol compared with the previously published results: results published in [16] and results presented on the GPDS website(7), which have been achieved using the system published in [6]....
[...]
...Graph Parameter Validation For the keypoint graph extraction, we use D = 25, which has been proposed in [16]....
[...]

Proceedings Article•DOI•

Offline Signature Verification Via Structural Methods: Graph Edit Distance and Inkball Models

[...]

Paul Maergner¹, Nicholas R. Howe², Kaspar Riesen³, Rolf Ingold¹, Andreas Fischer⁴ - Show less +1 more•Institutions (4)

University of Fribourg¹, Smith College², University of Applied Sciences and Arts Northwestern Switzerland FHNW³, Swiss National Bank⁴

01 Aug 2018

TL;DR: This paper investigates two recently presented structural methods for handwriting analysis: keypoint graphs with approximate graph edit distance and inkball models and proposes a combined verification system, which demonstrates an excellent performance on the MCYT and GPDS benchmark data sets when compared with the state of the art.

...read moreread less

Abstract: For handwritten signature verification, signature images are typically represented with fixed-sized feature vectors capturing local and global properties of the handwriting. Graph-based representations offer a promising alternative, as they are flexible in size and model the global structure of the handwriting. However, they are only rarely used for signature verification, which may be due to the high computational complexity involved when matching two graphs. In this paper, we take a closer look at two recently presented structural methods for handwriting analysis, for which efficient matching methods are available: keypoint graphs with approximate graph edit distance and inkball models. Inkball models, in particular, have never been used for signature verification before. We investigate both approaches individually and propose a combined verification system, which demonstrates an excellent performance on the MCYT and GPDS benchmark data sets when compared with the state of the art.

...read moreread less

15 citations

Cites background or methods or result from "A Structural Approach to Offline Si..."

...For more details, we refer the reader to [10]....
[...]
...Each verification score (either dGED or dinkball) is divided by the average dissimilarity score of the reference signatures of the current user as suggested in [10]....
[...]
...[10] have introduced a general framework for graph-based signature verification based on the graph edit distance between labeled graphs....
[...]
...For the graph edit distance based dissimilarity, we use the configuration proposed in [10] where they optimized the parameters on GPDS-75....
[...]
...We compare our results against the results published in [10] and the results presented on the GPDS website1....
[...]

Book Chapter•DOI•

A Survey of State of the Art Methods Employed in the Offline Signature Verification Process

[...]

Michael Stauffer¹, Paul Maergner², Andreas Fischer², Kaspar Riesen¹•Institutions (2)

University of Applied Sciences and Arts Northwestern Switzerland FHNW¹, University of Fribourg²

01 Jan 2021

TL;DR: The present chapter reviews the field of offline signature verification and presents a comprehensive overview of methods typically employed in the general process of offline signing verification.

...read moreread less

Abstract: Handwritten signatures are of eminent importance in many business and legal activities around the world. That is, signatures have been used as authentication and verification measure for several centuries. However, the high relevance of signatures is accompanied with a certain risk of misuse. To mitigate this risk, automatic signature verification was proposed. Given a questioned signature, signature verification systems aim to distinguish between genuine and forged signatures. In the last decades, a large number of different signature verification frameworks have been proposed. Basically, these frameworks can be divided into online and offline approaches. In the case of online signature verification, temporal information about the writing process is available, while offline signature verification is limited to spatial information only. Hence, offline signature verification is generally regarded as the more challenging task. The present chapter reviews the field of offline signature verification and presents a comprehensive overview of methods typically employed in the general process of offline signature verification.

...read moreread less

13 citations

1
2
3
4
…

References

PDF

Open Access

More filters

Journal Article•DOI•

Online and off-line handwriting recognition: a comprehensive survey

[...]

Réjean Plamondon¹, Sargur N. Srihari²•Institutions (2)

École Normale Supérieure¹, University at Buffalo²

01 Jan 2000-IEEE Transactions on Pattern Analysis and Machine Intelligence

TL;DR: The nature of handwritten language, how it is transduced into electronic data, and the basic concepts behind written language recognition algorithms are described.

...read moreread less

Abstract: Handwriting has continued to persist as a means of communication and recording information in day-to-day life even with the introduction of new technologies. Given its ubiquity in human transactions, machine recognition of handwriting has practical significance, as in reading handwritten notes in a PDA, in postal addresses on envelopes, in amounts in bank checks, in handwritten fields in forms, etc. This overview describes the nature of handwritten language, how it is transduced into electronic data, and the basic concepts behind written language recognition algorithms. Both the online case (which pertains to the availability of trajectory data during writing) and the off-line case (which pertains to scanned images) are considered. Algorithms for preprocessing, character and word recognition, and performance with practical systems are indicated. Other fields of application, like signature verification, writer authentification, handwriting learning tools are also considered.

...read moreread less

2,653 citations

"A Structural Approach to Offline Si..." refers methods in this paper

...Due to this lack of information, offline signature verification is considered as the more difficult task [2], but it also applies to more use cases....
[...]

Journal Article•DOI•

A fast parallel algorithm for thinning digital patterns

[...]

T. Y. Zhang¹, Ching Y. Suen²•Institutions (2)

Chongqing University¹, Concordia University²

01 Mar 1984-Communications of The ACM

TL;DR: A fast parallel thinning algorithm that consists of two subiterations: one aimed at deleting the south-east boundary points and the north-west corner points while the other one is aimed at deletion thenorth-west boundarypoints and theSouth-east corner points.

...read moreread less

Abstract: A fast parallel thinning algorithm is proposed in this paper It consists of two subiterations: one aimed at deleting the south-east boundary points and the north-west corner points while the other one is aimed at deleting the north-west boundary points and the south-east corner points End points and pixel connectivity are preserved Each pattern is thinned down to a skeleton of unitary thickness Experimental results show that this method is very effective 12 references

...read moreread less

2,243 citations

"A Structural Approach to Offline Si..." refers methods in this paper

...For skeletonization, we use the algorithm proposed in [17]....
[...]

Journal Article•DOI•

Thirty years of graph matching in pattern recognition

[...]

Donatello Conte¹, Pasquale Foggia², Carlo Sansone², Mario Vento¹•Institutions (2)

University of Salerno¹, University of Naples Federico II²

01 May 2004-International Journal of Pattern Recognition and Artificial Intelligence

TL;DR: This paper will try to characterize the role that graphs play within the Pattern Recognition field, and presents two taxonomies that include almost all the graph matching algorithms proposed from the late seventies and describes the different classes of algorithms.

...read moreread less

Abstract: A recent paper posed the question: "Graph Matching: What are we really talking about?". Far from providing a definite answer to that question, in this paper we will try to characterize the role that graphs play within the Pattern Recognition field. To this aim two taxonomies are presented and discussed. The first includes almost all the graph matching algorithms proposed from the late seventies, and describes the different classes of algorithms. The second taxonomy considers the types of common applications of graph-based techniques in the Pattern Recognition and Machine Vision field.

...read moreread less

1,517 citations

"A Structural Approach to Offline Si..." refers background in this paper

...One possible reason for the lack of graph-based signature verification systems might be the high computational complexity of general graph matching procedures [9]....
[...]
...Graph matching has been the topic of numerous studies over the last decades [9], [18]....
[...]

Journal Article•DOI•

Automatic signature verification and writer identification — the state of the art

[...]

Réjean Plamondon¹, Guy Lorette•Institutions (1)

École Polytechnique de Montréal¹

01 Jan 1989-Pattern Recognition

TL;DR: A survey of the literature on automatic signature verification and writer identification by computer, and an overview of achievements in static and dynamic approaches to solving these problems, with a special focus on preprocessing techniques, feature extraction methods, comparison processes and performance evaluation.

...read moreread less

981 citations

"A Structural Approach to Offline Si..." refers background in this paper

...State-of-the-art systems for offline signature verification usually rely either on local information like histogram of oriented gradients (HOG) or local binary patterns (LBP) [3], or they take global information into account, for example using a large number of geometrical features like number of holes, moments, projections, distributions, position of barycenter, number of branches in the skeleton, Fourier descriptors, tortuosities, directions, curvatures and chain codes, and many others [1], [4]....
[...]

Journal Article•DOI•

Automatic Signature Verification: The State of the Art

[...]

Donato Impedovo, Giuseppe Pirlo¹•Institutions (1)

University of Bari¹

01 Sep 2008

TL;DR: This paper presents the state of the art in automatic signature verification and addresses the most valuable results obtained so far and highlights the most profitable directions of research to date.

...read moreread less

Abstract: In recent years, along with the extraordinary diffusion of the Internet and a growing need for personal verification in many daily applications, automatic signature verification is being considered with renewed interest. This paper presents the state of the art in automatic signature verification. It addresses the most valuable results obtained so far and highlights the most profitable directions of research to date. It includes a comprehensive bibliography of more than 300 selected references as an aid for researchers working in the field.

...read moreread less

688 citations

"A Structural Approach to Offline Si..." refers background in this paper

...State-of-the-art systems for offline signature verification usually rely either on local information like histogram of oriented gradients (HOG) or local binary patterns (LBP) [3], or they take global information into account, for example using a large number of geometrical features like number of holes, moments, projections, distributions, position of barycenter, number of branches in the skeleton, Fourier descriptors, tortuosities, directions, curvatures and chain codes, and many others [1], [4]....
[...]
...But current state-of-theart in automatic signature verification still obtains a high level of accuracy, which is on a par with other biometric authentication frameworks [1]....
[...]