Multi-Representational Learning for Offline Signature Verification using Multi-Loss Snapshot Ensemble of CNNs

Offline Signature Verification (OSV) is a challenging pattern recognition task, especially in presence of skilled forgeries that are not available during training. This study aims to tackle its challenges and meet the substantial need for generalization for OSV by examining different loss functions for Convolutional Neural Network (CNN). We adopt our new approach to OSV by asking two questions: 1. which classification loss provides more generalization for feature learning in OSV? , and 2. How integration of different losses into a unified multi-loss function lead to an improved learning framework? These questions are studied based on analysis of three loss functions, including cross entropy, Cauchy-Schwarz divergence, and hinge loss. According to complementary features of these losses, we combine them into a dynamic multi-loss function and propose a novel ensemble framework for simultaneous use of them in CNN. Our proposed Multi-Loss Snapshot Ensemble (MLSE) consists of several sequential trials. In each trial, a dominant loss function is selected from the multi-loss set, and the remaining losses act as a regularizer. Different trials learn diverse representations for each input based on signature identification task. This multi-representation set is then employed for the verification task. An ensemble of SVMs is trained on these representations, and their decisions are finally combined according to the selection of most generalizable SVM for each user. We conducted two sets of experiments based on two different protocols of OSV, i.e., writer-dependent and writer-independent on three signature datasets: GPDS-Synthetic, MCYT, and UT-SIG. Based on the writer-dependent OSV protocol, we achieved substantial improvements over the best EERs in the literature. The results of the second set of experiments also confirmed the robustness to the arrival of new users enrolled in the OSV system.

[1]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[2]  Jesús Martínez del Rincón,et al.  Person Reidentification Using Deep Convnets With Multitask Learning , 2017, IEEE Transactions on Circuits and Systems for Video Technology.

[3]  Pascal Vincent,et al.  Dropout as data augmentation , 2015, ArXiv.

[4]  Subhash C. Bagui,et al.  Combining Pattern Classifiers: Methods and Algorithms , 2005, Technometrics.

[5]  Miguel A. Ferrer,et al.  Static Signature Synthesis: A Neuromotor Inspired Approach for Biometrics , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  Lorenzo Rosasco,et al.  Are Loss Functions All the Same? , 2004, Neural Computation.

[7]  Pengtao Xie,et al.  Diversifying Restricted Boltzmann Machine for Document Modeling , 2015, KDD.

[8]  Mustafa Berkay Yilmaz,et al.  Score level fusion of classifiers in off-line signature verification , 2016, Inf. Fusion.

[9]  Mustafa Berkay Yilmaz,et al.  Hybrid User-Independent and User-Dependent Offline Signature Verification with a Two-Channel CNN , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[10]  Yichuan Tang,et al.  Deep Learning using Linear Support Vector Machines , 2013, 1306.0239.

[11]  Luiz Eduardo Soares de Oliveira,et al.  Learning features for offline handwritten signature verification using deep convolutional neural networks , 2017, Pattern Recognit..

[12]  Lucas Beyer,et al.  In Defense of the Triplet Loss for Person Re-Identification , 2017, ArXiv.

[13]  Kiyoharu Aizawa,et al.  Significance of Softmax-Based Features in Comparison to Distance Metric Learning-Based Features , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[14]  Reza Ebrahimpour,et al.  Mixture of experts: a literature survey , 2014, Artificial Intelligence Review.

[15]  Imran Siddiqi,et al.  Towards the design of an offline signature verifier based on a small number of genuine samples for training , 2018, Expert Syst. Appl..

[16]  Robert Sabourin,et al.  Fixed-sized representation learning from offline handwritten signatures of different sizes , 2018, International Journal on Document Analysis and Recognition (IJDAR).

[17]  Loris Nanni,et al.  An On-Line Signature Verification System Based on Fusion of Local and Global Information , 2005, AVBPA.

[18]  Babak Nadjar Araabi,et al.  Deep Multitask Metric Learning for Offline Signature Verification , 2016, Pattern Recognit. Lett..

[19]  Luiz Eduardo Soares de Oliveira,et al.  Reducing forgeries in writer-independent off-line signature verification through ensemble of classifiers , 2010, Pattern Recognit..

[20]  Babak Nadjar Araabi,et al.  Learning Representations from Persian Handwriting for Offline Signature Verification, a Deep Transfer Learning Approach , 2019, 2019 4th International Conference on Pattern Recognition and Image Analysis (IPRIA).

[21]  Anastasios L. Kesidis,et al.  A Comprehensive Study of Sparse Representation Techniques for Offline Signature Verification , 2018, IEEE Transactions on Biometrics, Behavior, and Identity Science.

[22]  N. Otsu A threshold selection method from gray level histograms , 1979 .

[23]  Robert Sabourin,et al.  Offline handwritten signature verification — Literature review , 2017, 2017 Seventh International Conference on Image Processing Theory, Tools and Applications (IPTA).

[24]  Shuicheng Yan,et al.  Multi-loss Regularized Deep Neural Network , 2016, IEEE Transactions on Circuits and Systems for Video Technology.

[25]  Kilian Q. Weinberger,et al.  Snapshot Ensembles: Train 1, get M for free , 2017, ICLR.

[26]  Andrew Beng Jin Teoh,et al.  Image-based handwritten signature verification using hybrid methods of discrete Radon transform, principal component analysis and probabilistic neural network , 2016, Appl. Soft Comput..

[27]  Manabu Okawa,et al.  From BoVW to VLAD with KAZE features: Offline signature verification considering cognitive processes of forensic experts , 2018, Pattern Recognit. Lett..

[28]  Babak Nadjar Araabi,et al.  UTSig: A Persian offline signature dataset , 2016, IET Biom..

[29]  Wojciech Czarnecki,et al.  On Loss Functions for Deep Neural Networks in Classification , 2017, ArXiv.

[30]  Tianqi Chen,et al.  Empirical Evaluation of Rectified Activations in Convolutional Network , 2015, ArXiv.

[31]  Kiyoharu Aizawa,et al.  Significance of Softmax-Based Features over Metric Learning-Based Features , 2017 .

[32]  Christoph Meinel,et al.  Signature Embedding: Writer Independent Offline Signature Verification with Deep Metric Learning , 2016, ISVC.

[33]  Luiz Eduardo Soares de Oliveira,et al.  Writer-independent feature learning for Offline Signature Verification using Deep Convolutional Neural Networks , 2016, 2016 International Joint Conference on Neural Networks (IJCNN).

[34]  Jian Wang,et al.  Deep Metric Learning with Angular Loss , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[35]  Richard Socher,et al.  Using Mode Connectivity for Loss Landscape Analysis , 2018, ArXiv.

[36]  Cheng-Lin Liu,et al.  Offline signature verification using convolution Siamese network , 2018, International Conference on Graphic and Image Processing.

[37]  Ilias Theodorakopoulos,et al.  Parsimonious Coding and Verification of Offline Handwritten Signatures , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[38]  Julian Fiérrez,et al.  An Off-line Signature Verification System Based on Fusion of Local and Global Information , 2004, ECCV Workshop BioAW.

[39]  David Barber,et al.  Nesterov's accelerated gradient and momentum as approximations to regularised update descent , 2016, 2017 International Joint Conference on Neural Networks (IJCNN).

[40]  Muhammad Sharif,et al.  A framework for offline signature verification system: Best features selection approach , 2018, Pattern Recognit. Lett..

[41]  Yan Cui,et al.  Multi-phase Offline Signature Verification System Using Deep Convolutional Generative Adversarial Networks , 2016, 2016 9th International Symposium on Computational Intelligence and Design (ISCID).

[42]  Xin Yao,et al.  Diversity creation methods: a survey and categorisation , 2004, Inf. Fusion.

[43]  Ilias Theodorakopoulos,et al.  Offline Handwritten Signature Modeling and Verification Based on Archetypal Analysis , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[44]  Kaiqi Huang,et al.  Beyond Triplet Loss: A Deep Quadruplet Network for Person Re-identification , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[45]  Hassiba Nemmour,et al.  New off-line Handwritten Signature Verification method based on Artificial Immune Recognition System , 2016, Expert Syst. Appl..

[46]  Xiaogang Wang,et al.  Deep Learning Face Representation by Joint Identification-Verification , 2014, NIPS.

[47]  Jesús Francisco Vargas-Bonilla,et al.  Off-line signature verification based on grey level information using texture features , 2011, Pattern Recognit..

[48]  Robert Sabourin,et al.  A Writer-Independent Approach for Offline Signature Verification using Deep Convolutional Neural Networks Features , 2018, 2018 7th Brazilian Conference on Intelligent Systems (BRACIS).

[49]  Jian Sun,et al.  Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[50]  Hayaru Shouno,et al.  Analysis of Dropout Learning Regarded as Ensemble Learning , 2016, ICANN.

[51]  Alireza Alaei,et al.  An Efficient Signature Verification Method Based on an Interval Symbolic Representation and a Fuzzy Similarity Measure , 2017, IEEE Transactions on Information Forensics and Security.

[52]  Réjean Plamondon,et al.  Handwritten Signature Verification: New Advancements and Open Issues , 2012, 2012 International Conference on Frontiers in Handwriting Recognition.

[53]  Bernadette Dorizzi,et al.  Online Handwritten Signature Verification , 2009 .

[54]  Babak Nadjar Araabi,et al.  Active Transfer Learning for Persian Offline Signature Verification , 2019, 2019 4th International Conference on Pattern Recognition and Image Analysis (IPRIA).

[55]  Musa Mailah,et al.  Biometric signature verification , 2008 .

[56]  Manabu Okawa,et al.  Synergy of foreground-background images for feature extraction: Offline signature verification using Fisher vector with fused KAZE features , 2018, Pattern Recognit..

[57]  Reza Ebrahimpour,et al.  Incorporation of a Regularization Term to Control Negative Correlation in Mixture of Experts , 2012, Neural Processing Letters.

[58]  Hassiba Nemmour,et al.  Handwritten signature verification using the quad-tree histogram of templates and a Support Vector-based artificial immune classification , 2017, Image Vis. Comput..

[59]  Luiz Eduardo Soares de Oliveira,et al.  Analyzing features learned for Offline Signature Verification using Deep CNNs , 2016, 2016 23rd International Conference on Pattern Recognition (ICPR).

[60]  Robert Sabourin,et al.  Dynamic selection of generative-discriminative ensembles for off-line signature verification , 2012, Pattern Recognit..

[61]  Alex Alexandridis,et al.  Writer independent offline signature verification based on asymmetric pixel relations and unrelated training-testing datasets , 2019, Expert Syst. Appl..

[62]  George Economou,et al.  Hierarchical Dictionary Learning and Sparse Coding for Static Signature Verification , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[63]  Mustafa Berkay Ylmaz,et al.  Score level fusion of classifiers in off-line signature verification , 2016 .

[64]  Umapada Pal,et al.  SigNet: Convolutional Siamese Network for Writer Independent Offline Signature Verification , 2017, ArXiv.

[65]  George Economou,et al.  Offline signature verification and quality characterization using poset-oriented grid features , 2016, Pattern Recognit..

[66]  Reza Ebrahimpour,et al.  Combining features of negative correlation learning with mixture of experts in proposed ensemble methods , 2012, Appl. Soft Comput..

[67]  José Carlos Príncipe,et al.  Closed-form cauchy-schwarz PDF divergence for mixture of Gaussians , 2011, The 2011 International Joint Conference on Neural Networks.