Offline Handwriting Writer Identification using Depth-wise Separable Convolution with Siamese Network

Wirmanto Suteddy - Universitas Pendidikan Indonesia, Bandung, 40625, Indonesia
Devi Agustini - Universitas Pendidikan Indonesia, Bandung, 40625, Indonesia
Dastin Atmanto - Universitas Pendidikan Indonesia, Bandung, 40625, Indonesia


Citation Format:



DOI: http://dx.doi.org/10.62527/joiv.8.1.2148

Abstract


Offline handwriting writer identification has significant implications for forensic investigations and biometric authentication. Handwriting, as a distinctive biometric trait, provides insights into individual identity. Despite advancements in handcrafted algorithms and deep learning techniques, the persistent challenges related to intra-variability and inter-writer similarity continue to drive research efforts. In this study, we build on well-separated convolution architectures like the Xception architecture, which has proven to be robust in our previous research comparing various deep learning architectures such as MobileNet, EfficientNet, ResNet50, and VGG16, where Xception demonstrated minimal training-validation disparities for writer identification. Expanding on this, we use a model based on similarity or dissimilarity approaches to identify offline writers' handwriting, known as the Siamese Network, that incorporates the Xception architecture. Similarity or dissimilarity measurements are based on the Manhattan or L1 distance between representation vectors of each input pair. We train publicly available IAM and CVL datasets; our approach achieves accuracy rates of 99.81% for IAM and 99.88% for CVL. The model was evaluated using evaluation metrics, which revealed only two error predictions in the IAM dataset, resulting in 99.75% accuracy, and five error predictions for CVL, resulting in 99.57% accuracy. These findings modestly surpass existing achievements, highlighting the potential inherent in our methodology to enhance writer identification accuracy. This study underscores the effectiveness of integrating the Siamese Network with depth-wise separable convolution, emphasizing the practical implications for supporting writer identification in real-world applications.

Keywords


Offline handwriting; writer identification; siamese network; similarity approach; depth-wise separable convolution

Full Text:

PDF

References


N. Purohit and S. Panwar, “State-of-the-Art: Offline Writer Identification Methodologies,†in 2021 International Conference on Computer Communication and Informatics (ICCCI), Coimbatore, India: IEEE, Jan. 2021, pp. 1–8. doi: 10.1109/ICCCI50826.2021.9402539.

T. Bahram, “A texture-based approach for offline writer identification,†Journal of King Saud University - Computer and Information Sciences, vol. 34, no. 8, pp. 5204–5222, Sep. 2022, doi: 10.1016/j.jksuci.2022.06.003.

M. Chammas, A. Makhoul, and J. Demerjian, “Writer identification for historical handwritten documents using a single feature extraction method,†in 2020 19th IEEE International Conference on Machine Learning and Applications (ICMLA), Miami, FL, USA: IEEE, Dec. 2020, pp. 1–6. doi: 10.1109/ICMLA51294.2020.00010.

V. Christlein, D. Bernecker, F. Hönig, A. Maier, and E. Angelopoulou, “Writer Identification Using GMM Supervectors and Exemplar-SVMs,†Pattern Recognition, vol. 63, pp. 258–267, Mar. 2017, doi: 10.1016/j.patcog.2016.10.005.

M. Shabir, N. Islam, Z. Jan, and I. Khan, “Transformation Invariant Pashto Handwritten Text Classification and Prediction,†J CIRCUIT SYST COMP, vol. 32, no. 02, p. 2350020, Jan. 2023, doi: 10.1142/S0218126623500202.

A. Rehman, S. Naz, M. I. Razzak, and I. A. Hameed, “Automatic Visual Features for Writer Identification: A Deep Learning Approach,†IEEE Access, vol. 7, pp. 17149–17157, 2019, doi: 10.1109/ACCESS.2018.2890810.

A. Semma, Y. Hannad, I. Siddiqi, C. Djeddi, and M. El Youssfi El Kettani, “Writer Identification using Deep Learning with FAST Keypoints and Harris corner detector,†Expert Systems with Applications, vol. 184, p. 115473, Dec. 2021, doi: 10.1016/j.eswa.2021.115473.

S. He and L. Schomaker, “Deep adaptive learning for writer identification based on single handwritten word images,†Pattern Recognition, vol. 88, pp. 64–74, Apr. 2019, doi: 10.1016/j.patcog.2018.11.003.

S. He and L. Schomaker, “GR-RNN: Global-Context Residual Recurrent Neural Networks for Writer Identification.†arXiv, Apr. 11, 2021. Accessed: Aug. 01, 2023. [Online]. Available: http://arxiv.org/abs/2104.05036

L. G. Helal, D. Bertolini, Y. M. G. Costa, G. D. C. Cavalcanti, A. S. Britto, and L. E. S. Oliveira, “Representation Learning and Dissimilarity for Writer Identification,†in 2019 International Conference on Systems, Signals and Image Processing (IWSSIP), Osijek, Croatia: IEEE, Jun. 2019, pp. 63–68. doi: 10.1109/IWSSIP.2019.8787293.

V. Christlein, M. Gropp, S. Fiel, and A. Maier, “Unsupervised Feature Learning for Writer Identification and Writer Retrieval,†in 2017 14th IAPR International Conference on Document Analysis and Recognition (ICDAR), Kyoto: IEEE, Nov. 2017, pp. 991–997. doi: 10.1109/ICDAR.2017.165.

S. Chen, Y. Wang, C.-T. Lin, W. Ding, and Z. Cao, “Semi-supervised Feature Learning For Improving Writer Identification,†2018, doi: 10.48550/ARXIV.1807.05490.

S. He and L. Schomaker, “FragNet: Writer Identification Using Deep Fragment Networks,†IEEE Trans.Inform.Forensic Secur., vol. 15, pp. 3013–3022, 2020, doi: 10.1109/TIFS.2020.2981236.

P. Kumar and A. Sharma, “Segmentation-free writer identification based on convolutional neural network,†Computers & Electrical Engineering, vol. 85, p. 106707, Jul. 2020, doi: 10.1016/j.compeleceng.2020.106707.

L. Xing and Y. Qiao, “DeepWriter: A Multi-Stream Deep CNN for Text-independent Writer Identification.†arXiv, Aug. 02, 2016. Accessed: Aug. 16, 2023. [Online]. Available: http://arxiv.org/abs/1606.06472

A. Sulaiman, K. Omar, M. F. Nasrudin, and A. Arram, “Length Independent Writer Identification Based on the Fusion of Deep and Hand-Crafted Descriptors,†IEEE Access, vol. 7, pp. 91772–91784, 2019, doi: 10.1109/ACCESS.2019.2927286.

W. Suteddy, D. A. R. Agustini, A. Adiwilaga, and D. A. Atmanto, “End-To-End Evaluation of Deep Learning Architectures for Off-Line Handwriting Writer Identification: A Comparative Study,†JOIV : Int. J. Inform. Visualization, vol. 7, no. 1, p. 178, Feb. 2023, doi: 10.30630/joiv.7.1.1293.

S. Dey, A. Dutta, J. I. Toledo, S. K. Ghosh, J. Llados, and U. Pal, “SigNet: Convolutional Siamese Network for Writer Independent Offline Signature Verification.†arXiv, Sep. 30, 2017. Accessed: Aug. 01, 2023. [Online]. Available: http://arxiv.org/abs/1707.02131

W. Xiao and Y. Ding, “A Two-Stage Siamese Network Model for Offline Handwritten Signature Verification,†Symmetry, vol. 14, no. 6, p. 1216, Jun. 2022, doi: 10.3390/sym14061216.

N. Sharma et al., “Siamese Convolutional Neural Network-Based Twin Structure Model for Independent Offline Signature Verification,†Sustainability, vol. 14, no. 18, p. 11484, Sep. 2022, doi: 10.3390/su141811484.

C. Adak, S. Marinai, B. B. Chaudhuri, and M. Blumenstein, “Offline Bengali Writer Verification by PDF-CNN and Siamese Net,†in 2018 13th IAPR International Workshop on Document Analysis Systems (DAS), Vienna: IEEE, Apr. 2018, pp. 381–386. doi: 10.1109/DAS.2018.33.

V. Kumar and S. Sundaram, “Siamese based Neural Network for Offline Writer Identification on word level data,†2022, doi: 10.48550/ARXIV.2211.14443.

N. Dlamini and T. L. Van Zyl, “Author Identification from Handwritten Characters using Siamese CNN,†in 2019 International Multidisciplinary Information Technology and Engineering Conference (IMITEC), Vanderbijlpark, South Africa: IEEE, Nov. 2019, pp. 1–6. doi: 10.1109/IMITEC45504.2019.9015897.

U.-V. Marti and H. Bunke, “The IAM-database: an English sentence database for offline handwriting recognition,†International Journal on Document Analysis and Recognition, vol. 5, no. 1, pp. 39–46, Nov. 2002, doi: 10.1007/s100320200071.

F. Kleber, S. Fiel, M. Diem, and R. Sablatnig, “CVL-DataBase: An Off-Line Database for Writer Retrieval, Writer Identification and Word Spotting,†in 2013 12th International Conference on Document Analysis and Recognition, Washington, DC, USA: IEEE, Aug. 2013, pp. 560–564. doi: 10.1109/ICDAR.2013.117.

F. Chollet, “Xception: Deep Learning with Depthwise Separable Convolutions,†in 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI: IEEE, Jul. 2017, pp. 1800–1807. doi: 10.1109/CVPR.2017.195.

L. Utkin, M. Kovalev, and E. Kasimov, “An Explanation Method for Siamese Neural Networks,†in Proceedings of International Scientific Conference on Telecommunications, Computing and Control, N. Voinov, T. Schreck, and S. Khan, Eds., in Smart Innovation, Systems and Technologies, vol. 220. Singapore: Springer Singapore, 2021, pp. 219–230. doi: 10.1007/978-981-33-6632-9_19.

C. Saedi and M. Dras, “Siamese networks for large-scale author identification,†Computer Speech & Language, vol. 70, p. 101241, Nov. 2021, doi: 10.1016/j.csl.2021.101241.

V. Ruiz, I. Linares, A. Sanchez, and J. F. Velez, “Off-line handwritten signature verification using compositional synthetic generation of signatures and Siamese Neural Networks,†Neurocomputing, vol. 374, pp. 30–41, Jan. 2020, doi: 10.1016/j.neucom.2019.09.041.

B. Soujanya, C. Suresh, and T. Sitamahalakshmi, “Hand Written Character Identification of Kha Gunintham with Siamese CNN Network,†Advances in Dynamical Systems and Applications, vol. 16, no. 2, pp. 1333–1347, 2021.