Capsule-Based Persian/Arabic Robust Handwritten Digit Recognition Using EM Routing

In this paper, the problem of handwritten digit recognition has been addressed. However, the underlying language is Persian/Arabic, and the system with which this task is a capsule network (CapsNet) which has recently emerged as a more advanced architecture than its ancestor, namely CNN (Convolutional Neural Network). The training of the architecture is performed using Hoda dataset, which has been provided for Persian/Arabic handwritten digits. The output of the system, clearly outperforms the results achieved by its ancestors, as well as other previously presented recognition algorithms.

[1]  Laurens van der Maaten,et al.  Accelerating t-SNE using tree-based algorithms , 2014, J. Mach. Learn. Res..

[2]  Luiz Eduardo Soares de Oliveira,et al.  Handwritten digit segmentation: Is it still necessary? , 2018, Pattern Recognit..

[3]  Geoffrey E. Hinton,et al.  Visualizing Data using t-SNE , 2008 .

[4]  Yang Jin,et al.  Capsule Network Performance on Complex Data , 2017, ArXiv.

[5]  Geoffrey E. Hinton,et al.  Matrix capsules with EM routing , 2018, ICLR.

[6]  Chen Xu,et al.  MS-CapsNet: A Novel Multi-Scale Capsule Network , 2018, IEEE Signal Processing Letters.

[7]  Alireza Alaei,et al.  Using Modified Contour Features and SVM Based Classifier for the Recognition of Persian/Arabic Handwritten Numerals , 2009, 2009 Seventh International Conference on Advances in Pattern Recognition.

[8]  Xiaogang Wang,et al.  Neural Network Encapsulation , 2018, ECCV.

[9]  Ehsanollah Kabir,et al.  Introducing a very large dataset of handwritten Farsi digits and a study on their varieties , 2007, Pattern Recognit. Lett..

[10]  Alireza Alaei,et al.  Fine Classification of Unconstrained Handwritten Persian/Arabic Numerals by Removing Confusion amongst Similar Classes , 2009, 2009 10th International Conference on Document Analysis and Recognition.

[11]  J. Ó. Neill.  Siamese Capsule Networks , 2018, 1805.07242.

[12]  Ulas Bagci,et al.  Capsules for Object Segmentation , 2018, ArXiv.

[13]  Jun Li,et al.  On Learning and Learned Data Representation by Capsule Networks , 2019, IEEE Access.

[14]  Geoffrey E. Hinton,et al.  Dynamic Routing Between Capsules , 2017, NIPS.

[15]  Cheng-Lin Liu,et al.  Handwritten digit recognition: benchmarking of state-of-the-art techniques , 2003, Pattern Recognit..

[16]  Ching Y. Suen,et al.  Application of Support Vector Machines for Recognition of Handwritten Arabic/Persian Digits , 2003 .

[17]  Davar Giveki,et al.  Farsi/Arabic handwritten digit recognition based on ensemble of SVD classifiers and reliable multi-phase PSO combination rule , 2012, International Journal on Document Analysis and Recognition (IJDAR).

[18]  Jun Li,et al.  On Learning and Learned Representation with Dynamic Routing in Capsule Networks , 2018, ArXiv.

[19]  Ali Borji,et al.  Invariance analysis of modified C2 features: case study—handwritten digit recognition , 2009, Machine Vision and Applications.

[20]  Saeed Mozaffari,et al.  Structural decomposition and statistical description of Farsi/Arabic handwritten numeric characters , 2005, Eighth International Conference on Document Analysis and Recognition (ICDAR'05).

[21]  Harris Drucker,et al.  Comparison of learning algorithms for handwritten digit recognition , 1995 .