Open Access Open Access  Restricted Access Subscription or Fee Access

Performance Comparison of Handwritten Digit Recognition Using Machine Learning Techniques

Kilari Veera Swamy, D. Sai Pavithra

Abstract


Handwritten pattern, character and digit recognition has always been the most demanding and interesting fields of image processing. The objective of this work is representing and demonstrating the algorithms which are used to recognize handwritten digits. Recognition of handwritten digits has always a bit complex, but it is mostly used in various day-to-day applications. In the task of recognition, the digits are not correctly or accurately scripted or written as they vary in size, shape and other features which causes the extraction process of features and the segmentation process of handwritten digits to be challenging. The MNIST database is used as the database for the handwritten digit recognition. The work involves testing and training the dataset and feature extraction and classification. The horizontal and vertical projection techniques are used for segmentation in the proposed work. Hog features are used in this paper for SVM. SVM and CNN models are developed during training phase. MNIST database is used to train the model. SVM and CNN are used to recognize the handwritten digit. Accuracy of the both methods are compared.


Full Text:

PDF

References


M. Koteswara Rao, K Veera Swamy and K. Anithasheela, “Face Recognition based on unsupervised clustering feature extraction using Gaussian mixture model,” Journal of advanced research in dynamical & control systems. JARDCS.2019;Volume 11,(7):1626-1635.

Rohan Sethi, lla Kaushik, “Hand-Written Digit Recognition using Machine Learning: IEEE 9th International Conference on Communication Systems and Network Technologies (CSNT),2020. 2020 April 10-12; Gwalior, India; US;IEEE Press2020.

B. Sneha Bharathi and K Veera Swamy, “Effective Image segmentation using modified K-means Technique,” Fourth International Conference on Trends in Electronics and Informatics, pp. 757-762, 2020.

Zhao, K., “Handwritten Digit Recognition and Classification Using Machine Learning”. M.Sc. in Computing (Data Analytics), Technological University; Dublin,2018.

Vansh Jatana, “Machine Learning Algorithms”, Survey Paper,2020.

Machine learning: the power and promise of computers that learn by example – report, April 2017.

Support Vector Machine-Theory by Data science lovers in Machine Learning, Mar 30,2020.

S Thirumal Devi, K Veera Swamy, et al. “Feature concatenation for scene classification of VHR images based on canonical correlation analysis,” IOP Conference Series: Materials Science and Engineering,:2020; Volume 1070, doi:10.1088/1757-899.

Savita Ahlawat, Amit Choudhary, e “Improved Handwritten Digit Recognition Using Convolutional Neural Networks (CNN). June 2020 , Article in Sensors (Basel, Switzerland).

Vijayalaxmi, R Rudraswamimath, et al, “ Handwritten Digit Recognition using CNN”, International Journal of Innovative Science and Research Technology, IJISRT.June 2019. Volume (4): 182-187.

Le, H and Borji, A, “ What are the receptive, effective receptive, and projective fields of neurons in convolutional neural networks”, arXiv:1705.07049, 2018.

Luo, W. Li, Y. Urtasun, R and Zemel R, “Understanding the effective receptive field in deep convolutional neural networks”, Conference on Neural Information Processing Systems, NIPS 2016; Barcelona, Spain.2016.

T Siva Ajay ,“Handwritten Digit Recognition using Convolutional Neural Networks”, International Research Journal of Engineering and Technology (IRJET).July 2017; Volume: 04 Issue:( 07) :2971-2976.

Scherer. D, Muller. A, Sven Bhenke “Evaluation of pooling operations in convolutional architectures for object recognition”, Proceedings of the International Conference on Artificial Neural Networks,2010 September; Thessaloniki, Greece, ICANN.

“MNIST digit database”, https://www.kaggle.com/c/digit-recognizer/data.

MNIST training and testing digit database, https://github.com/RubixML/MNIST.

Soufiane Hamida, Bouchaib Cherradi,et al. “Performance Evaluation of Machine Learning Algorithm in Handwritten digits Recognition”, International conference on Smart Systems and Data Science; 2019 Oct 3-4; Rabat, Morocco:IEEE; 2019.


Refbacks

  • There are currently no refbacks.