D6305048419

Handwritten Digit Recognition Accuracy Comparison of Various Techniques
Arun.V¹, Abinesh P², Akash Kumar S³, Arvind Kumar D.R^4
1Arun V, Assistant Professor, SRM Institute of Science and Technology, Ramapuram, Chennai (Tamil Nadu), India.
²Abinesh P, UG Scholars, SRM Institute of Science and Technology, Ramapuram, Chennai (Tamil Nadu), India.
³Akash Kumar S, UG Scholars, SRM Institute of Science and Technology, Ramapuram, Chennai (Tamil Nadu), India.
⁴Arvind Kumar D R, UG Scholars, SRM Institute of Science and Technology, Ramapuram, Chennai (Tamil Nadu), India.
Manuscript received on 18 April 2019 | Revised Manuscript received on 25 April 2019 | Manuscript published on 30 April 2019 | PP: 1499-1502 | Volume-8 Issue-4, April 2019 | Retrieval Number: D6305048419/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: Hand written characters and numbers recognition methodology has sought more attention among the various developed learnings and algorithm in today’s current scenario. The learnings which are talking about are a machine and deep learning in this paper, we used a calculation of machine learning certain methods like KNN, RFC and SVM and made a comparison with deep learning CNN method calculation with help of utilizing tensor flow, keras and theano. Utilizing the above techniques we get a more accurate result on CNN using theano as contrasted with using SVM, KNN and RFC.
Keywords: Multilayer CNN, Supervised Vector Machine, K-Nearest Neighbors, Random Forest Classifier.
Scope of the Article: Pattern Recognition