Off-line Handwritten Character Recognition using Geometric Moment Features and Fuzzy Clustering
Amit Choudhary1, Savita Ahlawat2
1Amit Choudhary, Department of Computer Science, Maharaja Surajmal Institute, New Delhi, India.
2Savita Ahlawat, Department of Computer Science, Maharaja Surajmal Institute, New Delhi, India.
Manuscript received on 21 April 2019 | Revised Manuscript received on 26 May 2019 | Manuscript published on 30 May 2019 | PP: 3001-3006 | Volume-8 Issue-1, May 2019 | Retrieval Number: A2232058119/19©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: Now a days, the handwritten character recognition is a challenging task for the researchers because of the difficulty in identifying local invariant patterns due to different writing styles of every individual. In the present work, an off-line handwritten character recognition using geometric moment features is investigated. The geometric moment features are directly related to geometric features of a pattern and are also known as pattern sensitive features. They are invariant under change of size, orientation and translation. The Fuzzy C-means algorithm is proposed for clustering because the geometric invariant features contains a lot of redundant information. The proposed work employs geometric moment invariant features and achieved a classification accuracy of 86.50% which is quite satisfactory.
Index Terms: Geometric Moment Features, Fuzzy Clustering, Character Recognition, Neural Network.
Scope of the Article: Fuzzy Logics