HDNcfm: Handwritten Digit Recognition System Using No Combination of Feature Maps
Main Article Content
Abstract
Convolutional Neural Network (CNN) has proved its significance as a robust method for classification of various image based applications such as face recognition and handwritten digit recognition. Traditional neural netwotk’s input feature maps are convolved along with the kernal, which aids in increasing the performance of the system. However, the problem of some feature lost arise in traditional neural network while combining the feature maps, and also the application does not tend to do well at a large scale networks. Hence, to address these issues, in this paper, we proposed, HDNcfm: Handwritten Digit Recognition System using Combination of Feature Maps, which tend to converge faster than Combination of Feature Map(CFM) method, and also upgrade its performance. In this method as the input features doesn’t combined, the number of input feature and output feature can be same and therefore, this method can also be applied for relatively smaller dataset. Performance of the proposed approach has been evaluated using the parameters such as-convergence rate, and error rate. Results obtained clearly show the superior performance of the proposed scheme as compared to the traditional neural network based schemes.
Downloads
Download data is not yet available.
Article Details
Section
Articles
COPYRIGHT
Submission of a manuscript implies: that the work described has not been published before, that it is not under consideration for publication elsewhere; that if and when the manuscript is accepted for publication, the authors agree to automatic transfer of the copyright to the publisher.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work
- The journal allows the author(s) to retain publishing rights without restrictions.
- The journal allows the author(s) to hold the copyright without restrictions.