Comparison between Feature Based and Deep Learning Recognition Systems for Handwriting Arabic Numbers
Abstract
Feature extraction from images is an essential part of the recognition system. Calculating the appropriate features is critical to the part of the classification process. However, there are no standard features nor a widely accepted feature set exist applied to all applications, features must be application dependent. In contrast, deep learning extract features from an image without need for human hard-coding the features extraction process. This can be very useful to build a model for classification which can classify any type of images after trained with enough images with labels then the trained model can be used in different recognition applications to classify. This paper presents two techniques to build recognition system for Arabic handwriting numbers, the feature-based method shows accepted results. However, the deep learning method gives more accurate results and required less study on how Arabic number is written and no hand-coding algorithms needed for feature extraction to be used in the classification process.
Keywords: Handwriting Recognition, Image Processing, Features Extraction, Machine Learning, Deep Learning, Classification.
![Download the IISTE publication guideline!](https://www.iiste.org/sub/iiste_footer_journal.jpg)
To list your conference here. Please contact the administrator of this platform.
Paper submission email: CEIS@iiste.org
ISSN (Paper)2222-1727 ISSN (Online)2222-2863
Please add our address "contact@iiste.org" into your email contact list.
This journal follows ISO 9001 management standard and licensed under a Creative Commons Attribution 3.0 License.
Copyright © www.iiste.org