Performance of Machine Learning and Deep Learning on Arabic Handwritten Text Recognition

Author(s):  
Jawad H Alkhateeb ◽  
Aiman A Turani ◽  
AbdulRahman A. Alsewari
Author(s):  
Sri. Yugandhar Manchala ◽  
Jayaram Kinthali ◽  
Kowshik Kotha ◽  
Kanithi Santosh Kumar, Jagilinki Jayalaxmi ◽  

Author(s):  
Arthur Flor de Sousa Neto ◽  
Byron Leite Dantas Bezerra ◽  
Alejandro Hector Toselli ◽  
Estanislau Baptista Lima

Author(s):  
Yojana Swapneel Samant

The human race has shown a huge interest in machines over the years and has developed and advanced to a very large extent in this domain. Starting from the object identification and classification through pictures to editing for the captured image or video everything can be performed through machines and advanced systems, one such part of this advanced technology is deep learning or machine learning. which comes with its own individual set of modules, algorithms, and techniques. Similar to this domain one such idea which was discovered is handwritten digit recognition. This is one of such areas where development and research occur around the numerical also known as digits, where a number of possibilities, permutations, and combinations are attained to gain accurate results this can also be perceived as the ability of computers to interpret and understand the given input which is through number plates, photographs, documents or can be in a handwritten format and to convert it in digital format as an output through screens.


Author(s):  
Jebaveerasingh Jebadurai ◽  
Immanuel Johnraja Jebadurai ◽  
Getzi Jeba Leelipushpam Paulraj ◽  
Sushen Vallabh Vangeepuram

Author(s):  
Bayram Annanurov ◽  
Norliza Noor

<p>The motivation of this study is to develop a compact offline recognition model for Khmer handwritten text that would be successfully applied under limited access to high-performance computational hardware. Such a task aims to ease the ad-hoc digitization of vast handwritten archives in many spheres. Data collected for previous experiments were used in this work. The oneagainst-all classification was completed with state-of-the-art techniques. A compact deep learning model (2+1CNN), with two convolutional layers and one fully connected layer, was proposed. The recognition rate came out to be within 93-98%. The compact model is performed on par with the state-of-theart models. It was discovered that computational capacity requirements usually associated with deep learning can be alleviated, therefore allowing applications under limited computational power.</p>


2021 ◽  
Vol 8 (6) ◽  
pp. 870-881
Author(s):  
Rohini G. Khalkar ◽  
Adarsh Singh Dikhit ◽  
Anirudh Goel

Sign in / Sign up

Export Citation Format

Share Document