Attention-Based CNN-RNN Arabic Text Recognition from Natural Scene Images
According to statistics, there are 422 million speakers of the Arabic language. Islam is the second-largest religion in the world, and its followers constitute approximately 25% of the world’s population. Since the Holy Quran is in Arabic, nearly all Muslims understand the Arabic language per some a...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-07-01
|
Series: | Forecasting |
Subjects: | |
Online Access: | https://www.mdpi.com/2571-9394/3/3/33 |
_version_ | 1797519279071952896 |
---|---|
author | Hanan Butt Muhammad Raheel Raza Muhammad Javed Ramzan Muhammad Junaid Ali Muhammad Haris |
author_facet | Hanan Butt Muhammad Raheel Raza Muhammad Javed Ramzan Muhammad Junaid Ali Muhammad Haris |
author_sort | Hanan Butt |
collection | DOAJ |
description | According to statistics, there are 422 million speakers of the Arabic language. Islam is the second-largest religion in the world, and its followers constitute approximately 25% of the world’s population. Since the Holy Quran is in Arabic, nearly all Muslims understand the Arabic language per some analytical information. Many countries have Arabic as their native and official language as well. In recent years, the number of internet users speaking the Arabic language has been increased, but there is very little work on it due to some complications. It is challenging to build a robust recognition system (RS) for cursive nature languages such as Arabic. These challenges become more complex if there are variations in text size, fonts, colors, orientation, lighting conditions, noise within a dataset, etc. To deal with them, deep learning models show noticeable results on data modeling and can handle large datasets. Convolutional neural networks (CNNs) and recurrent neural networks (RNNs) can select good features and follow the sequential data learning technique. These two neural networks offer impressive results in many research areas such as text recognition, voice recognition, several tasks of Natural Language Processing (NLP), and others. This paper presents a CNN-RNN model with an attention mechanism for Arabic image text recognition. The model takes an input image and generates feature sequences through a CNN. These sequences are transferred to a bidirectional RNN to obtain feature sequences in order. The bidirectional RNN can miss some preprocessing of text segmentation. Therefore, a bidirectional RNN with an attention mechanism is used to generate output, enabling the model to select relevant information from the feature sequences. An attention mechanism implements end-to-end training through a standard backpropagation algorithm. |
first_indexed | 2024-03-10T07:40:39Z |
format | Article |
id | doaj.art-6cbfa48e9cd145d1aae0bf7c1bfb1204 |
institution | Directory Open Access Journal |
issn | 2571-9394 |
language | English |
last_indexed | 2024-03-10T07:40:39Z |
publishDate | 2021-07-01 |
publisher | MDPI AG |
record_format | Article |
series | Forecasting |
spelling | doaj.art-6cbfa48e9cd145d1aae0bf7c1bfb12042023-11-22T13:06:29ZengMDPI AGForecasting2571-93942021-07-013352054010.3390/forecast3030033Attention-Based CNN-RNN Arabic Text Recognition from Natural Scene ImagesHanan Butt0Muhammad Raheel Raza1Muhammad Javed Ramzan2Muhammad Junaid Ali3Muhammad Haris4Department of Computer Science, COMSATS University Islamabad, Islamabad 45550, PakistanDepartment of Software Engineering, College of Technology, Firat University, 23000 Elazig, TurkeyDepartment of Computer Science, COMSATS University Islamabad, Islamabad 45550, PakistanDepartment of Computer Science, COMSATS University Islamabad, Islamabad 45550, PakistanDepartment of Computer Science, COMSATS University Islamabad, Islamabad 45550, PakistanAccording to statistics, there are 422 million speakers of the Arabic language. Islam is the second-largest religion in the world, and its followers constitute approximately 25% of the world’s population. Since the Holy Quran is in Arabic, nearly all Muslims understand the Arabic language per some analytical information. Many countries have Arabic as their native and official language as well. In recent years, the number of internet users speaking the Arabic language has been increased, but there is very little work on it due to some complications. It is challenging to build a robust recognition system (RS) for cursive nature languages such as Arabic. These challenges become more complex if there are variations in text size, fonts, colors, orientation, lighting conditions, noise within a dataset, etc. To deal with them, deep learning models show noticeable results on data modeling and can handle large datasets. Convolutional neural networks (CNNs) and recurrent neural networks (RNNs) can select good features and follow the sequential data learning technique. These two neural networks offer impressive results in many research areas such as text recognition, voice recognition, several tasks of Natural Language Processing (NLP), and others. This paper presents a CNN-RNN model with an attention mechanism for Arabic image text recognition. The model takes an input image and generates feature sequences through a CNN. These sequences are transferred to a bidirectional RNN to obtain feature sequences in order. The bidirectional RNN can miss some preprocessing of text segmentation. Therefore, a bidirectional RNN with an attention mechanism is used to generate output, enabling the model to select relevant information from the feature sequences. An attention mechanism implements end-to-end training through a standard backpropagation algorithm.https://www.mdpi.com/2571-9394/3/3/33image text recognitiondeep learningrecurrent neural networks (RNNs)convolutional neural networks (CNNs)bidirectional RNNattention mechanism |
spellingShingle | Hanan Butt Muhammad Raheel Raza Muhammad Javed Ramzan Muhammad Junaid Ali Muhammad Haris Attention-Based CNN-RNN Arabic Text Recognition from Natural Scene Images Forecasting image text recognition deep learning recurrent neural networks (RNNs) convolutional neural networks (CNNs) bidirectional RNN attention mechanism |
title | Attention-Based CNN-RNN Arabic Text Recognition from Natural Scene Images |
title_full | Attention-Based CNN-RNN Arabic Text Recognition from Natural Scene Images |
title_fullStr | Attention-Based CNN-RNN Arabic Text Recognition from Natural Scene Images |
title_full_unstemmed | Attention-Based CNN-RNN Arabic Text Recognition from Natural Scene Images |
title_short | Attention-Based CNN-RNN Arabic Text Recognition from Natural Scene Images |
title_sort | attention based cnn rnn arabic text recognition from natural scene images |
topic | image text recognition deep learning recurrent neural networks (RNNs) convolutional neural networks (CNNs) bidirectional RNN attention mechanism |
url | https://www.mdpi.com/2571-9394/3/3/33 |
work_keys_str_mv | AT hananbutt attentionbasedcnnrnnarabictextrecognitionfromnaturalsceneimages AT muhammadraheelraza attentionbasedcnnrnnarabictextrecognitionfromnaturalsceneimages AT muhammadjavedramzan attentionbasedcnnrnnarabictextrecognitionfromnaturalsceneimages AT muhammadjunaidali attentionbasedcnnrnnarabictextrecognitionfromnaturalsceneimages AT muhammadharis attentionbasedcnnrnnarabictextrecognitionfromnaturalsceneimages |