Arabic Word Recognition Research Articles

The effect of dots on Arabic visual word recognition: a psycholinguistic perspective

The effect of dots on Arabic visual word recognition: a psycholinguistic perspective

Speaker independent recognition of low-resourced multilingual Arabic spoken words through hybrid fusion

Speaker independent recognition of low-resourced multilingual Arabic spoken words through hybrid fusion

Cursive Arabic handwritten word recognition system using majority voting and k-NN for feature descriptor selection

Cursive Arabic handwritten word recognition system using majority voting and k-NN for feature descriptor selection

A Novel Hybrid DL Model for Printed Arabic Word Recognition based on GAN

The recognition of printed Arabic words remains an open area for research since Arabic is among the most complex languages. Prior research has shown that few efforts have been made to develop models of accurate Arabic recognition, as most of these models have faced the increasing complexity of the performance and lack of benchmark Arabic datasets. Meanwhile, Deep learning models, such as Convolutional Neural Networks (CNNs), have been shown to be beneficial in reducing the error rate and enhancing accuracy in Arabic character recognition systems. The reliability of these models increases with the depth of layers. Still, the essential condition for more layers is an extensive amount of data. Since CNN generates features by analysing large amounts of data, its performance is directly proportional to the volume of data, as DL models are considered data-hungry algorithms. Nevertheless, this technique suffers from poor generalisation ability and overfitting issues, which affect the Arabic recognition models' accuracy. These issues are due to the limited availability of Arabic databases in terms of accessibility and size, which led to a central problem facing the Arabic language nowadays. Therefore, the Arabic character recognition models still have gaps that need to be bridged. The Deep Learning techniques are also to be improved to increase the accuracy by manipulating the strength of technique in a neural network for handling the lack of datasets and the generalisation ability of the neural network in model building. To solve these problems, this study proposes a hybrid model for Arabic word recognition by adapting a deep convolutional neural network (DCNN) to work as a classifier based on a generative adversarial network (GAN) work as a data augmentation technique to develop a robust hybrid model for improving the accuracy and generalisation ability. Each proposed model is separately evaluated and compared with other state-of-the-art models. These models are tested on the Arabic printed text image dataset (APTI). The proposed hybrid deep learning model shows excellent performance regarding the accuracy, with a score of 99.76% compared to 94.81% for the proposed DCNN model on the APTI dataset. The proposed model indicates highly competitive performance and enhanced accuracy compared to the existing state-of-the-art Arabic printed word recognition models. The results demonstrate that the generalisation of networks and the handling of overfitting have also improved. This study output is comparable to other competitive models and contributes an enhanced Arabic recognition model to the body of knowledge.

Improving end-to-end deep learning methods for Arabic handwriting recognition

The number of Arabic handwriting documents has increased greatly. Studies conducted in the Arabic handwriting recognition field have progressed significantly in recent years in different areas. The existing studies for Arabic language compared to others such as Latin remains insufficient. During the last decade, neural networks (NNs) have become the de facto standard for deep learning, and the combination of two or more NNs has proved its ability to learn complex objects, such as handwriting, and emulate human brains. We proposed an approach combining convolutional NN and bidirectional long short-term memory for recognition. This unique approach makes it possible to recognize Arabic handwriting words without segmentation. The proposed architecture is very efficient in terms of accuracy for Arabic word recognition. The hyperparameters set for our model were chosen based on a structured process of randomness and grid search to increase the accuracy of the model.

Efficient feature descriptor selection for improved Arabic handwritten words recognition

<span lang="EN-US">Arabic handwritten text recognition has long been a difficult subject, owing to the similarity of its characters and the wide range of writing styles. However, due to the intricacy of Arabic handwriting morphology, solving the challenge of cursive handwriting recognition remains difficult. In this paper, we propose a new efficient based image processing approach that combines three image descriptors for the feature extraction phase. To prepare the training and testing datasets, we applied a series of preprocessing techniques to 100 classes selected from the handwritten Arabic database of the </span><em><span lang="EN-US">Institut Für Nachrichtentechnik/Ecole Nationale d'Ingénieurs de Tunis</span></em><span lang="EN-US"> (IFN/ENIT). Then, we trained the k-nearest neighbor’s algorithm (k-NN) algorithm to generate the best model for each feature extraction descriptor. The best k-NN model, according to common performance evaluation metrics, is used to classify Arabic handwritten images according to their classes. Based on the performance evaluation results of the three k-NN generated models, the majority-voting algorithm is used to combine the prediction results. A high recognition rate of up to 99.88% is achieved, far exceeding the state-of-the-art results using the IFN/ENIT dataset. The obtained results highlight the reliability of the proposed system for the recognition of handwritten Arabic words.</span>

Handwritten Arabic and Roman word recognition using holistic approach

Handwritten Arabic and Roman word recognition using holistic approach

Recognition of Arabic handwritten words using convolutional neural network

<span>A new method for recognizing automatically Arabic handwritten words was presented using convolutional neural network architecture. The proposed method is based on global approaches, which consists of recognizing all the words without segmenting into the characters in order to recognize them separately. Convolutional neural network (CNN) is a particular supervised type of neural network based on multilayer principle; our method needs a big dataset of word images to obtain the best result. To optimize our system, a new database was collected from the benchmarking Arabic handwriting database using the pre-processing such as rotation transformation, which is applied on the images of the database to create new images with different features. The convolutional neural network applied on our database that contains 40320 of Arabic handwritten words (26880 images for training set and 13440 for test set). Thus, different configurations on a public benchmark database were evaluated and compared with previous methods. Consequently, it is demonstrated a recognition rate with a success of 96.76%.</span>

Arabic Handwritten Word Recognition Based on Stationary Wavelet Transform Technique using Machine Learning

This paper is aimed at improving the performance of the word recognition system (WRS) of handwritten Arabic text by extracting features in the frequency domain using the Stationary Wavelet Transform (SWT) method using machine learning, which is a wavelet transform approach created to compensate for the absence of translation invariance in the Discrete Wavelets Transform (DWT) method. The proposed SWT-WRS of Arabic handwritten text consists of three main processes: word normalization, feature extraction based on SWT, and recognition. The proposed SWT-WRS based on the SWT method is evaluated on the IFN/ENIT database applying the Gaussian, linear, and polynomial support vector machine, the k-nearest neighbors, and ANN classifiers. ANN performance was assessed by applying the Bayesian Regularization (BR) and Levenberg-Marquardt (LM) training methods. Numerous wavelet transform (WT) families are applied, and the results prove that level 19 of the Daubechies family is the best WT family for the proposed SWT-WRS. The results also confirm the effectiveness of the proposed SWT-WRS in improving the performance of handwritten Arabic word recognition using machine learning. Therefore, the suggested SWT-WRS overcomes the lack of translation invariance in the DWT method by eliminating the up-and-down samplers from the proposed machine learning method.

Offline Arabic handwritten word recognition: A transfer learning approach

Offline Arabic handwritten word recognition is still a challenging task. Many deep learning approaches perform admirably on this task if the lexicon size is not too large and the number of training samples is sufficient for the training process. The transfer learning technique is commonly used to compensate for the lack of training samples, but there is a wide controversy about the effectiveness of applying it to cross-domain tasks. In this paper, we examine the performance of three deep convolution neural networks that have been randomly initialized for recognizing Arabic handwritten words. Then, we evaluate the performance of the ResNet18 model that has been pre-trained on the ImageNet dataset for the same task. Finally, we propose an approach based on sequentially transferring the mid-level word image representations through two consecutive phases using the ResNet18 model. We carried out four different sets of experiments using two popular offline Arabic handwritten word datasets: the AlexU-W and the IFN/ENIT (v2.0p1e) to figure out the most effective way of applying transfer learning. Our results demonstrate that using the ImageNet as a source dataset improves the recognition accuracy of the ten frequently misclassified words in the IFN/ENIT dataset by 14%, while our proposed approach gives a rise of 35.45%. In the whole dataset, we achieved recognition accuracy up to 96.11%, which is nearly a 2.5% enhancement compared with other state-of-the-art methods.

Combine SVM and KNN classifiers for Handwriting Arabic Word Recognition based on Multifeatures

This paper presents a proposed system for recognizing the handwritten Arabic words. The proposed system recognized the Arabic word as one entity without using segmentation stage, which converted the word into parts. A proposed method for feature extraction stage used two groups of feature extraction techniques. First group combines two techniques and the second group used single technique. First group combines gradient (directional) feature method with the Run Length Count method and second group based on Discrete Cosine Translation technique. Classification stage is based on combined SVM with KNN classifiers. A standard data set which is AHDB database is used to simulate the proposed system. The recognition accuracy for the experimental results of the proposed system is 97.11 %.

Isolated Arabic handwritten words recognition using EHD and HOG methods

<span>Handwriting recognition is a growing field of study in computer vision, artificial intelligence and pattern recognition technology aimed to recognizing texts and handwritings of hefty amount of produced official documents and paper works by institutes or governments. Using computer to distinguish and make these documents accessible and approachable is the goal of these efforts. Moreover, recognition of text has accomplished practically a major progress in many domains such as security sector and e-government structure and more. A system for recognition text’s handwriting was presented here relied on edge histogram descriptor (EHD), histogram of orientated gradients (HOG) features extraction and support vector machine (SVM) as a classifier is proposed in this paper. HOG and EHD give an optimal features of the Arabic hand-written text by extracting the directional properties of the text. Besides that, SVM is a most common machine learning classifier that obtaining an essential classification results within various kernel functions. The experimental evaluation is carried out for Arabic handwritten images from IESK-ArDB database using HOG, EHD features and proposed work provides 85% recognition rate.</span>

A Review: Isolated Arabic Words Recognition Using Artificial Intelligent Techniques

In recent few years, deep learning has fast growing in many fields as natural language processing, image recognition, handwriting recognition, computer vision, and speech recognition. Automatic speech recognition (ASR) is a technique that refers to translating spoken words from an acoustic waveform into a text equivalent to what the speaker says. More recently, the advances in deep learning can support ASR in improving the performance of systems accuracies. Arabic is a Semitic language, one of the oldest used and most communicated languages in the world. But, it least concentrated in the case of Arabic speech recognition and under-resourced languages. This paper presents a survey that focuses on an automatic speech recognition system based on isolating words technique for Arabic speech. It also highlights the facilities and tools for developing speech recognition systems. This work is intended to be a useful starting point for those who are interested in ASR.

Reading Process of Arab Children: An Eye-Tracking Study on Saudi Elementary Students

Many previous eye-tracking studies were conducted to examine how adult readers process different written languages. Relatively, only few eye-tracking studies have been conducted to observe the reading process of Arab children. This study investigated the influence of orthographic regularity on Saudi elementary grades’ English and Arabic words recognition. The eye movements of 15 grade-four students and 15 grade-six students were recorded while they read words that differ in frequency and regularity. Analysis of the visual information from the word-recognition process shows differences in the students’ eye movements for the two languages. There were statistically significant differences in the total fixation duration and fixation count between the two languages and between both groups. All the students showed longer processing time for English sentences than Arabic ones. However, Arabic-speaking students were influenced by English orthography with more processing difficulty for English irregular words. The visual information shows that more cross-linguistic differences are found in grade-four students’ results. Grade-four students transferred their first language (L1) reading strategies to read English words; however, Arabic reading methods cannot be effectively applied to reading irregular orthographies like English. This explains the increased eye-movement measurements of grade-four students compared to grade-six students, who fixated more on unfamiliar English words. Although orthographic regularity had a major effect on the word-recognition process in this study, the development of the students’ Arabic and English orthographic knowledge affected the progress of their visual word recognition across the two levels.

Stacked autoencoder for Arabic handwriting word recognition

Stacked autoencoder for Arabic handwriting word recognition

Stacked autoencoder for Arabic handwriting word recognition

Stacked autoencoder for Arabic handwriting word recognition

Visual Arabic Word Recognition during Subliminal and Supraliminal Presentations: A Lexical Decision Study in Normal and Disabled Readers

This study aimed at assessing visual word recognition in normal skilled (NRs) and disabled adult Arabic readers (DRs) using masked subliminal and supraliminal word presentations in a lexical decision task (LDT). The main assumption of this investigation was that, as a consequence of their reading disability, DRs would exhibit slower reaction times (RTs) and lower performance than NR in the supraliminal condition. By contrast, since words identified subliminally are thought to occur thanks to advanced orthographic (not phonological skills), it was assumed that this presentation mode would diminish the differences between DRs and NRs. RTs and d’ detection measures were analyzed in 59 NRs and 21 DRs who had to decide whether visually presented letter strings corresponded to real Arabic words or to pseudowords. The statistical analyses conducted on RTs and d’ showed that, although both groups showed a lexicality effect (faster RTs for words than for pseudowords), the DRs compared to NRs showed longer RTs and lower detection measures in the supraliminal mode. Importantly, in the subliminal mode, no group effect was found in terms of RTs and the size of the difference between groups in terms of detection decreased. Also, in this latter presentation condition, we observed that the lexicality effect was significant in the DRs but not in the NRs. These findings are discussed in relation to other studies indicating that DRs might have benefited from good orthographic skills, on which they rely to compensate for their phonological deficits.

VISUALIZATION WITH VOCABULARY TRACKING FOR ARABIC WORD RECOGNITION AND READING COMPREHENSION

Comprehending text is the aim of reading; however, there is the phenomenon of non-Arabic speakers in Malaysia reading the Qur’an, written in Arabic, without comprehension. Word recognition (WR) theory, through word frequency effect (WFE) and word superiority effect (WSE), are used as a basis to achieve reading comprehension (RC) of the Qur'an. The Eye of Qur'an (EoQu) interface was developed, to visualise word occurrences and word morphology. This is achieved through parallel plot and word segmentation visualization. EoQu can track a user’s personal vocabulary with a presentation of percentage and word position in the Qur’an. Consequently, users know their ability to recognize Arabic words in relation to the whole Qur’an to achieve RC. An experimental study was set up with 90 Malaysian participants, starting with a pre-test, followed by a stratified sampling to divide participants into control and experimental groups (who used EoQu) for the post-test. Results showed evidence of improvement in WR based on scores and time taken to complete the Arabic Word Recognition Test.

Letter connectedness and Arabic visual word recognition.

This study investigates the processing consequences of letter connectedness during Arabic visual word recognition. Specifically, this study examined (a) whether there is a processing cost associated with letter connectedness during word-level reading and (b) whether this factor modulates form-level activation among words during lexical access. Experiment 1 tested one-, two-, and three-chunk Arabic words and nonwords in a lexical decision task with masked identity priming. Experiment 2 tested the same stimuli in a lexical decision task with masked form priming, in which prime-target pairs differed by a letter associated with the morphological root. In both experiments, there was a clear processing cost for letter connectedness-one-chunk words had longer processing times than two-chunk words, which had longer processing times than three-chunk words. Comparable processing time differences were also found for nonwords, suggesting that letter connectedness influences Arabic word recognition at a prelexical orthographic processing stage. Furthermore, although reliable priming was found in both the experiments, there was a suggestion that letter connectedness modulated form priming effects (Experiment 2), with the strongest effect for three-chunk word targets. These findings are taken to indicate that letter connectedness is an important factor that should be considered-and controlled for-in examinations of Arabic visual word recognition.

Arabic Diacritics and Their Role in Facilitating Reading Speed, Accuracy, and Comprehension by English L2 Learners of Arabic

AbstractThe present study investigated the role of Arabic diacritics in word recognition and their impact on Arabic L2 learners’ reading speed, accuracy, and comprehension at different stages of Arabic L2 acquisition. Fifty‐four English L2 learners of Arabic from 3 proficiency levels (beginner, intermediate, and advanced) participated in the study. They belonged to 2 sets of groups: half with exposure to instructional materials containing diacritics, vowelized textbook (VT), and the other half with exposure to instructional materials not containing diacritics, un‐vowelized textbook (UVT). Participants performed word‐list reading, text reading, and target‐word comprehension tasks under vowelized (V) and un‐vowelized (UV) conditions. The results revealed participants in all VT groups did consistently better than their UVT counterparts. This positive role of diacritics in terms of Arabic word recognition and reading performance implies that inclusion of diacritics in words and texts within instructional materials does not only benefit Arabic L2 learners by removing ambiguity from words, but it also positively impacts their reading and pronunciation performance in general. The study findings and pedagogical implications are also relevant to other languages, such as Farsi and Hebrew, which exhibit similar orthographic features.