A Printed Arabic Optical Character Recognition System using Deep Learning

Alghyaline, Salah

doi:10.3844/jcssp.2022.1038.1050

Cited by 5 publications

(3 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is clear from the table that there are many similarities between some Arabic characters. These similarities make recognising handwritten Arabic characters difficult and increase the recognition error rate [27,28]. Moreover, the handwritten character could suffer from rotation, and it is not identical to printed characters due to the different writing styles that people use.…”

Section: Related Work 21 the Arabic Languagementioning

confidence: 99%

Optimised CNN Architectures for Handwritten Arabic Character Recognition

Alghyaline

2024

CMC

View full text Add to dashboard Cite

Handwritten character recognition is considered challenging compared with machine-printed characters due to the different human writing styles. Arabic is morphologically rich, and its characters have a high similarity. The Arabic language includes 28 characters. Each character has up to four shapes according to its location in the word (at the beginning, middle, end, and isolated). This paper proposed 12 CNN architectures for recognizing handwritten Arabic characters. The proposed architectures were derived from the popular CNN architectures, such as VGG, ResNet, and Inception, to make them applicable to recognizing character-size images. The experimental results on three well-known datasets showed that the proposed architectures significantly enhanced the recognition rate compared to the baseline models. The experiments showed that data augmentation improved the models' accuracies on all tested datasets. The proposed model outperformed most of the existing approaches. The best achieved results were 93.05%, 98.30%, and 96.88% on the HIJJA, AHCD, and AIA9K datasets.

show abstract

Section: Related Work 21 the Arabic Languagementioning

confidence: 99%

Optimised CNN Architectures for Handwritten Arabic Character Recognition

Alghyaline

2024

CMC

View full text Add to dashboard Cite

show abstract

“…Based on the latest advancements in detection research (Liu et al, 2020;Krishna et al, 2022;Gai et al, 2023;(Alghyaline, 2022;Ponraj, 2023), the YOLO algorithm holds significance for several reasons.…”

Section: Yolo Detectionmentioning

confidence: 99%

Smart Harvesting Decision System for Date Fruit Based on Fruit Detection and Maturity Analysis Using YOLO and K-Means Segmentation

Ouhda,

Yousra,

Aksasse

2023

Journal of Computer Science

View full text Add to dashboard Cite

The date palm (Phoenix dactylifera) is a large palm with exotic fruits measuring up to 30 meters in height. The date palm produces fruits rich in nutrients provides a multitude of secondary products, and generates income necessary for the survival of a large population. Losses attributed to manual harvesting encompass both quantitative and qualitative aspects, with the latter measured through attributes such as appearance, taste, texture, and nutritional or economic value. These losses, in terms of both quantity and quality, are influenced by practices across all phases of the harvesting process. On the other hand, the risks of work accidents are high because of the length of the date palms. To reduce the losses and reduce risks, it is essential to propose a decision system for robotic harvesting to help farmers overcome the constraints during the harvest. The assessment of quality and maturity levels in various agricultural products is heavily reliant on the crucial attribute of color. In this study, an intelligent harvesting decision system is proposed to estimate the level of maturity based on deep learning, K-means clustering, and color analysis. The decision system's performance is assessed using the dataset of date fruit in the orchard and various metrics. Based on the experimental results, the proposed approach has been deemed effective and the system demonstrates a high level of accuracy. The system can detect, locate, and analyze the maturity stage to make a harvest decision.

show abstract

“…YOLOv4 has been used to train a CNN model for recognizing printed Arabic characters, achieving an accuracy of 82.4% [10]. Other researchers explored methods for developing an OCR system for Sanskrit Manuscripts, using conventional feature extraction, heuristic methods, and machine learning approaches such as CNN, LSTM, or Bidirectional LSTM [11].…”

Section: Introductionmentioning

confidence: 99%

YOLOv5 and U-Net-based Character Detection for Nusantara Script

Prasetiadi,

Saputra,

Kresna

et al. 2023

join

View full text Add to dashboard Cite

Indonesia boasts a diverse range of indigenous scripts, called Nusantara scripts, which encompass Bali, Batak, Bugis, Javanese, Kawi, Kerinci, Lampung, Pallava, Rejang, and Sundanese scripts. However, prevailing character detection techniques predominantly cater to Latin or Chinese scripts. In an extension of our prior work, which concentrated on the classification of script types and character recognition within Nusantara script systems, this study advances our research by integrating object detection techniques, employing the YOLOv5 model, and enhancing performance through the incorporation of the U-Net model to facilitate the pinpointing of fundamental Nusantara script's character locations within input document images. Subsequently, our investigation delves into rearranging these character positions in alignment with the distinctive styles of Nusantara scripts. Experimental results reveal YOLOv5's performance, yielding a loss rate of approximately 0.05 in character location detection. Concurrently, the U-Net model exhibits an accuracy ranging from 75% to 90% for predicting character regions. While YOLOv5 may not achieve flawless detection of all Nusantara scripts, integrating the U-Net model significantly enhances the detection rate by 1.2%.

show abstract

A Printed Arabic Optical Character Recognition System using Deep Learning

Cited by 5 publications

References 48 publications

Optimised CNN Architectures for Handwritten Arabic Character Recognition

Optimised CNN Architectures for Handwritten Arabic Character Recognition

Smart Harvesting Decision System for Date Fruit Based on Fruit Detection and Maturity Analysis Using YOLO and K-Means Segmentation

YOLOv5 and U-Net-based Character Detection for Nusantara Script

Contact Info

Product

Resources

About