Sailfish Optimizer with Deep Transfer Learning-Enabled Arabic Handwriting Character Recognition

Maray, Mohammed; Al-onazi, Badriyya B.; Alshahrani, Saeed M.; Alotaibi, Najm; Alazwari, Sana; Othman, Mahmoud M.; Hamza, Manar Ahmed

doi:10.32604/cmc.2023.033534

Cited by 4 publications

(3 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…modified the MobileNetV2 model using transfer learning techniques to recognize handwritten Arabic digits and characters. Since the number of character categories was small and the sample size per category was large, the recognition accuracy was high, but it significantly increased the time cost.Maray et al [20]. proposed a novel SFODTL-AHCR model for recognizing handwritten Arabic characters.…”

Section: B Transfer Learning In Character Recognitionmentioning

confidence: 99%

Incremental Recognition of Multi-Style Tibetan Character Based on Transfer Learning

Zhao,

Wang,

Wang

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Tibetan script possesses a distinctive artistic form of writing, intricate glyph structures, and diverse stylistic variations. In the task of text recognition, effectively handling the recognition of Tibetan script with significantly different stylistic fonts remains a challenge. Existing research has made considerable progress in recognizing Tibetan script within a single style using techniques such as convolutional neural networks and convolutional recurrent neural networks. However, when dealing with multi-style Tibetan script recognition, the standard approach involves training models using a multi-label joint training method. This approach annotates the style and class of different font style samples and merges them into a single dataset for model training. Nevertheless, as the amount of data and performance requirements increase, this approach gradually faces issues such as decreasing accuracy, insufficient generalization capability, and poor adaptability to new style samples. In this paper, we propose a transfer learning-based method for incremental recognition of multi-style Tibetan script, referred to as "multi-style Tibetan script incremental recognition." This method learns and completes the recognition of new style Tibetan script gradually from models that already possess recognition capabilities for a specific style. The proposed method consists of the following stages: 1) A multi-style classifier based on the VGG-16 network is designed to extract features that enable precise differentiation between different font styles of Tibetan script; 2) A pre-training phase is conducted using a residual network on the Tibetan script's standard Uchen style, resulting in an efficient feature extractor and a classifier capable of effectively recognizing the standard style, serving as a baseline model; 3) Through transfer learning, the feature extraction part of the baseline model is frozen, forming a shared feature extractor used for feature extraction of Tibetan script in the Uchen variant and antiquarian character styles. Corresponding classifiers are trained based on different styles; 4) By combining the multistyle classifier, shared feature extractor, and category classifier, recognition of Tibetan script characters is achieved. This method incrementally recognizes new styles based on the existing recognition model for Tibetan script in the Uchen standard style. Experimental results demonstrate that this method enhances the recognition accuracy of Tibetan script in Uchen standard, variant, and antiquarian character styles to 99.04%, 98.05%, and 97.66% respectively, using the TCDB and HUTD datasets. Compared to the traditional multitask recognition approach, the overall recognition accuracy improved from 90.14% to 98.40%. This method exhibits high accuracy, strong generalization capability, and good adaptability to new style samples in multistyle Tibetan script character recognition. Furthermore, it can be applied to other tasks involving multi-style, multi-font, and multi-script recognition.

show abstract

Section: B Transfer Learning In Character Recognitionmentioning

confidence: 99%

Incremental Recognition of Multi-Style Tibetan Character Based on Transfer Learning

Zhao,

Wang,

Wang

et al. 2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Then the updated global best solution is sent to the WOA method and therefore, WOA has the absence of exploration problems for selecting process. So, the diversity of the solution is increased, helping to reduce the chance of WOA being trapped in the local optima which also increases the convergence speed [45]. So, the MWOA is proposed to make new WOA techniques to find a better fitness function.…”

Section: 32modified Woa Optimization (Mwoa)mentioning

confidence: 99%

Handwritten character recognition using optimization based skewed line segmentation method and multi-class support vector machine

2023

IJATEE

View full text Add to dashboard Cite

show abstract

“…They use multiple layers of interconnected neurons to Arabic, a Semitic language from the Arabian Peninsula, is spoken by millions, particularly in the Middle East and North Africa. It is one of six UN-recognized official languages and has significant cultural, religious, and historical significance [1,12].…”

Section: Introductionmentioning

confidence: 99%

The Impact of Various Factors on the Convolutional Neural Networks Model on Arabic Handwritten Character Recognition

Alsayed,

Li,

Fat’hAlalim

et al. 2024

ijacsa

View full text Add to dashboard Cite

Recognizing Arabic handwritten characters (AHCR) poses a significant challenge due to the intricate and variable nature of the Arabic script. However, recent advancements in machine learning, particularly through Convolutional Neural Networks (CNNs), have demonstrated promising outcomes in accurately identifying and categorizing these characters. While numerous studies have explored languages like English and Chinese, the Arabic language still requires further research to enhance its compatibility with computer systems. This study investigates the impact of various factors on the CNN model for AHCR, including batch size, filter size, the number of blocks, and the number of convolutional layers within each block. A series of experiments were conducted to determine the optimal model configuration for the AHCD dataset. The most effective model was identified with the following parameters: Batch Size (BS) = 64, Number of Blocks (NB) = 3, Number of Convolution Layers in Block (NC) = 3, and Filter Size (FS) = 64. This model achieved an impressive training accuracy of 98.29% and testing accuracy of 97.87%.

show abstract

Sailfish Optimizer with Deep Transfer Learning-Enabled Arabic Handwriting Character Recognition

Cited by 4 publications

References 18 publications

Incremental Recognition of Multi-Style Tibetan Character Based on Transfer Learning

Incremental Recognition of Multi-Style Tibetan Character Based on Transfer Learning

Handwritten character recognition using optimization based skewed line segmentation method and multi-class support vector machine

The Impact of Various Factors on the Convolutional Neural Networks Model on Arabic Handwritten Character Recognition

Contact Info

Product

Resources

About