Search Results (2)

Learning to write the Arabic alphabet is crucial for Arab children’s cognitive development, enhancing their memory and retention skills. However, the lack of Arabic language educational applications may hamper the effectiveness of their learning experience. To bridge this gap, SamAbjd was developed, an interactive web application that leverages deep learning techniques, including air-writing recognition, to teach Arabic letters. SamAbjd was tailored to user needs through extensive surveys conducted with mothers and teachers, and a comprehensive literature review was performed to identify effective teaching methods and models. The development process involved gathering data from three publicly available datasets, culminating in a collection of 31,349 annotated images of handwritten Arabic letters. To enhance the dataset’s quality, data preprocessing techniques were applied, such as image denoising, grayscale conversion, and data augmentation. Two models were experimented with using a convolution neural network (CNN) and Visual Geometry Group (VGG16) to evaluate their effectiveness in recognizing air-written Arabic characters. Among the CNN models tested, the standout performer was a seven-layer model without dropout, which achieved a high testing accuracy of 96.40%. This model also demonstrated impressive precision and F1-score, both around 96.44% and 96.43%, respectively, indicating successful fitting without overfitting. The web application, built using Flask and PyCharm, offers a robust and user-friendly interface. By incorporating deep learning techniques and user feedback, the web application meets educational needs effectively. Full article

Air writing is one of the essential fields that the world is turning to, which can benefit from the world of the metaverse, as well as the ease of communication between humans and machines. The research literature on air writing and its applications shows significant work in English and Chinese, while little research is conducted in other languages, such as Arabic. To fill this gap, we propose a hybrid model that combines feature extraction with deep learning models and then uses machine learning (ML) and optical character recognition (OCR) methods and applies grid and random search optimization algorithms to obtain the best model parameters and outcomes. Several machine learning methods (e.g., neural networks (NNs), random forest (RF), K-nearest neighbours (KNN), and support vector machine (SVM)) are applied to deep features extracted from deep convolutional neural networks (CNNs), such as VGG16, VGG19, and SqueezeNet. Our study uses the AHAWP dataset, which consists of diverse writing styles and hand sign variations, to train and evaluate the models. Prepossessing schemes are applied to improve data quality by reducing bias. Furthermore, OCR character (OCR) methods are integrated into our model to isolate individual letters from continuous air-written gestures and improve recognition results. The results of this study showed that the proposed model achieved the best accuracy of 88.8% using NN with VGG16. Full article