Sign language is the only means of communication for deaf and hearing-disabled people in their communities. It uses body language and gestures, such as hand shapes and facial expressions, to convey a message. It is important to note that sign language is specific to the region; that is, Arabic sign language (ArSL) is different from English sign language. Therefore, this research proposes a way to improve the translation of ArSL using a new artificial intelligence (AI) architecture. Specifically, a convolutional neural network (CNN) based on fine-tuning of the SSD-ResNet50 V1 FPN is applied to build a real-time ArSL recognition and translation system with fast and accurate results. The proposed AI architecture can provide translation of sign language in real-time to enhance communication in the deaf community. We achieved an average F-score of 86% and an average accuracy of 94%.

The use of human damage assessments and expert subjective guilt determinations in modern traffic accident management leads to systemic inefficiencies, which provoke impacted drivers and extend resolution timelines. To overcome these obstacles, this study introduces DeepOntoAccident, an intelligent system that transforms accident reporting by combining semantic reasoning with cutting-edge computer vision. Fundamentally, our system uses an optimized SSD-MobileNet-V2 architecture to identify and categorize vehicle damage into 18 different categories that record location, severity, and impacted components, with remarkable performance (94.7% accuracy, 97.7% mAP). The system’s ontological reasoning architecture, which places visual data within defined traffic norms and environmental variables using SWRL rules, is what truly innovates beyond just recognizing damage. This approach enables culpability reasoning (via a 5-level scale) by integrating contextual factors (weather/environment) and fraudulent driver behaviors, using the identified damage types as evidence. The framework aligns deep learning with ontological reasoning and SWRL rules, producing auditable reasoning chains and nuanced culpability assessments that imitate expert decision-making. The resultant technology addresses major challenges for insurance companies and traffic authorities, such as Saudi Arabia’s Najm services, by cutting processing times from hours to minutes and introducing unprecedented transparency and consistency to accident investigations. Our work establishes a new paradigm for automated accident management, bridging the gap between regulatory-aware reasoning and data-driven pattern recognition while balancing computational efficiency with explainable, legally grounded outcomes.

Deaf and hearing-impaired individuals rely on sign language, a visual communication system using hand shapes, facial expressions, and body gestures. Sign languages vary by region. For example, Arabic Sign Language (ArSL) is notably different from American Sign Language (ASL). This project focuses on creating an Arabic Sign Language Recognition (ArSLR) System tailored for healthcare, aiming to bridge communication gaps resulting from a lack of sign-proficient professionals and limited region-specific technological solutions. Our research addresses limitations in sign language recognition systems by introducing a novel framework centered on ResNet50ViT, a hybrid architecture that synergistically combines ResNet50’s robust local feature extraction with the global contextual modeling of Vision Transformers (ViT). We also explored a tailored Vision Transformer variant (SignViT) for Arabic Sign Language as a comparative model. Our main contribution is the ResNet50ViT model, which significantly outperforms existing approaches, specifically targeting the challenges of capturing sequential hand movements, which traditional CNN-based methods struggle with. We utilized an extensive dataset incorporating both static (36 signs) and dynamic (92 signs) medical signs. Through targeted preprocessing techniques and optimization strategies, we achieved significant performance improvements over conventional approaches. In our experiments, the proposed ResNet50-ViT achieved a remarkable 99.86% accuracy on the ArSL dataset, setting a new state-of-the-art, demonstrating the effectiveness of integrating ResNet50’s hierarchical local feature extraction with Vision Transformer’s global contextual modeling. For comparison, a fine-tuned Vision Transformer (SignViT) attained 98.03% accuracy, confirming the strength of transformer-based approaches but underscoring the clear performance gain enabled by our hybrid architecture. We expect that RAFID will help deaf patients communicate better with healthcare providers without needing human interpreters.