ARCHIVES

Visually challenged individuals face significant challenges in identifying colours during everyday activities, such as selecting clothing, recognizing household objects, and interpreting visual information in their surroundings. Although several assistive technologies have been developed to support navigation and object detection, real-time colour recognition remains a relatively underexplored area. This paper presents an AI-enabled smart spectacle system designed to provide real-time colour detection and audio feedback for visually challenged users. The proposed system integrates a wearable camera, deep learning-based image processing techniques, and a text-to-speech module to identify and communicate colour information effectively. Images captured by the camera are processed using a trained deep learning model to recognize dominant colours and associated objects under varying environmental conditions. The detected colour information is then converted into speech and delivered through a Bluetooth-enabled audio device, allowing users to receive immediate auditory feedback. The system is designed to be lightweight, portable, cost-effective, and suitable for daily use. Experimental results demonstrate reliable colour recognition performance with low response time and improved usability in different lighting environments. The proposed solution enhances user independence and accessibility by enabling accurate colour identification in real-world scenarios. This research contributes to the advancement of intelligent wearable assistive technologies aimed at improving the quality of life of visually challenged individuals