Integration of AI Virtual Assistant and Deep Learning for Visually Challenged People
M. Tamilarasi, Rajeshwari Sajith, V S Pragaadeshvar · 2025
The proposed system aims to enhance the lives of individuals with partial vision impairment by introducing a comprehensive web application that seamlessly integrates various cutting-edge technologies. The platform incorporates a user-friendly chatbot, text extraction using Optical Character Recognition (OCR), object detection, and face recognition features. Through the web application, users can interact effortlessly, receiving vital information and assistance tailored to their specific needs. The chatbot serves as a virtual AI assistant, providing real-time support and facilitating intuitive communication. Additionally, the text extraction feature utilizes OCR to convert printed or handwritten text into digital content, which is then audibly announced, ensuring accessible information for users. By leveraging this functionality, the system allows users to comprehend text from images or documents effectively. Leveraging the ESP-32 camera feed, the object detection module identifies objects in the user's environment and audibly reveals the objects, thereby enhancing environmental awareness and alerting users to obstacles or items in their surroundings. Similarly, the face recognition technology, operating through the ESP-32 camera, identifies individuals and announces their names, personalizing interactions and enabling users to recognize people around them. By amalgamating these technologies, the system strives to create an inclusive digital environment, fostering independence and empowerment for individuals with partial vision impairment.