MUMBAI, India, Aug. 12 -- Intellectual Property India has published a patent application (202621073212 A) filed by Prashant Sheshrao Titare; Mrs. Jaya S Wakode; Harsh Subodh Joshi; Sanvidhan Sanjay Donde; and Viraj Balasaheb Jadhav on June 12, 2026, for Wearable Artificial Vision E-Assistant For Scene Understanding.
Inventors include Prashant Sheshrao Titare; Mrs. Jaya S Wakode; Harsh Subodh Joshi; Sanvidhan Sanjay Donde; and Viraj Balasaheb Jadhav.
The application for the patent was published on August 07, 2026, under issue no. 32/2026.
Abstract: The rapid advancement of artificial intelligence and embedded systems has opened new possibilities for assistive technologies aimed at improving the quality of life for visually impaired individuals. Despite existing solutions, many visually impaired people still face significant challenges in understanding their surroundings, avoiding obstacles, recognizing individuals, reading textual information, and navigating unfamiliar environments. These limitations restrict their independence and mobility in daily life. Therefore, there is a growing need for an integrated, real-time, and user-friendly assistive system that can provide comprehensive environmental awareness through natural interaction methods. In this context, the proposed system introduces a voice controlled, AI powered assistive solution designed specifically for visually impaired users. The system leverages advanced computer vision, natural language processing, and speech technologies to deliver multiple functionalities within a unified framework. It incorporates scene understanding using a vision-language model to generate descriptive insights of the surroundings, obstacle detection using real-time object detection models to ensure user safety, face recognition and identification through embedding-based comparison techniques, and optical character recognition (OCR) for reading textual content from images. Additionally, the system integrates GPS-based navigation using external routing services to guide users toward desired destinations. All functionalities are accessible through voice commands and support multilingual interaction, enhancing usability across diverse user groups. The system architecture consists of a wearable hardware setup, including a camera and processing unit mounted on a cap, coupled with a backend server for model execution. A central AI orchestration model manages tool selection, parallel execution of tasks, and general conversational queries. The system is designed to operate efficiently with low latency while sallowing multiple assistive features to run simultaneously in the background, with user-controlled activation and suspension. Experimental evaluation focuses on real-time responsiveness, detection accuracy, and system usability under practical conditions. The ultimate goal of this project is to develop a scalable, cost-effective, and intelligent assistive solution that enhances autonomy, safety, and accessibility for visually impaired individuals. By integrating multiple AI capabilities into a single cohesive system, this work contributes toward the development of next-generation smart assistive technologies aligned with the broader vision of inclusive and human-centric innovation.
Disclaimer: Curated by HT Syndication.
