Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{183557,
        author = {Spurti and Mahesh Dixit},
        title = {AI Assistant for visually impaired},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {3},
        pages = {1957-1964},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=183557},
        abstract = {Visual impairment ranks among the top ten disabilities globally, with India bearing the highest population of visually impaired individuals. To address this challenge, we propose an AI-assisted framework designed to empower visually impaired users through real-time object detection and recognition, enabling safer and more independent navigation. The system captures images via a camera and processes them using Single Shot Multi-Box Detector (SSD) architecture, leveraging deep neural networks for accurate object identification. The model is trained on the COCO dataset, a comprehensive collection embedded within the Tensor Flow library that encompasses approximately 90% of real-world object categories. To estimate object proximity, depth estimation techniques are integrated, enhancing spatial awareness. The final output is delivered through audio feedback using voice assistance modules, providing intuitive and accessible guidance. The entire system is developed in Python, utilizing its extensive ecosystem of libraries to streamline implementation and reduce code complexity.},
        keywords = {AI Assistance, Visual Impairment, Object Detection, TensorFlow, SSD Architecture, COCO Dataset, Depth Estimation, Voice Feedback, Assistive Technology, Python Programming.},
        month = {August},
        }