Visionbridge: Enabling Independence Through Object, Face and Currency Recognition for Blind

Abstract

Navigating daily life poses significant challenges for blind and visually impaired individuals, particularly in identifying obstacles, recognizing familiar faces, and handling currency transactions. These everyday tasks often require external assistance, leading to a dependency on others and a reduced sense of autonomy. Traditional tools like white canes and guide dogs offer limited functionalities and cannot address the dynamic and complex challenges faced by visually impaired individuals in real-time environments. This study introduces an innovative interface designed to empower blind and visually impaired individuals by enhancing their independence and safety. The proposed system integrates advanced AI powered functionalities such as face detection, obstacle detection, and currency recognition. By utilizing real-time image capturing and processing, the system provides users with immediate, context sensitive audio feedback to assist them in navigating their surroundings and performing essential tasks. The development of such a system is necessary to bridge the gap left by existing assistive technologies, which are often limited in functionality, integration, or affordability. By leveraging advancements in artificial intelligence, computer vision, and wearable technology, the proposed solution addresses critical challenges, fostering greater autonomy and confidence for visually impaired individuals in their daily lives.

Introduction

Visually impaired individuals face significant challenges in daily tasks such as recognizing faces, detecting obstacles, and handling currency, which often require assistance and limit their independence. Traditional aids like white canes and guide dogs primarily support physical navigation but lack real-time environmental awareness and object identification capabilities. This gap reduces autonomy and increases risks.

Recent advances in AI, computer vision, and wearable tech offer promising solutions to integrate multiple assistive functions—such as face recognition, obstacle detection, and currency identification—into a single device. Existing systems mostly focus on isolated functions and suffer from limited adaptability, object differentiation, and cost issues.

The proposed AI-powered system combines deep learning models like YOLO for object detection, Grassmann algorithms for face recognition, and CNNs for currency detection. It processes live video from wearable cameras and provides real-time audio feedback, helping users identify people, avoid obstacles, and manage money independently.

The system features a modular design optimized for embedded devices (e.g., Raspberry Pi) with text-to-speech feedback, aiming to enhance safety and autonomy. It undergoes thorough development and testing, including user trials, to ensure reliability in diverse real-world conditions.

Conclusion

This project introduces a cutting-edge solution designed to empower visually impaired individuals by integrating face detection, obstacle detection, and currency recognition into a single wearable device. By combining these essential functionalities, the system provides real-time support, helping users overcome various challenges in their daily lives and enhancing their independence. The system utilizes advanced artificial intelligence (AI) technologies, including the Grassmann model for face recognition, YOLO (You Only Look Once) for object detection, and Convolutional Neural Networks (CNNs) for currency recognition. Each of these models has been meticulously implemented to ensure accuracy and real-time performance. The audio feedback module plays a critical role in delivering immediate and context-specific auditory information, guiding users through complex environments. This feature ensures that users can respond to obstacles, recognize familiar faces, and identify currency notes without needing visual assistance. One of the major strengths of this project is its focus on usability and practicality. The wearable device is designed with user convenience in mind, offering a lightweight and portable solution that can be easily adopted in everyday life.

References

[1] Javed, Sajid, et al. \"Moving object detection in complex scene using spatiotemporal structured-sparse RPCA.\" IEEE Transactions on Image Processing 28.2 (2018): 1007-1022. [2] Ren, Shaoqing, et al. \"Faster r-cnn: Towards real-time object detection with region proposal networks.\" Advances in neural information processing systems 28 (2015). [3] He, Kaiming, et al. \"Deep residual learning for image recognition.\" Proceedings of the IEEE conference on computer vision and pattern recognition. 2016. [4] Zhang, Han, et al. \"Spda-cnn: Unifying semantic part detection and abstraction for fine-grained recognition.\" Proceedings of the IEEE conference on computer vision and pattern recognition. 2016. [5] Ouyang, Wanli, et al. \"Deepid-net: Deformable deep convolutional neural networks for object detection.\" Proceedings of the IEEE conference on computer vision and pattern recognition. 2015. [6] Elgendy, Mostafa, Cecilia Sik-Lanyi, and Arpad Kelemen. \"Making shopping easy for people with visual impairment using mobile assistive technologies.\" Applied Sciences 9.6 (2019): 1061. [7] Awad, Milios, et al. \"Intelligent eye: A mobile application for assisting blind people.\" 2018 IEEE Middle East and North Africa Communications Conference (MENACOMM). IEEE, 2018.

Copyright

Copyright © 2025 Nandhini E, Nanjundeswari M, Sreehari P, Yokesh V S, Swathi H. This 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.