Smart Glasses for Visually Impaired People

Abstract

The paper represents smart glasses develop to provide support to visually impaired people by giving them real-time awareness of the surrounding, including obstacles and all the surface conditions. Equipped with sensors for object detection, water detection and death estimation, the system provides alerts to user in the form of gentle audio clues whenever there is an obstacle nearby or a slippery water covered area that may cause a fall. The combination of lightweight design, simple voice guidance and low-cost hardware makes the glasses usable in everyday situation. The reliable support while walking, communicating or navigating unfamiliar places. Combining practicality with safety focused technologies these manmade glasses also promised to enhance mobility independence and overall confidence of the visually impaired

Introduction

This paper presents a Smart Glasses Assistance System for Visually Impaired People, designed to improve independent navigation and safety through a combination of sensors, embedded AI, and real-time audio feedback. Traditional mobility aids such as white canes and guide dogs are helpful but have limitations in detecting head-level obstacles, dynamic objects, water hazards, and sudden environmental changes. The proposed system addresses these challenges by integrating multiple sensors into wearable smart glasses.

The system uses an ESP32 microcontroller as the central processing unit and combines a camera module, VL53L0X laser distance sensor, moisture sensor, IR reflectance sensor, and MPU-6050 accelerometer. These components work together to detect obstacles, identify objects, recognize water hazards, monitor surface conditions, and detect falls. A lightweight deep-learning model running directly on the ESP32 performs object recognition offline, eliminating the need for cloud connectivity and reducing response time. Audio feedback is delivered through a speaker or bone-conduction device, providing immediate guidance to the user. The MPU sensor also enables fall detection and SOS message transmission, while gesture-based shake detection can be used to mute or unmute audio alerts.

The methodology focused on understanding the daily challenges faced by visually impaired individuals, including obstacle avoidance, terrain changes, and wet surfaces. A lightweight glasses prototype was developed with integrated sensors and optimized software that fuses sensor data to classify hazards and generate meaningful audio alerts. Extensive laboratory and field testing helped refine sensor placement, alert clarity, and wearing comfort.

The hardware includes an ESP32, camera module, laser distance sensor, water detection sensors, IR reflectance sensor, audio feedback module, potentiometer, MPU-6050 accelerometer, rechargeable battery system, and MicroSD card containing prerecorded object-specific voice messages. The software combines sensor fusion and embedded deep learning for real-time environmental awareness.

Experimental results demonstrate strong performance, with 92% object detection accuracy, obstacle detection up to 3 meters with a response time of 0.3 seconds, and 85% water detection accuracy. The system successfully operates offline while providing real-time obstacle warnings and hazard alerts. User evaluations showed improved navigation accuracy compared to traditional mobility aids, along with positive feedback regarding usability and audio guidance. The compact, lightweight, and low-cost design makes the proposed smart glasses a practical and affordable assistive technology for enhancing the independence, safety, and mobility of visually impaired individuals.

Conclusion

The EDI review successfully presented the feasibility and design of smart classes for visually impaired people in general the project achieves its central object designing a low-cost system that integrates multiple assistive features into one component Portable device specially the smart glasses filled with the ESP32 microcontroller and distance sensor sensors provides essential functionalities offline object detection and obstacle avoidance In short the project offers an economic alternative to costly commercial solutions Field test with visually embed participants produce better navigation accuracy compared to traditional aids and positive remarks regarding the clarity of auditory and tackle response Since the smartglasses works effectively offline cloud services are out of questions therefore the usability is quite strong anywhere The overall design provides such a low cost effective multi feature assisted device is indeed practical Core Functionalities and Hardware Integration The basic functionalities of smart glasses are based on robust and hardware components within asp 32 microcontroller and ultrasound sensors and device sensors serving as the core the system effectively fulfils two most important features which include offline detection of an obstacle and detection of water The incorporation of several feature Intricate functionalities into one compact and lightweight variable and comfortable design was another huge design advantage beside the capability for obstacle avoiding the design also encompasses moisture sensors for water detection and IR reflectance sensors for sensing the surface type Addressing a wide range of real world hazards related to mobility multisensor input is well integrated with one device processing to ensure complete environmental awareness The implementation demonstrated promising performance Matrix crucial for user safety confidence in field test conducted with visually impaired participants the system should gain in navigation accuracy of 85 over state of the art aid User gave positive feedback on the clarity of auditory and tackle response the core technology involves the integration of distance sensor which allows for fast obstacle detection up to three metres with low latency of only 03 seconds This is an extremely important speed for intermediate alerts via auditory feedback in the interest of user safety Offline operation and practical reality Therefore one of the main goal and accomplishment of Design was the elimination of dependency on any external infrastructure The deep learning module runs on the ASP32 micro controller thus the system operates effectively offline such a feature assures strong usability anywhere independent of Internet connectivity and enhances the reliability and day life unity of devices Confirmation of all these features validates the potential of smartglasses to improve mobility independence and self-assurance of people with visually impairments

References

[1] M. A. Addow et al., “A low-cost IoT-based smart home automation system for modern homes in Mogadishu, Somalia,” Springer Journal, 2025. [2] S. Chowdhury, “Enhanced IoT-based smart agro-farm security system with multiple advanced sensors,” ScienceDirect, 2025. [3] “IoT-enabled smart lighting control system with motion detection using PIR technology,” New Horizons of Science, Technology, 2025. [4] M. Alazab, A. Alazab, and I. Khan, “Secure data transmission framework for IoT devices using blockchain and edge computing,” Sensors, vol. 25, no. 4, p. 1892, 2025. [5] H. Rahman and M. Hasan, “An IoT-based home automation and security system using ESP32 and Firebase,” Procedia Computer Science, vol. 232, pp. 135–142, 2025. [6] F. Nawaz and T. Qureshi, “IoT-based smart energy management system using real-time data analytics,” Journal of Ambient Intelligence and Humanized Computing, Springer, 2025. [7] R. Yadav and A. Bhatnagar, “Real-time monitoring and control of home appliances using IoT and mobile application,” International Journal of Emerging Trends in Engineering Research, vol. 13, no. 1, pp. 15–22, 2025. [8] S. S. Nalawade, D. Biradar, R. N. Desai, M. G. Kulkarni, and P. G. Gawande, “Design and implementation of a microcontroller-based temperature control and monitoring system for refrigerated trucks,” International Journal for Multidisciplinary Research, vol. 7, no. 3, pp. 1–8, 2025. [9] H. Manjhi, “IoT-based patient health monitoring system using ESP32 and Blynk app,” International Journal for Research in Applied Science and Engineering Technology, vol. 13, no. 5, pp. 202–209, 2025. [10] C. Karpodini, T. Michailidis, C. Creed, and I. Williams, “Haptic feedback to overcome barriers for visually impaired users in digital audio workstations,” International Journal of Human–Computer Interaction, 2025. [11] J. Feng, G. Hamilton-Fletcher, T. E. Hudson, M. Beheshti, M. Porfiri, and J.-R. Rizzo, “Haptics-based, higher-order sensory substitution designed for object negotiation in blindness and low vision: Virtual whiskers,” Taylor & Francis Online, 2025. [12] J. G. Ramani, P. Yamini, G. Sajeni, G. Vijesh, and M. Tilak, “Gen-AI powered glasses for visually impaired,” International Journal of Innovative Research in Technology, vol. 11, pp. 1–9, 2025. [13] G. Rjoub, S. Islam, J. Bentahar, M. A. Almaiah, and R. Alrawashdeh, “Enhancing IoT intelligence: A transformer-based reinforcement learning methodology,” arXiv preprint, 2024. [14] S. M. Anik, X. Gao, H. Zhong, X. Wang, and N. Meng, “Automation configuration in smart home systems: Challenges and opportunities,” arXiv preprint, 2024. [15] “Efficient and secure IoT-based smart home automation using multisensor, ML, and blockchain,” CMES Journal, 2024. [16] R. Kumar, A. Singh, and P. Verma, “Design and implementation of ESP32-based smart energy monitoring and control system,” IEEE Access, vol. 12, pp. 45781–45793, 2024. [17] V. Sharma, R. Gupta, and D. Jain, “Development of low-cost IoT-based smart irrigation system using soil and climate sensors,” International Journal of Advanced Computer Science and Applications, vol. 15, no. 2, pp. 120–129, 2024. [18] Y. Zhou, X. Li, and L. Chen, “Integration of artificial intelligence in IoT-enabled smart cities: Challenges and future prospects,” IEEE Internet of Things Journal, vol. 11, no. 6, pp. 5784–5797, 2024. [19] K. Patel and S. Mehta, “Optimization of smart home energy consumption using machine learning algorithms,” Sustainable Computing: Informatics and Systems, vol. 43, p. 101028, 2024. [20] J. M. Feghali, C. Feng, A. Majumdar, and W. Y. Ochieng, “High-performance positioning systems for navigation and wayfinding for visually impaired people: A comprehensive review,” Sensors, vol. 24, p. 7020, 2024. [21] R. Singh and H. Lee, “Energy-efficient IoT architectures for remote monitoring,” Sustainable Computing: Informatics and Systems, vol. 38, p. 100876, 2023. [22] S. A. Gholamreza et al., “Haptic feedback devices: A comprehensive review of current technologies and applications,” IEEE Transactions on Haptics, vol. 15, no. 1, pp. 1–15, 2022. [23] S. Patel, H. Park, P. Bonato, L. Chan, and M. Rodgers, “Low-cost IoT-based patient monitoring: A review,” Journal of Medical Systems, vol. 45, no. 3, p. 45, 2021. [24] L. M. Dang, M. J. Piran, D. Han, K. Min, and H. Moon, “A survey on Internet of Things and cloud computing for healthcare,” Electronics, vol. 9, no. 10, p. 1719, 2020. [25] X. Li et al., “Navigation aid for visually impaired people based on AI and AR technologies,” Journal of Enterprise Information Management, vol. 33, no. 3, pp. 170–192, 2020. [26] A. Ullah et al., “An overview of deep learning-based object detection techniques,” IEEE Access, vol. 7, pp. 12345–12354, 2019. [27] G. R. Miah and M. S. Hussain, “A unique smart eye glass for visually impaired people,” International Journal of Computer Science and Network Security, vol. 18, no. 12, pp. 124–129, 2018. [28] M. R. Miah and M. S. Hussain, “Smart cane for blind people using Raspberry Pi and Arduino,” International Journal of Pure and Applied Mathematics, vol. 119, no. 16, pp. 3485–3489, 2018.

Copyright

Copyright © 2026 Ishan Jadhav, Vedant Jadhav, Abhimanyu Ingawale, Minal Deshmukh, Dr. Minal Deshmukh. 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.