SMART SENSE NAVIGATION: A REAL-TIME OBSTACLE DETECTION AND DISTANCE ESTIMATION SYSTEM FOR THE VISUALLY IMPAIRED

Read full text Summarise with VIRAT
Cited by: 0
References: 5

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Independent navigation is the major challenge for all the visually challenged individuals due to environmental unawareness.In this paper, we present a mobile-based application called Smart Sense Navigation where we combined privacy integrating real-time object detection and monocular depth estimation.We constructed a framework which employs a lightweight deep learning models which is optimized for Android Devices by TensorFlow Lite.In the obstacle-depth fusion mechanism, the model enables the adaptive risk classification i.e., safe, caution, high-risk based on the threshold levels.The experimental results states that the real-time performance is 18-24 FPS, detection accuracy is 91.3%, and risk classification accuracy is 88.7%.Meanwhile it also maintains low latency and efficient energy consumption.This proposed solution enhances independence, safety, and mobility for visually impaired users.

Fields of Study & Subjects

Assistive Technology in Communication and Mobility Advanced Neural Network Applications Autonomous Vehicle Technology and Safety

Publication details

Journal / source
International Research Journal of Modernization in Engineering Technology and Science
Publication date
26 March 2026
Language
EN