2025· Proceedings of the 1st International Conference on Interdisciplinary Research in Science, Engineering, and Technology· 0 citations· 33 references
TL;DR
An enhanced road safety system that combines motion sensor-based detection with computer vision algorithms to create a more comprehensive hazard alert system and demonstrates significant improvements in detection accuracy.
Abstract
: Road safety remains a critical global concern, with road hazards and sudden braking incidents contributing significantly to accidents. This research introduces an enhanced road safety system that combines motion sensor-based detection with computer vision algorithms to create a more comprehensive hazard alert system. The system employs smartphone sensors including accelerometers, gyroscopes, GPS, and the device camera to detect road hazards, provide lane departure warnings, and alert drivers to potential collision risks. Our application, 'RoadAware' aims to reduce accidents by offering real-time hazard alerts, mapping dangerous road conditions, and providing visual driving assistance through multiple modes including dash-mount computer vision and heads-up display reflection. This paper details the implementation of computer vision models for lane detection and forward collision warning, performance optimization techniques across various devices, and integration with existing motion-based hazard detection. Testing results demonstrate significant improvements in detection accuracy, with pothole identification reaching 93% accuracy and false positive rates for sudden braking detection reduced to 0.5%. The multi-modal approach addresses various driving contexts and environmental conditions, enhancing the system's versatility and effectiveness.
Overspeeding of vehicles causing road accidents is still a major issue in transportation safety in urban and highway environments. Traditional monitoring systems are often based on manual surveillance, expensive infrastructure, or delayed reporting mechanisms that obstruct preventive real-time action. We propose SpeedGuard, a lightweight, real-time, Global Positioning System (GPS)-based overspeed detection and driver alert system using mobile GPS sensors, accelerometer data, and Pythonbased real-time processing techniques. The SpeedGuard system continuously tracks the speed and location of a vehicle using live sensor data from a mobile device, and instant warning alerts are generated when vehicle speed exceeds pre-defined safety thresholds to enhance driver awareness and transportation safety. This system integrates Flask-based communication, real-time monitoring, acceleration and braking analysis, and dashboard alert visualization. Experimental evaluation demonstrates successful live speed monitoring, overspeed detection, sensor event analysis, and lightweight deployment without expensive hardware infrastructure. The proposed SpeedGuard system offers a portable, low-cost, scalable, and efficient solution for intelligent transportation safety and real-time monitoring applications.
Prachi Sahu, Bansod Sneha Bharat, K. Reddy et al.· 2026 International Conferenc...· 0 citations
Nowadays, one of the main reasons for traffic accidents that result in fatalities, serious injuries, and large financial losses is driver fatigue. The challenge for researchers is to develop an accurate fatigue detection system that will maintain an eye on physical conditions and ensure that drivers are paying attention. This study presents a smart and flexible real-time vehicle safety prototype capable of detecting driver fatigue and implementing preventive measures before an accident happens. The suggested approach tracks fatigue indicators using an eye-blink sensor built into spectacles. When the system detects the driver’s tiredness, it swiftly stops the vehicle, triggers a buzzer, and displays a "Danger" notice to alert other drivers and passengers regarding the driver's status. Once the driver is back to normal, the system stops the alarm and permits the vehicle to safely resume movement. So, the accuracy of the system is more than 80%. The percentage of accuracy at night is more than at day light. The purpose of the system is not only to alert the driver and surrounding individuals but also to actively prevent accidents by controlling vehicle under fatigue conditions. This study aids in decreasing traffic accidents, boosting driver alertness, and improving overall road safety.
S. Raju, Md. Abu Bakkar Sikder, Rishikesh Chowdhury· Engineering· 0 citations
Experimental evaluation across varied traffic and lighting conditions confirms reliable accident detection, fast alert dispatch, and consistent forensic report generation, demonstrating the system's potential to shorten emergency response times and streamline post-accident investigation.
Vidya M N, Prajwal Raj V, Dr Manjunath B· International Journal of Adv...· 0 citations
Pre-2018 approaches to traffic accident detection using video surveillance show high detection accuracy with low false alarms, especially in controlled environments like highways and intersections, but challenges remain in real-time implementation due to lighting, occlusion, and camera angle issues.
Nimal Perera, Tharindu Jayasinghe· International Journal of Mod...· 0 citations
Road traffic accidents caused by driver fatigue continue to be a major public safety concern, highlighting the need for intelligent and real-time monitoring systems. This paper presents a Smart Driver Drowsiness Detection and Advanced Emergency Communication System that combines computer vision, deep learning, and automated emergency response to enhance driver safety. The proposed system employs MediaPipe Face Mesh to accurately detect facial landmarks and localize the driver’s eye regions from live video captured through a standard webcam. A Convolutional Neural Network (CNN) is utilized to classify the eye state as open or closed, while an adaptive eyescore mechanism continuously evaluates the driver’s alertness by analyzing consecutive eye-closure patterns. When signs of prolonged drowsiness are detected, the system immediately activates an audible warning to regain the driver’s attention. If the driver fails to respond after multiple warning cycles, the system initiates an automated emergency communication process by obtaining the current GPS location, generating a Google Maps navigation link, transmitting an emergency email to predefined contacts, and placing an emergency phone call using Android Debug Bridge (ADB). Unlike conventional driver monitoring systems that focus solely on fatigue detection, the proposed framework integrates an intelligent emergency response mechanism to provide timely assistance during critical situations. The system is entirely nonintrusive, operates in real time using inexpensive hardware, and does not require wearable sensors, making it practical for deployment in modern vehicles. Experimental evaluation demonstrates that the integration of MediaPipe and CNN enables reliable eyestate recognition with low computational complexity, providing an effective solution for reducing fatigue-related accidents and improving road safety.
Dr M Venkata, Yamuna Chirumalla· International Journal of Eng...· 0 citations
The proposed framework uses the data from the vehicle's accelerometer, gyroscope and Global Positioning System to continuously monitor the specific dynamics of the vehicle, recognizing the abnormal patterns of movement involved in road accidents and shows a high accuracy of detection with a low false-positive rate.
J. Sravanthi, Bolla Bhagya Lakshmi· International Journal for Re...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
