The document proposes a framework for detecting road accidents using video surveillance. It uses bounding box detection and a Gaussian Mixture Model algorithm to identify accidents. When an accident is detected, it will capture an image, determine the location using geopy library, and send notifications to nearby police stations and hospitals. This allows emergency services to respond faster. The proposed system was found to achieve high detection rates and low false alarm rates across different conditions.
2. ABSTRACT
• Computer vision-based accident detection through video surveillance has become a beneficial but
daunting task. In this project, detection of road accidents is proposed.
• The proposed framework capitalizes on axis bounding box technique for accurate object detection
followed by an efficient centroid based GMM algorithm for surveillance footage.
• The proposed framework provides a robust method to achieve a high Detection Rate and a low False Alarm
Rate on general road-traffic CCTV surveillance footage.
• This framework was evaluated on diverse conditions such as broad daylight, low visibility, rain, hail, and
snow using the proposed dataset.
• This framework was found effective and paves the way to the development of general-purpose vehicular
accident detection algorithms in real-time.
• This project also use geopy library to capture the live location and we can send notification to the nearby
police station and hospital with the snap of accident image.
• So by seeing the image they can take necessary resource allocation and the recovery is made very easy in
less time. Alarm buzzer is also included to notify the nearby people.
2
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING
3. INTRODUCTION
• Road-side traffic flow has turned out to be an elementary share of human lifestyles
and has an impact on several services and activities on a day-to-day basis.
• It can be seen that 1.25 million human beings lose their lives due to vehicular
accidents.
• Several human casualties occur due to delays in reporting accident cases in a timely
fashion causing further delays for receiving medical assistance.
• So finding the accident and alerting the nearby hospital play an important role.
• Detection of moving objects, is an important research area in computer vision
research, which is applied to more and more video surveillance systems.
• To obtain the exact outline of the object after the object tracking and processing is
very important, affecting the performance of the whole system. Background
subtraction method is one of the commonly used methods for moving object
detection which helps us to find the moving vehicle. And the angle cordinate
intersection helps us to find the accident.
3
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING
4. EXISTING SYSTEM
Collision detection based on exceeding the threshold of acceleration using an
accelerometer to measure the dynamic force caused by movement and the
gravity force.
The rollover detection based on exceeding the threshold of the vehicle angle of
inclination.
Artificial Neural Networks (ANN), Support Vector Machine (SVM), and Random
Forests (RF)
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING 4
5. DISADVANTAGE
• The occurrence of an accident and the dispatch of emergency medical
services.
• Costly, power- consuming and inefficient.
• Completely depend on the sensors.
• Crash Detection Algorithm
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING 5
6. 6
OBJECTIVE:
• To know the accident by feature extraction from the image and video as input
format.
• Based on the injury as it normal or in serious condition sending an alert message to
nearby hospital.
• Using CCTV surveillance detect the accident.
• It reduce the late communication timing among hospitals and accident spot.
7. PROPOSED SYSTEM
• GAUSSIAN MIXTURE MODEL – GMM for background subtraction.
• Moving vehicle detection and bounding box interaction detection.
• Geopy library to capture the live location
• Smtp for gmail notification
• Send notification to the nearby police station and hospital with the snap of
accident image.
• So by seeing the image they can take necessary resource allocation and the
recovery is made very easy in less time.
• Alarm buzzer is also included to notify the nearby people near to cctv camera.
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING 7
8. GMM
Background Sub
Feature
Extraction - Find
Contour
Set Threshold
Alert
Attach Snapshot
Alert
Hospital/police
station
Geopy share Location
Bounding Box
Accident Detection
No
Yes
10. CONCLUSION
• The proposed solution is implemented on python, using the OpenCV bindings.
• The traffic camera footages from variety of sources are in implementation.
• A simple interface is developed for the user to select the region of interest to be analyzed and then image processing
techniques are applied to detect the accident
• Currently proposed system works with already captured videos but it can be modified to be used for processing live video
streams
• One of the limitations of the system is that it is not efficient at detection of occlusion of the vehicles which affects the
counting as well as classification.
• This problem could be solved by introducing the second level feature classification such as the classification on the bases
of color.
• Another limitation of the current system is that it needs human supervision for defining the region of interest. The user has
to define an imaginary line where centroid of the contours intersects for the counting of vehicles hence the accuracy is
dependent on the judgment of the human supervisor.
• Furthermore the camera angle also affects the system hence camera calibration techniques could be used for the detection
of the lane for the better view of the road and increasing the efficiency.
• The system is not capable of detection of vehicles in the night as it needs the foreground objects to be visible for extraction
of contour properties as well as features for the classification using SIFT features.
10
11. REFERENCES
• T. Rahman, "Road Accidents in Bangladesh: An Alarming Issue", the World Bank, 2012. [Online].
• K. M. Habibullah, A. Alam, S. Saha, A. Amin and A. K. Das, " A Driver-Centric Carpooling: Optimal RouteFinding Model using Heuristic
Multi–Objective Search," 2019 4th International Conference on Computer and Communication Systems (ICCCS), Singapore, 2019.
• K. M. Habibullah, A. Alam, S. Saha, A. Amin and A. K. Das, " A Driver-Centric Carpooling: Optimal RouteFinding Model using Heuristic
Multi–Objective Search," 2019 4th International Conference on Computer and Communication Systems (ICCCS), Singapore, 2019
• M. S. Satu, S. Ahamed, F. Hossain, T. Akter and D. M. Farid, "Mining traffic accident data of N5 national highway in Bangladesh
employing decision trees," 2017 IEEE Region 10 Humanitarian Technology Conference (R10-HTC), Dhaka, 2017, pp. 722-725.
• Nicky Kattukkaran, Arun George and T P. Mithun Haridas, Intelligent accident detection and alert system for emergency medical
assistance, pp. 1-6, 2017.
• Y. Yorozu, M. Hirano, K. Oka and Y. Tagawa, "Electron spectroscopy studies on magneto-optical media and plastic substrate
interface", IEEE Transl. J. Magn. Japan, vol. 2, pp. 740-741, August 1987.
• S. Santos, Guide to NEO-6M GPS Module Arduino. Random Nerd Tutorials, June 2019, [online] Available:
https://randomnerdtutorials.com/guide-to-neo-6m-gps-module-with-arduino/.
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING 11