Abstract
Imagine being stuck in a traffic jam, it’s pretty aggravating and all of a sudden you hear a siren of an ambulance and everyone around us start making space for the ambulance to pass and relived to reach its destination, but it is unfeasible in every circumstance. Ambulance services are one of the mainly affected services in traffic jams. Let’s talk about few statistics, according to a report published by Times of India about 146,133 people were killed in road accidents in India in the year 2016. Unfortunately, about 30% of deaths are caused due to delayed ambulance. Another Indian government data shows more than 50% of heart attack cases reach hospital late. Providing prompt and effective emergency health services by the ambulances is a challenge in country like India. So, the main core is to make sure that ambulances should be prime concerned during traffic jams and help them reach their destined hospitals instantaneously. The exercise behind this is whenever the ambulance finds a nearby traffic signal within the radius of 500 metres, it will immediately send the request to the control room for checking the current status of the signal in which the ambulance has to cross. What it takes is that mainly we need to train an SVM classifier which is a machine learning module on extracted features to differentiate between Vehicle and Non-Vehicle objects. For feature extraction to have a robust feature set and to increase accuracy level we will be using Histogram of Oriented Gradients (HOG). Using this we should be able to get further details of vehicles like their colour. Scikit-image python library provides us with the necessary API for calculating HOG feature. Once we have the prediction model, it’s time to use it on our test images. Prediction model will be applied in a special technique called Sliding Windows. By Eliminating false positives, we should be able to identify the Ambulance. Now we improve traffic clearance by also using street lights for detecting the approach of ambulance by placing lightweight detector CSL-YOLO on street lights and train Machine Learning module to identify the ambulance and share its location with control room to intimate its approach. Now when it is within 500 meters, we display ambulance symbol onto traffic lights and instruct the citizens to move towards the right allowing the free space on the left for the ambulance to reach the destination. By this we can conclude that this development will help in management of services and reduces risk of life.
References
1. K. Simonyan and A. Zisserman, “Very deep convolution networks for large-scale image recognition,” arXiv preprintarXiv: 1409.1556, 2014.
2. Yolo Algorithm Somya Goel, A. Baghel, Aprajita Srivastava, Aayushi Tyagi, P. Nagrath ,computer Science ,Advances in Intelligent Systems and Computing 2019.
3. J. Vignesh, E. Gajendran, Sameeruddin Khan, N. Balakumar and S. R. Boselin Prabhu, “Design and implementation of hybrid cascaded energy Efficient Kogge Stone Adder” ARPN Journal of Engineering and Applied Sciences Vol 12 Issue no 21 2017.
4. Greeshma C , Nidhindas K , Parvathi Kishore P, Sreejith P, “Traffic Control using Computer Vision” International Journal of Advanced Research in Computer and Communication EngineeringVol 8 Issue no 4 2019.
5. An object detection system based on YOLO in traffic scene Jing Tao, Hongbo Wang, Xinyu Zhang, Xiaoyu Li, Hua-wei Yang, Computer Science.
6. https://arxiv.org/pdf/2108.12118.pdf
7. https://www.youtube.com/watch?v=XoMiveY_1Z4 [How to train the ML module using python]
8. https://www.youtube.com/watch?v=t65xGPpDVCw [how recognition is executed in YOLO
9. https://www.youtube.com/watch?v=XRVzuV9RexY [Preparing dataset for custom YOLO v3 object detector].
10. Viswanathan, A., Arunachalam, V. P., & Karthik, S. (2012). Geographical division traceback for distributed denial of service. Journal of Computer Science, 8(2), 216.
11. Anurekha, R., K. Duraiswamy, A. Viswanathan, V.P. Arunachalam and K.G. Kumar et al., 2012. Dynamic
approach to defend against distributed denial of service attacks using an adaptive spin lock rate control mechanism. J. Comput. Sci., 8: 632-636.
12. Umamaheswari, M., & Rengarajan, N. (2020). Intelligent exhaustion rate and stability control on underwater wsn with fuzzy based clustering for efficient cost management strategies. Information Systems and e-Business Management, 18(3), 283-294.
13. Babu, G., & Maheswari, M. U. (2014). Bandwidth Scheduling for Content Delivery in VANET. International Journal of Innovative Research in Computer and Communication Engineering IJIRCCE, 2(1), 1000-1007.
14. Viswanathan, A., Kannan, A. R., & Kumar, K. G. (2010). A Dynamic Approach to defend against anonymous DDoS flooding Attacks. International Journal of Computer Science & Information Security.
15. Kalaivani, R., & Viswanathan, A. HYBRID CLOUD SERVICE COMPOSITION MECHANISM WITH SECURITY AND PRIVACY FOR BIG DATA PROCESS., International Journal of Advanced Research in Biology Engineering Science and Technology, Vol. 2,Special Issue 10, ISSN 2395-695X.
16. Ardra, S., & Viswanathan, A. (2012). A Survey On Detection And Mitigation Of Misbehavior In Disruption Tolerant Networks. IRACST-International Journal of Computer Networks and Wireless Communications (IJCNWC), 2(6).