Inspiration
The motivation behind this project is:
Traffic signals are used to control the vehicles in order to ensure safe and smooth traffic flow as it informs the driver about the lanes to drive in. Knowing traffic rules and following them reduces accidents.
In certain instances the time spent by the vehicles in the traffic is more which can cost a life also. For example, an ambulance needs to reach the hospital without wasting any time at the traffic. This can only be possible when the ambulance is identified and the lane is clear to drive in.
Reducing the amount of time spent by the vehicle at the traffic and identifying the type of vehicle to handle special cases needs traffic extraction at the intersections.
Send the output after processing video to the traffic signal monitors which vary traffic signal.
What it does
The main objectives of this project are:
• To extract data from video surveillance at intersections.
• To identify the vehicles from the extracted data.
• To classify the vehicles identified.
• To estimate the count of each type of vehicle.
How we built it
We went through the articles available in the internet to explore and analyse the solution we have. We captured a traffic video to test the code. The technologies used are -Machine Learning Techniques,OpenCV ,Jupyter Notebook,Python
Challenges we ran into
Since our team is new to machine learning understanding the algorithm took time. We invested time in learning and understanding the ML concepts. While running the collab notebook, the runtime used to get disconnected often which caused delays in writing code
Accomplishments that we're proud of
We are able to identify the vehicles in the video. The video id divided into frames in which vehicles are identified and shown. This is the first step towards adaptive traffic signal system as it will help us to find which lane/road is congested
What we learned
- We are happy to participate in this datathon event and learn more about Machine learning and what and all can be achieved through that. We also learnt more about python3.
- We understood how to collaborate within our team and also reach out when we needed help. This will help us in future hackthons and also build projects in future
What's next for Adaptive Traffic Signal System
- Data Gathering
- Analyzing Data
- Traffic Simulation
- Signal Adaptation Control
- Communication
- Emergency Reaction
- Friendly Interfaces for Users
- Evaluation and Enhancement
- Expandability and Prospectivity
- Compliance with Safety and Regulations
- Public Knowledge