Inspiration
We believe that our hack will help the authorities, business owners, and landlords to be aware of any mishaps like shooting or fire accidents prior, avoiding any severe damage to people or properties.
What it does
It uses weapon detection to detect if anyone is carrying any weapon in the frames and sends alerts via Email to the authorities and the owner.
How we built it
We used openCV, and yolov3 to determine the weapon which includes classes like Gun, Rifle, and Fire.
Challenges we ran into
The amount of time the detector took to train was a lot, and it slowed us down, and due to the time constraint we were not able to refine the dataset or use a bigger dataset due to which the model could not reach its peak accuracy, and sometimes the program would not detect properly because of poor quality video source from cameras.
Accomplishments that we're proud of
We are very proud of what we have learned and implemented during this entire competition with a group of 4 novice programmers.
What we learned
We learned a great deal about working with existing APIs and resources on computer vision and we are excited to witness how the future is going to be with a deeper penetration of computer vision into mainstream markets.
Log in or sign up for Devpost to join the conversation.