Catch Ya

App Demo home page
App Demo complaints page
image to test ML model
ML model filtering
ML model character tracking
ML model character identification
ML model flow for phase 2
Prototype of Smartwatch for phase 3

Inspiration

We saw the lack of crime reporting in India as well as the amount of effort it took the officials to actually identify a crime and then report it. We wanted to think of a solution that would target both of these issues as well as provide common people a means of side earning, so we killed two birds with one stone, and started thinking of a solution by which the users can earn while helping the government officials at the same time

What it does

Catch Ya! reports the crimes by automatically detecting and reporting complaints How?

The user uploads a photo/video depicting the crime that occurred Then the report is submitted to the authorised official for further evaluation. if it is a traffic rule violation then a ML/DL model will look for the HSRP or it will recognize the face in it.

How we built it

We built it with the combined power of ML/DL and IoT

Challenges we ran into

Model training with proper images Creating an accurate flow for the application keeping it simple and fast to report crimes.

Accomplishments that we're proud of

We cracked the concept to attract the user base by providing rewards. F1 score of 96-98% for our ML/DL Model
Created design for the smart band for embedding IoT sensors

What we learned

We learned how to work with Progressive Web Application (PWA) Training and creation of CNN Model and its application How we can use IoT devices in a cyber-physical system to gauge the emotions of the wearer