The problem InBIN solves

The problems that we are tackling are Improper waste disposal, Unsegregated waste and Over Flowing Dust Bins,
The Existing solutions seem to always focus on the benefit of the nation as a whole but not the personal benefit of the people involved. People will defenitely be more motivated towards the cause if there is some kind of personal benefit for them. Hence we came up with the Incentivised Tech Bin.
It is a reward based system on the right disposal of waste, implemented using a face recogniton system to add reward points to a user account. The reward points can be exchanged for multiple goodies on our website.
The dustbin will have multiple compartments for segregation of waste. A Machine Learning algorithm will segregate the waste into organic and recyclable. Once the type of waste is detected, the waste will go into the right compartment. Along with this, the system will keep track of the amount of waste in the bin and notifications will be sent out accordingly in order avoid waste accumulation for long durations of time. We have also designed a path planning algorithm that will result in efficient waste collection to avoid overflowing dustbins.

Challenges we ran into

The CNN network took too long to train. We had planned to train 25,000 images for 100 epoches to obtain the desrired accuracy but we had to reduce the number of images and number of epoches dude to time constraints.
One of the python packages was not installing in the laptop of the web devoloper. Hence we had to host the facial recognition and deep learning on another system,while having the actual website elsewhere.
One of the systems was windows and we were not experienced with adding exception to the windows firewall hence it took sometime to figure it out.