WAMS | Devpost

Training and Validation Accuracy
Training and Validation Loss
Dataset

Inspiration

Non-biodegradable waste can cause -Pollution -Block drains -Harm to animals -Air pollution WAMS enables you to find if trash is biodegradable and helps you help the environment.

What it does

We have created an application to distinguish between degradable and non-degradable waste, including machine learning, where we have used object detection using the Tensorflow application. Non-biodegradable waste is those that don’t decompose easily by natural agents and hence act as a source of pollution which can further lead to deadly consequences for the environment. Some of the negatives include reduced soil fertility, death of cattle, biomagnification of poisonous chemicals, and destruction of ecological balance.

How we built it

We researched the data to be trained and collected it. The data pre-processing was done on the entire dataset. Subsequently, the model was decided and we finalized on Convolutional Neural Networks.

Challenges we ran into

Researching the dataset was time-consuming as we had to look for images with trash
Integrating the android app with the ML model was a complicated procedure because of time constraint.
CNN was a relatively new concept and we had to learn about it, hence it was tedious

Accomplishments that we're proud of

We are proud of the fact that we developed this application in order to help do our part to save the environment. The project was fun and required every person on the team to commit to it. Challenges were tough but we didn't let that stop us

What we learned

We learned that it is very important to classify trash and to know the importance of distinguishing between degradable and non-degradable waste. Also, we learned a lot through the challenges faced.