Inspiration
The global population is rising (8.5 billion in 2030, and projected to increase further to 9.7 billion in 2050) and the demands on our food production will only continue to grow. To prevent a possible food shortage from happening, we want to make sure that foods grown are in the most optimal conditions. One currently problem we face today are weeds/invasive plants that take up nutrients, water, fertilizer, and space while growing with food crops. Our project attempts to remove these weeds so there is greater efficiency in crop growth.
What it does
Our software will utilize a RCNN to identify weeds types and remove them.
How we built it
We used a kaggle dataset for weed types, and attempted to train a RCNN model to identify these weeds.
Challenges we ran into
input types for the model training
Accomplishments that we're proud of
Reading through the documentation and attempting to figure out how to resolve training and configuration errors
What we learned
Practiced more with tensorflow and model training
What's next for WeedRemover
expanding from our initial kaggle dataset to other weed types.
Built With
- python
- tensorflow
Log in or sign up for Devpost to join the conversation.