Inspiration
My team members and I are passionate about food security and preventing food production loss due to crop diseases. Approximately 220 billion lost or 40% of global food production is lost to crop diseases, leading to decreased access to safe and nutritious food.
What it does
Our app, InfoCrop, uses drone imaging and computer vision to detect 26 different classes of crop diseases. Additionally, the predicted class of the model is used to generate personalized treatment recommendations.
How we built it
We used a ResNet 9 Model and a web scraping algorithm to find webpages with treatments for the predicted disease.
Challenges we ran into
We had trouble installing the QGIS platform to use NDVI on drone images because of conflicting dll files on our laptops.
Accomplishments that we're proud of
We are proud of the web scraping algorithm and app we were able to built despite having no previous coding experience in Swift.
What we learned
We learned about different ResNet models, how to build a web scraping algorithm, and code an app in Swift using XCode.
What's next for InfoCrop
We want to use a drone with NDVI cameras to run our computer vision model with drone images and display the data to our app.
Log in or sign up for Devpost to join the conversation.