-
-
Test data for application
-
Test data for application
-
Test data for application
-
Test data for application
Inspiration
Growing up in a family of farmers, I heard stories of how pests and diseases harmed our crops. Botanists charged a lot for help. So, I created an app to detect plant diseases just by uploading a picture.
What it does
AgroDoc is a comprehensive agricultural app that offers a range of helpful features. It uses CNN image recognition to identify plant diseases from uploaded images and also get appropriate solution for the disease. Additionally, farmers can connect with experts through the expert advice feature for personalized consultations. The AI chatbot provides instant assistance and answers common queries. The community post feature allows farmers to share knowledge, ask questions, and learn from each other's experiences. Overall, AgroDoc serves as a one-stop solution for farmers, offering tools for disease detection, expert guidance, and community support. Note : Currently it detects upto 200 diseases in 4 plants - Cherry ,tomato , potato and pepper
How we built it
My partner, Wasif, and I built it using various frameworks like React Native and Django. We utilized the CNN algorithm to detect plant diseases by uploading an image.
Challenges we ran into
We faced challenges in developing the image recognition feature and making it accurate. Further we faced challenges while merging the database with the frontend as we had never worked with large datasets that were provided during training.
Accomplishments that we're proud of
We're proud to have developed a tool that can help farmers quickly identify and address crop diseases, saving them time and money. Furthermore, our application stood 1st in the internal hackathon held at our college. Additionally, we won the Appreciation award for the best revolutionary application at Anveshna 2023-2024.
What we learned
We learned a lot about image recognition technology and how it can be applied to solve real-world problems in agriculture.
What's next for AgroDoc
Currently, our ML algorithm is 76% accurate; we aim to improve it to at least 85% accuracy. Our plan is to add some extra features to our application, including Market Locator where farmers can sell their crops, Seasonal Guidance where farmers get advice about their crops according to the season, and News related to agriculture.
Log in or sign up for Devpost to join the conversation.