AI Weeder is an open-source project aimed at automating weed detection using machine learning techniques. The project leverages image recognition and computer vision to identify and classify weeds in real-time, potentially helping to reduce the manual effort required in agriculture and improve the efficiency of crop management.
- Real-Time Weed Detection: Utilizes a neural network model to segment and detect weeds from plant images.
- High Accuracy: The model is trained to achieve a high degree of accuracy in differentiating between crops and weeds.
- Scalable and Flexible: Can be adapted to different environments and scales, from small farms to larger agricultural operations.
The data used in this project is available at the following link:
This dataset includes images used for training and testing the weed detection model. Make sure to download the data and place it in the appropriate directory before running the code.
To set up the AI Weeder locally, you will need to clone the repository and install the necessary dependencies:
git clone https://github.com/dorwein/ai_weeder.git
cd ai_weeder
pip install -r requirements.txt-
Download the Dataset: Link above ☝️
-
Run the Jupyter Notebooks: Navigate to the
notebooks/directory and open the relevant Jupyter notebooks to preprocess the data, train the model, and evaluate its performance. -
Save the Model: After training, make sure to save the trained model to
models/
Once you have trained the model, you can use the API provided in the ai_weeder_package/api/ directory to make predictions