This project aims to analyze the maturity of okra plants using thermal imaging and machine learning. It leverages a pre-trained TensorFlow Lite model to classify the maturity stage of okra plants based on their thermal images.
- Thermal Image Classification: Utilizes a TensorFlow Lite model to categorize okra plant maturity into stages such as "young", "developed", and "average".
- Model Training: The project includes a Python script (
okra_model_trainer.py) for training the machine learning model. - Model Deployment: A TensorFlow Lite model (
image_maturity_model.tflite) is provided for deployment and inference. - Streamlit Web App: A Streamlit web application (
streamlit_site.py) allows for interactive analysis and visualization of thermal images.
-
Install dependencies:
pip install -r requirements.txt
-
Train the model :
python okra_model_trainer.py
-
Run the Streamlit web app:
streamlit run streamlit_site.py
-
Run the inference script with the pre-trained model:
python Run_with_model.py
This will load the pre-trained model (image_maturity_model.tflite) and apply it to the thermal image data located in the Thermal_image directory.
Note: The Thermal_image directory contains image data labeled with maturity stages ("young", "developed", and "average"). These images serve as the input for the model.