A Convolutional Neural Network with TensorFlow and OpenCV using a dataset that contains different leafs of different plants. Repository has the dataset under plant_disease_classification/datasets folder.
- tensorflow
- numpy
- scikit-learn
- opencv-python
- opencv-contrib-python
Call trainer.py with train method with training and tests datasets.
python plant_disease_classification/trainer.py --train plant_disease_classification/datasets/train --val plant_disease_classification/datasets/test --num_classes 38
After successfull training, model is created under plant_disease_classification/ckpts/ folder.
Call classify function of classifier.py with selected image file path. Below there is sample command for CLI for default image path.
python plant_disease_classification/classifier.py --classify
Another version of this project which is implemented using PyTorch and same datasets is available https://github.com/abdullahselek/plant-disease-classification-pytorch.