Finalist in #ATBDATHATON held on 30 March,2019 in Edmonton, Alberta, CANADA.
We created a real-time computer vision classification model to estimate plant growth stages and did a live demo on stage.
Full Demo: https://www.youtube.com/watch?v=G6xBIVzubFk
Presentation: Crop growth stage modeling and classification
Blog about the competition: https://www.linkedin.com/pulse/computer-vision-technology-project-recognized-finalist-kurrant/
- Download the Aberystwyth Leaf Evaluation Dataset
wget https://zenodo.org/record/168158/files/images_and_annotations.zip?download=1
unzip images_and_annotations.zip
Note: Be careful about the size of the dataset(61GB before extraction).
- Preprocess the data and split it into different growth stages:
python preProcess.py
Note: You will need to edit the pathes. Manually look at the data delete the outliers especially in the '4' folder.
We used tensorflow 1.12. To train run the following code:
python train.py
Note: You will to edit the path to dataset and you might want to increase the batch size.
You can download our trained model from here:
to run the inference code using webcam run the following:
Will be updated