/DL4BudDetection

Towards Practical 2D Bud Detection With Fully Convolutional Neural Networks

Primary LanguagePython

Deep Learning for 2D Grapevine Bud Detection

Pre-print available at: https://arxiv.org/abs/2008.11872

Data:

  • Classification and segmentation data here

  • Pre-processed data here.

Running training and inference scripts.

  • Install dependencies
bash:~$ pip install -r requirements.txt
  • Download and extract pre-processed data on /images directory or just run the following get_data.py script
bash:~$ python get_data.py
  • Run training scripts for all architectures (sample call shows default values)
bash:~$ python train.py --epochs 200 --lr 0.0001 --bs 4 --savemodel --csv train.csv --imgpath ./images
  • Run inference scripts for a desired model, generating prediction masks for each threshold and optionally (--valid flag) generating a csv component report (call shows sample values, --model must be specified)
bash:~$ python inference.py --model FCMN8rmsprop_lr0.0001_prep_keras_dp0.001_ep200.h5 \ 
--output ./output/validation/ --csv test.csv --imgpath ./images --valid