This repo is used to serve as a flow on PlantIT that can conduct the semantic section pre-trained model on IPPN dataset.

Deep plant phenomics is a python package developed and published by Ubbens and Stavness.

A truncated version of deepplantphenomics python package is in the other_module dir for the usage of docker image building recipe.

[] check how to correctly save the output [] build singularity image [] prepare other scripts

• Any image in the specified dir with suffix '_rgb.png' can be read in the model.
• Use -p to specify the input path dir. The -ft will be removed.
• Local testing command (script correctly run but cannot obtain output files):

docker run -v /path/to/image/:/home/ docker://bendjamin101001/plantit_dpp:v1 python3 semantic_section_example.py -p /home/ -ft jpg