PlantCV v3.13.0
PlantCV Version 3.13 Updates
- Update imports to discontinue the deprecation warnings in
pcv.watershed
- Update
scikit-image
requirement toscikit-image>=0.13
- Reorganizes our tutorials in several ways:
- There is now a main tutorials page that is organized as a gallery of tutorial "cards" that can be filtered by keyword tags. Each card has a launch Binder button to access the interactive tutorial and a link to the static tutorial.
- The tutorial card images and links to notebooks are remote and can be hosted from any GitHub (or other) repository.
- The static tutorial pages are now grouped in a directory called "tutorials."
- The static tutorial pages now only have a launch Binder button and render the complete Jupyter notebooks using nbviewer, rather than having a page that recreates the workflow and has a script version of the workflow.
- Added
pcv.transform.gamma_correct
which performs gamma correction on the input image (wrapper of the skimage gamma correction function). - Updated the
debug
method in the backend within more miscellaneous functions. - Expand the functionality of the metadata matcher portion of
plantcv-workflow
to support the matching of multiple metadata values.- Syntax at the command line
(--match id:1,id:2,id:3)
- Also supports lists in configuration file based parallelization
- Syntax at the command line
- Updates plantcv.hyperspectral.read_data to support Band Sequential (BSQ) in addition to Band Interleaved by Line (BIL) raw data formats for ENVI type multi/hyperspectral datasets.
- Adds
pcv.visualize.obj_sizes
function for annotating the sizes of separate objects onto a visualization. - Add
pcv.visualize. obj_size_ecdf
for a new way to visualize: empirical cumulative distribution function (eCDF). - Converted to base python classes
int
andbool
since numpy is deprecatingnp.int
andnp.bool
datatypes. - Update the fill_segments function in the morphology sub-package
- The added observations are corrected.
- Also return the
filled_mask
(which is a label image as an output) along with thefilled_image
as outputs. - The
filled_img
is generate by calling the addedcolorize_label_img
visualization function.