This function uses watershed segmentation to label individual objects in a series of images.
plantcv.segment_image_series(imgs_paths, masks_paths, rois, save_labels=True, ksize=3)
returns out_labels
-
Parameters:
- imgs_paths = List of paths to the images in the time series. Ordered by time
- masks_paths = List of paths to the masks in the time series. Each mask should correspond to the image in imgs_paths for the same index
- rois = List of roi contours
- save_labels = Optional, saves the labels of each image independently (default: True)
- ksize = int, tuple, or numpy.ndarray. Specifies the size of the block in the time dimension to propagate the labels. If a tuple the first element is used, if array then the first dimension is used (default: 3)
-
Context:
- Used to obtain individual masks for objects, like full plants or leaves, that grow in a time series of images from being separated to touching
Early time point and corresponding mask
Later time point and corresponding mask
from plantcv import plantcv as pcv
# Set global debug behavior to None (default), "print" (to file), or "plot" (Jupyter Notebooks or X11)
pcv.params.debug = "print"
# Using segment image series
analysis_image = pcv.segment_image_series(imgs_paths=paths_to_imgs_list, masks_paths=paths_to_masks_list, rois=valid_rois, save_labels=True, ksize=3)Image Series Segmentation
Source Code: Here