Read point data into Napari Format¶

Read points from a file or dictionary into Napari format

plantcv.napari_read_coor(coor, dataformat = 'yx')

returns dictionary of points labeled by class

• Parameters:

  • coor - dictionary object of coordinates, or a path to json datafile with dictionary of point coordinates
  • dataformat - either 'yx', 'xy', or 'sam', Napari takes data as y,x format. If data is 'xy' data is converted from x,y to y,x. If data is 'sam' point data is formatted for input into ultralytics sam3 functions. If 'sam' format is selected the function does expect a dictionary with 'pos' and 'neg' points as labelled classes.

• Context:

  • Import previously labeled points, or points from other functions (e.g. pcvan.napari_save_coor)

• Example use:

  • Below

import plantcv.plantcv as pcv 
import plantcv.annotate as pcvan

# read in data

data = pcvan.napari_read_coor(coor ='coor.json', dataformat = 'xy')

• Example use for training Segment Anything Model:
  • Below

from ultralytics import SAM

model = SAM("sam3.pt")
results = model.predict(source="./Example_image.jpg", 
                        points=data["points"], 
                        labels=data["labels"])
results[0].show()

Source Code: Here