You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We have been talking about building an interactive graphic for object counting.
As we have discussed, the user would mark the objects by clicking on them but the function would also be able to take a list of coordinates and 'pre-mark' the corresponding objects. This would have the advantage of automating as much as possible and have an easy way to correct wrong detection and select difficult objects.
A basic implementation would require:
A function that takes a mask and returns a list of coordinates of the centroids of the objects/blobs.
The function with the interactive graphic. It that takes the original image and optionally the list of coordinates and displays them. Some features:
Click on new objects to add them to the list.
Be able to remove objects.
Export the results on csv?
Support different classes?
Example: In the following picture, most of the grains can be detected automatically. Here, the segmentation was performed using thresholding in the blue channel followed by eliminating objects depending on the eccentricity (favoring circles). The user then would just have to select the few missed grains and correct the mistaken selections.
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
We have been talking about building an interactive graphic for object counting.
As we have discussed, the user would mark the objects by clicking on them but the function would also be able to take a list of coordinates and 'pre-mark' the corresponding objects. This would have the advantage of automating as much as possible and have an easy way to correct wrong detection and select difficult objects.
A basic implementation would require:
Example: In the following picture, most of the grains can be detected automatically. Here, the segmentation was performed using thresholding in the blue channel followed by eliminating objects depending on the eccentricity (favoring circles). The user then would just have to select the few missed grains and correct the mistaken selections.
Beta Was this translation helpful? Give feedback.
All reactions