if you see the attached two pictures,then in "the mask" section you can see 2 binary images where white background are 0 and black marked area of object is full of values with 1
now given images like that,i want to calculate the height,width of the object we can see in the mask,note that : "masks can contain very small object,sometimes very large objects too and most of the times shapes of the objects can be different too,i mean object can be found in different area of the mask picture for different image,for all those kind of different shape objects i need to use technique or algorithm that can always find the height and width of those objects using python?how can i do that? a working example with code or pseudo code will be highly appreciated.thanks
[Note : i have seen this post : https://stackoverflow.com/questions/54785782/crop-binary-image-on-object-height-and-width which doesn't answer to my question/problem]
If by length you just mean how many rows or columns in that image contain mask elements this should work:
import numpy as np
def findlenght(matrix,ax):
others=[ a for a in reversed([ k for k in range(len(matrix.shape)) if k != ax])]
matrix=np.where(matrix==0,1,0)
for axis in others: matrix=np.sum(matrix,axis)
LoNgNeSs=np.sum(matrix!=0)
return LoNgNeSs
Will return the length of the 0-mask matrix along axis ax. Of course this is completely insensitive to the actual size: in the top row image it might give the same numbers for a mask taking up the entire lung.
This might help, the problem seems similar:
https://stackoverflow.com/questions/24897092/python-opencv-detect-a-white-object-from-a-binary-image-and-crop-it
Related to your objectives in this query, I would like to suggest you : https://towardsdatascience.com/image-segmentation-using-pythons-scikit-image-module-533a61ecc980Article Skin Lesion Segmentation in Dermoscopic Images with Combinat...
https://stackoverflow.com/questions/59048089/image-segmentation-using-python-and-opencvSEE ALSO: https://www.analyticsvidhya.com/blog/2019/04/introduction-image-segmentation-techniques-python/
Use MATLAB function 'regionprops' which returns various properties like area, centroid, bounding box etc. for binary image regions
https://in.mathworks.com/help/images/ref/regionprops.html
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Programming Languages
- Python