Skip to content

Latest commit

 

History

History
27 lines (14 loc) · 965 Bytes

File metadata and controls

27 lines (14 loc) · 965 Bytes

Region-Growing

Implementation of Region Growing Algorithm on RGB Colour Image with Parallel Processing.

RegionGrowing.py

Input: Colour Image (1.tif) Manually labelled Image (less accurately segmented) (2.tif)

Output: Precisely segmented Image (out.tif)

Use Function region_growing() to just implement region growing algorithm on a single coloured image.

Methodology:

alt text

Results:

Vadodara city: the road which was neglected in manually labelled image, has detected after applying the unsupervised segmentation, region growing and progressive thresholding techniques.

Sagar city: the urban area which was overestimated has been taken care.

Jabalpur city: the visibility of road has been increased as compare to manually labelled image.

alt text