Back to Aurora Vision Library website

You are here: Start » Function Reference » Computer Vision » Image Segmentation » FindMaxStableExtremalRegions


Header: AVL.h
Namespace: avl
Module: FoundationPro

Segments an image by binarizing it with many different thresholds and by looking which blobs appear "stable".

Applications: Most frequently used for finding correspondence points between two images.


void avl::FindMaxStableExtremalRegions
	const avl::Image& inImage,
	const int inDelta,
	const int inMinArea,
	const int inMaxArea,
	const float inMaxVariation,
	const float inMinDiversity,
	avl::RegionConnectivity::Type inConnectivity,
	atl::Array<avl::Region>& outRegions


Name Type Range Default Description
Input value
inImage const Image& Input image
Input value
inDelta const int 1 - 255 30 Area variance is calculated against ancestor with color difference of delta
Input value
inMinArea const int 0 - 50 Minimum area of stable region
Input value
inMaxArea const int 0 - 2000 Maximum area of stable region
Input value
inMaxVariation const float 0.0 - 0.1f Maximum area variance with containing larger region specified by delta parameter, for region to be considered as stable
Input value
inMinDiversity const float 0.0 - 2.0f Minimum area diversity that region must have in order to be stable when compared to stable regions within it
Input value
inConnectivity RegionConnectivity::Type EightDirections
Output value
outRegions Array<Region>&


For input inImage only pixel formats are supported: 1⨯uint8.

Read more about pixel formats in Image documentation.


List of possible exceptions:

Error type Description
DomainError Not supported inImage pixel format in FindMaxStableExtremalRegions. Supported formats: 1xUInt8.