Back to Adaptive Vision Studio website

You are here: Start » Filter Reference » Feature Detection » cvCornerMinEigenVal


Calculates the minimal eigenvalue of gradient matrices for corner detection.

Name Type Range Description
inImage Image Input single-channel UINT8 or REAL image.
inBlockSize Integer 2 - Neighborhood size.
inApertureSize Integer 1 - 31 Aperture parameter for the Sobel operator, may be 3, 5 or 7.
inBorderType CvBorderType Pixel extrapolation method.
outImage Image Output image with minimal eigenvalues.


The operation detects corners within inImage image using the Kanade-Tomasi corner response method. For more information on the method please refer to the DetectCorners_CornerResponse article. outImage is a corner response image of type Real with pixel values from 0 to 255.


cvCornerMinEigenVal performed on the sample image. Output image is normalized.


This filter can throw an exception to report error. Read how to deal with errors in Error Handling.

List of possible exceptions:

Error type Description
DomainError Input image must be single-channel UINT8 or REAL in cvCornerMinEigenVal.

Complexity Level

This filter is available on Basic Complexity Level.

See Also