goodfeaturestotrackcv

Finds prominent corners in an image.

Syntax

R = goodfeaturestotrackcv(handle, maxCorners, quality, minDistance...)

R = goodfeaturestotrackcv(handle, maxCorners, quality, minDistance, mask, blockSize, useHDetector, freeParam)

Inputs

handle
Handle of an image.
Type: integer
maxCorners
Maximum prominent corners to detect. If value is less less than 1, all prominent corners will be detected.
Type: integer
quality
Minimal accepted quality of detected corners.
Type: scalar
minDistance
Minimum possible Euclidean distance between the detected corners.
Type: scalar
mask
Optional handle of an 8-bit single channel image or a 2D matrix of natural numbers representing the region of interest where prominent corners are detected in R. If specified it must have the same dimensions as handle.
Type: integer | mat
blockSize
Optional parameter specifying average block size for computation of a derivative covariation matrix in each pixel neighborhood. Defaults to 3.
Type: integer
useHDetector
Optional value set to true if Harris detector should be used. Default value is false.
Type: logical
freeParam
Optional parameter specifying the free parameter of the Harris detector. Default value is 0.04.
Type: scalar

Outputs

R
Handle of the resulting gray scale image showing prominent corners.
Type: integer

Example

Compute prominent corners in an image with default parameters: 

handle = imreadcv('cv4.jpg');
maxCorners = 0;
qualityLevel = 0.01;
minDistance = 10;
R = goodfeaturestotrackcv(handle, maxCorners, qualityLevel, minDistance);
Figure 1. Input image


Figure 2. Output image