Main Content


Object for storing corner points


This object stores information about feature points detected from a 2-D grayscale image.




points = cornerPoints(location) constructs a cornerPoints object from an M-by-2 array [x y] of location coordinates.

points = cornerPoints(location,Name,Value) sets properties using one or more name-value pairs. Enclose each property name in quotes. For example, points = cornerPoints('Metric',0.0)

Input Arguments

expand all

Location of points, specified as an M-by-2 array of [x y] coordinates.


expand all

Number of points held by the corner points object, specified as a numeric value.

Strength of detected feature, specified as a numeric value. The algorithm uses a determinant of an approximated Hessian.

Object Functions

plotPlot points
isemptyDetermine if points object is empty
lengthNumber of stored points
selectStrongestSelect points with strongest metrics
sizeReturn the size of a points object
selectUniformSelect uniformly distributed subset of feature points
gatherRetrieve cornerPoints from the GPU


collapse all

Read an image.

I = imread('cameraman.tif');

Detect feature points.

points = detectHarrisFeatures(I);

Display the ten strongest points.

strongest = selectStrongest(points,10);
hold on

Figure contains an axes. The axes contains 2 objects of type image, line.

Display the (x,y) coordinates of the strongest points.

ans = 10x2 single matrix

  112.4516  208.4412
  108.6510  228.1681
  136.6969  114.7962
  181.4160  205.9876
  135.5823  123.4529
  100.4951  174.3253
  146.7581   94.7393
  135.2899   92.6485
  129.8439  110.0350
  130.5716   91.0424

Create a checkerboard image.

I = checkerboard(50,2,2);

Load the locations of corner points.

location = [51    51    51   100   100   100   151   151   151; ...
            50   100   150    50   101   150    50   100   150]';

Save the points in a cornerPoints object.

points = cornerPoints(location);

Display the points on the checkerboard.

hold on

Figure contains an axes. The axes contains 2 objects of type image, line.


Although cornerPoints may hold many points, it is a scalar object. Therefore, numel(cornerPoints) always returns 1. This value may differ from length(cornerPoints), which returns the true number of points held by the object.

Extended Capabilities

Introduced in R2012a