Main Content

imhmax

Suppress regional maxima in image using H-maxima transform

Description

example

J = imhmax(I,H) suppresses regional maxima in the grayscale image I by using the H-maxima transform. The H-maxima transform decreases the height of all regional maxima by an amount up to H. As a result, the transform fully suppresses regional maxima whose height is less than H. Regional maxima are connected pixels with the same intensity value, t, that are surrounded by pixels with an intensity value less than t.

J = imhmax(I,H,conn) additionally specifies the connectivity value used to identify the regional maxima in I.

Examples

collapse all

Create a sample 10-by-10 image. Add two regional maxima, each consisting of an area of connected pixels surrounded by lower intensity values.

a = 10*ones(10,10);
a(2:4,2:4) = 13;  
a(6:8,6:8) = 18;

This image is a grayscale representation of the pixel values. The height of each maximum depends on the surrounding pixel values.

Grayscale representation of the original image, with callouts labeling the height of the maxima as 3 and 8.

Apply the H-maxima transform that decreases the height of regional maxima by up to 4.

b = imhmax(a,4);

This image is a grayscale representation of the transformed image. The transform fully suppresses one of the maxima. The transform partially suppresses the taller maximum, and subtracts 4 from the intensity values of the pixels in that maximum.

Grayscale representation of the transformed image.

You can suppress small regional maxima to identify the brightest peaks in an image.

Read a grayscale image of snowflakes into the workspace and display it.

I = imread("snowflakes.png");
imshow(I,InitialMagnification=200)

Find the regional maxima in the image.

regmax = imregionalmax(I);

Display a mask of the regional maxima pixels as an overlay on the original image. The regional maxima correspond to shallow intensity fluctuations, rather than the snowflakes.

overlay = imoverlay(I,regmax,"green");
imshow(overlay,InitialMagnification=200)

Apply the H-maxima transform to remove the shallow intensity peaks. Display the filtered image.

h = 75;
B = imhmax(I,h);
imshow(B,InitialMagnification=200)

Find the regional maxima of the filtered image. Display a mask of the maxima as an overlay on the original image.

regmaxfilt = imregionalmax(B);
overlayfilt = imoverlay(I,regmaxfilt,"green");
imshow(overlayfilt,InitialMagnification=200)

Note that you can alternatively use the imextendedmax function to apply the H-maxima transform and calculate the regional maxima in one step.

Input Arguments

collapse all

Input image, specified as a numeric array of any dimension.

Data Types: single | double | int8 | int16 | int32 | int64 | uint8 | uint16 | uint32 | uint64

H-maxima transform, specified as a nonnegative scalar.

Data Types: single | double | int8 | int16 | int32 | int64 | uint8 | uint16 | uint32 | uint64

Pixel connectivity, specified as one of the values in this table. The default connectivity is 8 for 2-D images, and 26 for 3-D images.

Value

Meaning

Two-Dimensional Connectivities

4

Pixels are connected if their edges touch. The neighborhood of a pixel are the adjacent pixels in the horizontal or vertical direction.

3-by-3 pixel neighborhood with four pixels connected to the center pixel

Current pixel is shown in gray.

8

Pixels are connected if their edges or corners touch. The neighborhood of a pixel are the adjacent pixels in the horizontal, vertical, or diagonal direction.

3-by-3 pixel neighborhood with 8 pixels connected to the center pixel

Current pixel is shown in gray.

Three-Dimensional Connectivities

6

Pixels are connected if their faces touch. The neighborhood of a pixel are the adjacent pixels in:

  • One of these directions: in, out, left, right, up, and down

3-by-3-by-3 pixel neighborhood with 6 pixels connected to the faces of the center pixel

Current pixel is shown in gray.

18

Pixels are connected if their faces or edges touch. The neighborhood of a pixel are the adjacent pixels in:

  • One of these directions: in, out, left, right, up, and down

  • A combination of two directions, such as right-down or in-up

3-by-3-by-3 pixel neighborhood with 6 pixels connected to the faces and 12 pixels connected to the edges of the center pixel

Current pixel is center of cube.

26

Pixels are connected if their faces, edges, or corners touch. The neighborhood of a pixel are the adjacent pixels in:

  • One of these directions: in, out, left, right, up, and down

  • A combination of two directions, such as right-down or in-up

  • A combination of three directions, such as in-right-up or in-left-down

3-by-3-by-3 pixel neighborhood with 6 pixels connected to the faces, 12 pixels connected to the edges, and 8 pixels connected to the corners of the center pixel

Current pixel is center of cube.

For higher dimensions, imhmax uses the default value conndef(ndims(I),"maximal").

Connectivity can also be defined in a more general way for any dimension by specifying a 3-by-3-by- ... -by-3 matrix of 0s and 1s. The 1-valued elements define neighborhood locations relative to the center element of conn. Note that conn must be symmetric about its center element. See Specifying Custom Connectivities for more information.

Data Types: single | double | int8 | int16 | int32 | int64 | uint8 | uint16 | uint32 | uint64

Output Arguments

collapse all

Transformed image, returned as a numeric array of the same size and class as I.

References

[1] Soille, P. Morphological Image Analysis: Principles and Applications. Springer-Verlag, 1999, pp. 170-171.

Extended Capabilities

Version History

Introduced before R2006a