imclearborder

Suppress light structures connected to image border

Syntax

J = imclearborder(I)

J = imclearborder(I,conn)

J = imclearborder(___,Name=Value)

Description

J = imclearborder(I) suppresses structures in the image I that are lighter than their surroundings and connected to the image border. Use this function to clear the image border or select image borders. For grayscale images, imclearborder tends to reduce the overall intensity level in addition to suppressing border structures. The output image J is grayscale or binary, depending on the input.

J = imclearborder(I,conn) specifies the pixel connectivity, conn.

J = imclearborder(___,Name=Value) specifies options for border structure selection using one or more name-value arguments. For example, imclearborder(I,Borders=["left" "right"]) removes only the structures touching the left or right border of an image.

Examples

Remove Objects Connected to Image Border

Read a binary image (a postprocessed image of microscopic quartz columnar grains [2]) into the workspace and display it.

originalBW = imread("quartz_columns.png");
imshow(originalBW)

Clear all light objects in the image that are connected to the image border, and display the adjusted image.

BWclearB = imclearborder(originalBW);
imshow(BWclearB)

Remove Dark Border from Image

Read a grayscale image containing an object with a dark border into the workspace, and display it.

I = imread("logo.png");
imshow(I)

Complement the image and remove the border. Display the image.

J = imcomplement(I);
JNoBorder = imclearborder(J);
imshow(JNoBorder)

Complement the image again to return it to the original input image contrast.

INoBorder = imcomplement(JNoBorder);
imshow(INoBorder)

Remove Objects Connected to Select Image Borders

Read a binary image (a postprocessed image of microscopic quartz columnar grains [2]) into the workspace, and display it.

originalBW = imread("quartz_columns.png");
imshow(originalBW)

Remove only the objects which are connected to the top or bottom border of the image.

BWclear2B = imclearborder(originalBW, Borders=["top" "bottom"]);
imshow(BWclear2B)

Impact of Connectivity on Clearing the Border

Create a simple binary image.

BW = [0     0     0     0     0     0     0     0     0
      0     0     0     0     0     0     0     0     0
      0     0     0     0     0     0     0     0     0
      1     0     0     1     1     1     0     0     0
      0     1     0     1     1     1     0     0     0
      0     0     0     1     1     1     0     0     0
      0     0     0     0     0     0     0     0     0
      0     0     0     0     0     0     0     0     0
      0     0     0     0     0     0     0     0     0];

Clear pixels on the border of the image using 4-connectivity. Note that imclearborder does not clear the pixel at (5,2) because, with 4-connectivity, it is not considered connected to the border pixel at (4,1).

BWc1 = imclearborder(BW,Connectivity=4)

BWc1 = 9×9

     0     0     0     0     0     0     0     0     0
     0     0     0     0     0     0     0     0     0
     0     0     0     0     0     0     0     0     0
     0     0     0     1     1     1     0     0     0
     0     1     0     1     1     1     0     0     0
     0     0     0     1     1     1     0     0     0
     0     0     0     0     0     0     0     0     0
     0     0     0     0     0     0     0     0     0
     0     0     0     0     0     0     0     0     0

Now clear pixels on the border of the image using 8-connectivity. imclearborder clears the pixel at (5,2) because, with 8-connectivity, it is considered connected to the border pixel (4,1).

BWc2 = imclearborder(BW,Connectivity=8)

BWc2 = 9×9

     0     0     0     0     0     0     0     0     0
     0     0     0     0     0     0     0     0     0
     0     0     0     0     0     0     0     0     0
     0     0     0     1     1     1     0     0     0
     0     0     0     1     1     1     0     0     0
     0     0     0     1     1     1     0     0     0
     0     0     0     0     0     0     0     0     0
     0     0     0     0     0     0     0     0     0
     0     0     0     0     0     0     0     0     0

Input Arguments

I — Grayscale or binary image

numeric array | logical array

Grayscale or binary image, specified as a numeric or logical array.

Example: I = imread('pout.tif');

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

conn — Pixel connectivity

4 | 8 | 6 | 18 | 26 | 3-by-3-by- ... -by-3 matrix of 0s and 1s

Pixel connectivity, specified as one of the values in this table or a 3-by-3-by- ... -by-3 matrix of 0s and 1s. 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.	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.	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	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	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	Current pixel is center of cube.

For higher dimensions, imclearborder 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.

Note

A pixel on the edge of the input image might not be considered to be a border pixel if you specify a nondefault connectivity. For example, if conn = [0 0 0; 1 1 1; 0 0 0], elements on the first and last row are not considered to be border pixels because, according to that connectivity definition, they are not connected to the region outside the image.

Note

If you specify both the conn argument and the Connectivity name-value argument, then imclearborder sets the connectivity according to Connectivity and ignores the value of conn.

Data Types: double | logical

Name-Value Arguments

Specify optional pairs of arguments as Name1=Value1,...,NameN=ValueN, where Name is the argument name and Value is the corresponding value. Name-value arguments must appear after other arguments, but the order of the pairs does not matter.

Example: imclearborder(I,Borders=["left" "right"]) removes light structures touching the left or right border of an image.

Borders — Image borders to remove structures from

vector of strings | N-by-2 matrix of 0s and 1s

Since R2023b

Image borders to remove structures from, specified as a vector of strings or an N-by-2 matrix of 0s and 1s:

• Vector of strings — Specifies which borders of a 2-D image to remove structures from as any combination of "left", "right", "top", and "bottom". When you specify I as a 2-D image, the default value of Borders is ["left" "right" "top" "bottom"].

• N-by-2 matrix of 0s and 1s—Specifies which borders of an N-dimensional image to remove structures from, where the first element of each row represents the first border in the corresponding dimension and the second element represents the second border in that dimension. For example, if Borders(k,1) is 1, then structures which touch the first border in the k-th dimension are selected. If Borders(k,2) is 1, then structures which touch the second border in the k-th dimension are selected. For example, specifying Borders = [0 0; 1 1; 0 0] is equivalent to specifying Borders = ["left" "right"]. The default value of Borders for N-dimensional images is ones(ndims(I),2), which specifies to remove structures touching all borders of the image.

Connectivity — Pixel connectivity

4 | 8 | 6 | 18 | 26 | 3-by-3-by- ... -by-3 matrix of 0s and 1s

Since R2023b

Pixel connectivity, specified as 4, 8, 6, 18, 26, or a 3-by-3-by- ... -by-3 matrix of 0s and 1s. For more information, see conn.

Data Types: double | logical

Output Arguments

J — Processed image

numeric array | logical array

Processed grayscale or binary image, returned as numeric or logical array, depending on the input image you specify.

Algorithms

imclearborder uses morphological reconstruction where:

• The mask image is the input image.

• The marker image is 0 everywhere except along the border, where it equals the mask image.

References

[1] Soille, Pierre. Morphological Image Analysis: Principles and Applications Berlin ; New York: Springer, 1999, 164–165.

[2] Molnar, Ian. Uniform quartz - Silver nanoparticle injection experiment, Digital Rocks Portal (April 2016). Accessed March 10, 2023. https://www.digitalrocksportal.org/projects/44, made available for documentation use under the ODC-BY 1.0 Attribution License.

Extended Capabilities

C/C++ Code Generation

Generate C and C++ code using MATLAB® Coder™.

Usage notes and limitations:

• imclearborder supports the generation of C code (requires MATLAB^® Coder™). Note that if you choose the generic MATLAB Host Computer target platform, imclearborder generates code that uses a precompiled, platform-specific shared library. Use of a shared library preserves performance optimizations but limits the target platforms for which code can be generated. For more information, see Types of Code Generation Support in Image Processing Toolbox.

• Supports only up to 3-D inputs.

• The name-value argument Connectivity must be a compile-time constant.

• If the Borders value is not in numeric matrix form, then you must specify it as a cell array.

Version History

Introduced before R2006a

R2023b: Improved performance

The imclearborder function shows improved performance. For example, in the code being timed below, the call to imclearborder is about 1.7x faster than in the previous release.

function t = imclearborderTimingTest 
    A = imbinarize(imread("rice.png"));
    f = @() imclearborder(A);
    t = timeit(f);
end

The approximate execution times are:

R2023a: 0.85 ms

R2023b: 0.51 ms

The code was timed on a macOS 12.5.1, Intel^® Core i9 CPU @ 3.6 GHz test system.

R2023b: Option to specify at which borders to remove structures

Specify the borders at which to remove connected structures by using the Borders name value argument. For example, imclearborder(I,Borders=["left" "right"]) removes light structures touching only the left or right image border.