grayconnected

R2026b

Select contiguous image region with similar gray values using flood-fill technique

Syntax

BW = grayconnected(I,row,column)

BW = grayconnected(I,row,column,tolerance)

BW = grayconnected(___,Connectivity=4)

Description

BW = grayconnected(I,row,column) finds a connected region of similar intensity in the grayscale image I. Specify the row and column indices of the starting point, the seed pixel. The function returns a binary mask, BW, that indicates which pixels are 8-connected to the seed pixel with a similar intensity.

BW = grayconnected(I,row,column,tolerance) specifies the range of intensity values to include in the mask, as in [(seedvalue-tolerance),(seedvalue+tolerance)].

BW = grayconnected(___,Connectivity=4) returns a mask that indicates which pixels are 4-connected to the seed pixel with a similar intensity (since R2026b). Specify the connectivity as 4 in addition to any combination of input arguments from previous syntaxes.

Examples

Segment Image Using Flood-Fill Technique

Read and display a grayscale image.

I = imread("cameraman.tif");
imageshow(I)

Segment the sky in the image by using the flood-fill technique. Select a pixel in the sky to be the seed location. This example uses the pixel with (row, column) coordinate (50, 50). Call the grayconnected function, specifying the image to be segmented and this seed location.

J = grayconnected(I,50,50);

Display the segmented region in color over the original image by using the labeloverlay function. The segmented region includes sky pixels that are 8-connected to the seed pixel. The region does not include pixels of similar intensity that are disconnected, such as the sky visible between the legs of the tripod.

imageshow(labeloverlay(I,J))

Segment the jacket of the cameraman by using the flood-fill technique. Select a pixel in the jacket to be the seed location. This example specifies the seed pixel with (row, column) coordinate (110, 65). Call the grayconnected function, specifying the image to be segmented and this seed location.

J2 = grayconnected(I,110,65);

Display the segmented image in color over the original image. The segmented region includes all pixels that are 8-connected to the seed pixel. The tripod and the hair of the cameraman have similar intensity to the jacket, so they are included in the segmented region.

imageshow(labeloverlay(I,J2))

Create Binary Mask from Connected Pixels

Create small sample image.

I = uint8([20 22 24 23 25 20 100
           21 19 12 13 12 30 6
           22 11 13 12 24 25 5
           23 13 13 13 24 25 5
           24 27 13 12 12 13 5
           25 26  5 28 29 50 6]);

Specify the row and column indices of the seed location. The value at the seed location is 23.

seedrow = 4;
seedcol = 1;

Specify the tolerance.

tol = 3;

Create mask image, specifying the seed location and tolerance. The mask includes all pixels that are 8-connected to the seed pixel and have a value in the range [20, 26]. The mask excludes pixels with grayscale values within the tolerance range that are not 8-connected, such as the pixel with (row, column) coordinate (3, 6).

BW = grayconnected(I,seedrow,seedcol,tol)

BW = 6×7 logical array

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

Find Connected Region Using 4-Connectivity

Read and display a grayscale image.

I = imread("moon.tif");
imageshow(I)

Segment a crater by using the flood-fill technique. This example uses the pixel with (row, column) coordinate (220, 125). Call the grayconnected function, specifying the image to be segmented and this seed location. To constrain the mask within the crater, decrease the tolerance of intensity values. Specify a connectivity of 4.

tol = 20;
crater = grayconnected(I,220,125,tol,Connectivity=4);

Display the segmented crater in color over the original image.

imshow(labeloverlay(I,crater))

Input Arguments

I — Grayscale image

numeric matrix

Grayscale image, specified as a numeric matrix.

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

row — Row index of seed pixel

positive integer

Row index of seed pixel, specified as a positive integer.

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

column — Column index of seed pixel

positive integer

Column index of seed pixel, specified as a positive integer.

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

tolerance — Tolerance of intensity values

numeric scalar

Tolerance of intensity values to include in the mask, specified as a numeric scalar. The mask includes all pixels with a value in the range [(seedvalue-tolerance),(seedvalue+tolerance)]. By default, the tolerance is 32 for integer-valued images and 0.1 for floating point images.

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

Output Arguments

BW — Binary mask

logical matrix

Binary mask of the connected region, returned as a logical array of the same size as I. All of the foreground pixels indicate image pixels that are connected to the seed pixel with similar intensity.

Data Types: logical

Extended Capabilities

C/C++ Code Generation

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

Usage notes and limitations:

• grayconnected supports the generation of C code (requires MATLAB^® Coder™). Note that if you choose the generic MATLAB Host Computer target platform, grayconnected 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.

GPU Code Generation

Generate CUDA® code for NVIDIA® GPUs using GPU Coder™.

Refer to the usage notes and limitations in the C/C++ Code Generation section. The same usage notes and limitations apply to GPU code generation.

Version History

Introduced in R2015b

R2026b: Specify 4-connectivity

The grayconnected function now supports 4-connectivity regions. To calculate a mask that indicates which pixels are 4-connected to the seed pixel, specify the new name-value argument Connectivity=4.