lazysnapping

Segment image into foreground and background using graph-based segmentation

Syntax

BW = lazysnapping(A,L,foremask,backmask)

BW = lazysnapping(A,L,foreind,backind)

BW = lazysnapping(___,Name,Value)

Description

BW = lazysnapping(A,L,foremask,backmask) segments the image A into foreground and background regions using lazy snapping. The label matrix L specifies the subregions of the image. foremask and backmask are masks designating pixels in the image as foreground and background, respectively.

BW = lazysnapping(A,L,foreind,backind) segments the image A into foreground and background regions. foreind and backind specify the linear indices of the pixels in the image marked as foreground and background, respectively.

BW = lazysnapping(___,Name,Value) segments the image or volume using name-value pairs to control aspects of the segmentation.

Examples

Perform Lazy Snapping Using Foreground and Background Masks

Read and display an image.

RGB = imread('peppers.png');
imshow(RGB)

Create a label matrix.

L = superpixels(RGB,500);

Specify a rectangular ROI within the foreground by using the drawrectangle function. The 'Position' name-value pair argument specifies the upper left coordinates, width, and height of the ROI as the 4-element vector [xmin,ymin,width,height]. If you want to draw the rectangle interactively, then omit the 'Position' name-value pair argument.

f = drawrectangle(gca,'Position',[100 128 350 150],'Color','g');

Create a mask that contains the foreground pixels.

foreground = createMask(f,RGB);

Specify background ROIs. To improve the segmentation accuracy, this example specifies two rectangular ROIs in different areas of the background.

b1 = drawrectangle(gca,'Position',[130 30 40 30],'Color','r');
b2 = drawrectangle(gca,'Position',[6 368 500 10],'Color','r');

Create a mask that contains the background pixels. This mask is the union of the two background ROIs.

background = createMask(b1,RGB) + createMask(b2,RGB);

Perform lazy snapping.

BW = lazysnapping(RGB,L,foreground,background);

Visualize the result of the segmentation by highlighting the foreground in green.

imshow(labeloverlay(RGB,BW,'Colormap',[0 1 0]))

Create a masked image in which the background is black.

maskedImage = RGB;
maskedImage(repmat(~BW,[1 1 3])) = 0;
imshow(maskedImage)

Perform Lazy Snapping Using Pixel Indices

Read and display an image.

RGB = imread('peppers.png');
imshow(RGB)

Create a label matrix.

L = superpixels(RGB,500);

Specify the x- and y-coordinates of pixels in the foreground.

foregroundX = [34 114 195 259 392 467 483];
foregroundY = [298 140 135 200 205 283 104];

Convert the coordinates to linear indices. sub2ind takes (row, column) coordinates so specify the input arguments with the y-coordinates before the x-coordinates.

foregroundInd = sub2ind(size(RGB),foregroundY,foregroundX);

Specify the x- and y-coordinates of pixels in the background.

backgroundX = [130 170];
backgroundY = [52 32];

Convert the coordinates to linear indices.

backgroundInd = sub2ind(size(RGB),backgroundY,backgroundX);

Perform lazy snapping.

BW = lazysnapping(RGB,L,foregroundInd,backgroundInd);

Display the segmented mask. Foreground pixels are true and background pixels are false.

imshow(BW)

Display the mask over the original image, highlighting foreground pixels in green.

imshow(labeloverlay(RGB,BW,'Colormap',[0 1 0]))

Segment Volume in Foreground and Background

Load 3-D volumetric image into the workspace.

D = load('mri.mat');
V  = squeeze(D.D);  

Create a 2-D mask identifying initial foreground and background seed points.

 seedLevel = 10;
 fseed = V(:,:,seedLevel) > 75;
 bseed = V(:,:,seedLevel) == 0;
 figure; 
 imshow(fseed)

 figure; 
 imshow(bseed)

Place seed points into empty 3-D mask.

fmask = zeros(size(V));
bmask = fmask;
fmask(:,:,seedLevel) = fseed;
bmask(:,:,seedLevel) = bseed;

Generate a 3-D label matrix.

 L = superpixels3(V,500);

Segment the image into foreground and background using Lazy Snapping.

bw = lazysnapping(V,L,fmask,bmask);

Display the 3-D segmented image.

figure;
p = patch(isosurface(double(bw)));
p.FaceColor = 'red';
p.EdgeColor = 'none';
daspect([1 1 27/128]);
camlight; lighting phong

Input Arguments

A — Image to segment

2-D grayscale image | 2-D truecolor image | 2-D multispectral image | 3-D grayscale volume

Image to segment, specified as a 2-D grayscale, truecolor, or multispectral image or a 3-D grayscale volume. For double and single images, lazysnapping assumes the range of the image to be [0, 1]. For uint16, int16, and uint8 images, lazysnapping assumes the range to be the full range for the given data type. If the values do not match the expected range based on the data type, then scale the image to the expected range or adjust EdgeWeightScaleFactor to improve results.

Data Types: single | double | int16 | uint8 | uint16

L — Label matrix

numeric array

Label matrix of the input image or volume, specified as numeric array. For 2-D grayscale images and 3-D grayscale volumes, the size of L must match the size of the input image A. For color images and multichannel images, L must be a 2-D array where the first two dimensions match the first two dimensions of the input image A.

Do not mark a given subregion of the label matrix as belonging to both the foreground mask and the background mask. If a region of the label matrix contains pixels belonging to both the foreground mask and background mask, lazysnapping segments the region as background.

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

foremask — Mask image that defines foreground

logical array

Mask image that defines the foreground, specified as a logical array. For 2-D grayscale images and 3-D grayscale volumes, the size of foremask must match the size of the input image A. For color images and multichannel images, foremask must be a 2-D array where the first two dimensions match the first two dimensions of the input image A.

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

backmask — Mask image that defines background

logical array

Mask image that defines the background, specified as a logical array. For 2-D grayscale images and 3-D grayscale volumes, the size of backmask must match the size of the input image A. For color images and multichannel images, backmask must be a 2-D array where the first two dimensions match the first two dimensions of the input image A.

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

foreind — Linear index of foreground pixels

numeric vector

Linear index of pixels in the label matrix, specified as a numeric vector.

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

backind — Linear index of background pixels

numeric vector

Linear index of pixels that define the background, specified as a numeric vector.

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

Before R2021a, use commas to separate each name and value, and enclose Name in quotes.

Example: 'Connectivity',6

Connectivity — Connectivity of connected components

8 for 2-D images and 26 for 3-D images (default) | 4 | 6 | 18

Connectivity of connected components, specified as the comma-separated pair consisting of 'Connectivity' and one of the following: 4 or 8, for 2-D images, and 6, 18, or 26 for 3-D images (volumes).

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

EdgeWeightScaleFactor — Scale factor for edge weights

500 (default) | positive number

Scale factor for edge weights between the subregions of the label matrix, specified as the comma-separated pair consisting of 'EdgeWeightScaleFactor' and a positive number. Typical values range from [10, 1000]. Increasing this value increases the likelihood that lazysnapping labels neighboring subregions together as either foreground or background.

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

Output Arguments

BW — Segmented image

logical array

Segmented image, returned as a logical array of the same size as the label matrix, L.

Data Types: logical

Tips

• The lazy snapping algorithm developed by Li et al. clusters foreground and background values using the K-means method. This implementation of the lazy snapping algorithm does not cluster similar foreground or background pixels. To improve performance, reduce the number of pixels with similar values that are identified as foreground or background.

• To obtain masks foremask or backmask interactively, you can draw an ROI on the image then create a mask from the ROI by using the createMask function. For more information, see Create ROI Shapes.

• To obtain pixel indices foreind or backind interactively, you can draw a Polyline ROI object by using the drawpolyline function. Get the x- and y-coordinates of the vertices from the Position property of the Polyline. Finally, convert the coordinates to linear indices by using the sub2ind function. Note that the sub2ind function uses (row, column) coordinates instead of (x, y) coordinates.

References

[1] Y. Li, S. Jian, C. Tang, H. Shum, Lazy Snapping In Proceedings from the 31st International Conference on Computer Graphics and Interactive Techniques, 2004.

Version History

Introduced in R2017a