imageSet

(Not recommended) Define collection of images

The imageSet object is not recommended. Instead, manage collections of image data using an imageDatastore object. For more information, see Compatibility Considerations.

Description

An imageSet object stores information about an image data set or a collection of image data sets. You can use this object to manage your image data. The object contains image descriptions, locations of images, and the number of images in your collection.

Creation

Syntax

imgSet = imageSet(imageLocation)

imgSetVector = imageSet(imgFolder,'recursive')

Description

imgSet = imageSet(imageLocation) creates an imageSet object.

imgSetVector = imageSet(imgFolder,'recursive') returns a vector of image sets found through a recursive search starting from imgFolder. The imgSetVector output is a 1-by-NumFolders vector of imageSet objects, where NumFolders is the number of folders that contain at least one image.

Input Arguments

imageLocation — Image file location

character vector | cell array

Image file location, specified as a character vector or a cell array. The vector must specify the folder name that contains the images. The image files name extensions must be supported by imread. The cell array must contain image locations.

Example: {'imagePath1','imagePath2', ..., 'imagePathX'}, where each imagePath represents the path to an image.

imgFolder — Start recursive image search folder

character vector

Start recursive image search folder, specified as a character vector. The function searches the folder structure recursively, starting from imgFolder.

Properties

Description — Information about image set

character vector

Information about the image set, specified as a character vector. When you create an image set by recursively searching folders or by specifying a single folder location, the Description property is set to the folder name. When you specify individual image files, the Description property is not set. You can set the property manually.

Data Types: char

Count — Number of images in image set

positive integer

Number of images in the image set, specified as a positive integer.

Data Types: double | single

ImageLocation — Image locations

cell array of character vectors

Image locations, specified as a cell array of character vectors.

Data Types: cell

Object Functions

partition	(Not recommended) Divide image set into subsets
read	(Not recommended) Read image at specified index
select	(Not recommended) Select subset of images from image set

Examples

Create an Image Set From a Folder of Images

Read the folder of images.

imgFolder = fullfile(toolboxdir('vision'),'visiondata','stopSignImages');
imgSet = imageSet(imgFolder);

Display the first image in the image set collection.

imshow(read(imgSet,1));

Create an Array of Image Sets from Multiple Folders

Identify the path to the image sets.

imgFolder = fullfile(matlabroot, 'toolbox','vision',...
    'visiondata','imageSets');

Recursively scan the entire image set folder.

imgSets = imageSet(imgFolder,'recursive')

imgSets=1×2 object
  1x2 imageSet array with properties:

    Description
    ImageLocation
    Count

Display the names of the scanned folders.

{imgSets.Description}

ans = 1x2 cell
    {'books'}    {'cups'}

Display 2nd image from the 'cups' folder.

imshow(read(imgSets(2),2));

Create an Image Set by Specifying Individual Images

Specify individual images.

As an alternative to the method below, you can pick the files manually using imgetfile: imgFiles = imgetfile('MultiSelect',true);

imgFiles = { fullfile(matlabroot,'toolbox','vision','visiondata','stopSignImages','image001.jpg'),...
             fullfile(matlabroot,'toolbox','vision','visiondata','stopSignImages','image002.jpg') };

Create image set.

imgSet   = imageSet(imgFiles);

Compatibility Considerations

imageSet is not recommended

Not recommended starting in R2016b

imageSet is not recommended. Use imageDatastore instead, which offers more functionality. There are no plans to remove imageSet.