Main Content

augment

Apply identical random transformations to multiple images

所有的页面崩溃

Syntax

augI = augment(augmenter,I)

Description

example

augI= augment(augmenter,I)augments imageIusing a random transformation from the set of image preprocessing options defined by image data augmenter,augmenter. IfIconsists of multiple images, thenaugmentapplies an identical transformation to all images.

Examples

collapse all

Open Live Script

Create an image augmenter that rotates images by a random angle. To use a custom range of valid rotation angles, you can specify a function handle when you create the augmenter. This example specifies a function calledmyrange(defined at the end of the example) that selects an angle from within two disjoint intervals.

imageAugmenter = imageDataAugmenter('RandRotation',@myrange);

Read multiple images into the workspace, and display the images.

img1 = imread('peppers.png'); img2 = imread('corn.tif',2); inImg = imtile({img1,img2}); imshow(inImg)

Figure contains an axes object. The axes object contains an object of type image.

Augment the images with identical augmentations. The randomly selected rotation angle is returned in a temporary variable,angle.

outCellArray = augment(imageAugmenter,{img1,img2});
angle = 8.1158

View the augmented images.

outImg = imtile(outCellArray); imshow(outImg);

Figure contains an axes object. The axes object contains an object of type image.

Supporting Function

This example defines themyrangefunction that first randomly selects one of two intervals (-10, 10) and (170, 190) with equal probability. Within the selected interval, the function returns a single random number from a uniform distribution.

functionangle = myrange()ifrandi([0 1],1) a = -10; b = 10;elsea = 170; b = 190;endangle = a + (b-a).*rand(1)end

Input Arguments

collapse all

Augmentation options, specified as animageDataAugmenterobject.

Images to augment, specified as one of the following.

  • Numeric array, representing a single grayscale or color image.

  • Cell array of numeric and categorical images. Images can be different sizes and types.

Output Arguments

collapse all

Augmented images, returned as a numeric array or cell array of numeric and categorical images, consistent with the format of the input imagesI.

Tips

  • You can use theaugmentfunction to preview the transformations applied to sample images.

  • To perform image augmentation during training, create anaugmentedImageDatastoreand specify preprocessing options by using the'DataAugmentation'name-value pair with animageDataAugmenter. The augmented image datastore automatically applies random transformations to the training data.

版本历史

Introduced in R2018b

See Also

|

Topics