Skip to main content
This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHub
The Random Crop Images node randomly selects a rectangular section from each input image and crops it to a specified width and height. This is commonly used for data augmentation to create variations of training images. The random position for the crop is determined by a seed value, ensuring the same crop can be reproduced.

Inputs

ParameterData TypeRequiredRangeDescription
imageIMAGEYes-The image to be cropped.
widthINTNo1 - 8192The width of the crop area (default: 512).
heightINTNo1 - 8192The height of the crop area (default: 512).
seedINTNo0 - 18446744073709551615A number used to control the random position of the crop (default: 0).
Note: The width and height parameters must be less than or equal to the dimensions of the input image. If a specified dimension is larger than the image, the crop will be limited to the image’s boundary.

Outputs

Output NameData TypeDescription
imageIMAGEThe resulting image after the random crop has been applied.