ImageAugmenterBase<T>

Base Classes ยท AiDotNet.Augmentation.Image

Base class for image data augmentations.

Inheritance: AugmentationBase<T, ImageTensor<T>> โ†’ ImageAugmenterBase<T>

Known derived types: AdaIN<T>, AugMax<T>, AutoAugment<T>, AutoContrast<T>, BiasFieldCorrection<T>, BoxBlur<T>, CLAHE<T>, CenterCropOrPad<T>, ChannelDropout<T>, ChannelShuffle<T>, Closing<T>, Clouds<T>, CoarseDropout<T>, ColorConstancy<T>, CopyPaste<T>, DADA<T>, Defocus<T>, Denormalize<T>, Dilate<T>, Downscale<T>, ElasticTransform<T>, Emboss<T>, Equalize<T>, Erode<T>, FDA<T>, FancyPCA<T>, FastAutoAugment<T>, FiveCrop<T>, Fog<T>, Frost<T>, GammaCorrection<T>, GhostingArtifact<T>, GlassBlur<T>, GrayscaleToRgb<T>, GridDistortion<T>, GridDropout<T>, GridMask<T>, HideAndSeek<T>, HistogramColorTransfer<T>, HistogramEqualization<T>, HistogramMatching<T>, HueSaturationValue<T>, ISONoise<T>, ImageCompression<T>, IntensityNormalization<T>, Invert<T>, JpegCompression<T>, KSpaceMotion<T>, LongestMaxSize<T>, MaskDropout<T>, MedianBlur<T>, MorphologicalGradient<T>, Mosaic9<T>, Mosaic<T>, MotionBlur<T>, MultiplicativeNoise<T>, Normalize<T>, Opening<T>, OpticalDistortion<T>, PadToSquare<T> (+48 more)

For Beginners

Image augmentation transforms images to improve model robustness to variations in viewpoint, lighting, and appearance. Common techniques include:

  • Geometric transforms: flips, rotations, scaling, cropping
  • Color transforms: brightness, contrast, saturation, hue
  • Noise and blur: Gaussian noise, blur, sharpening
  • Regularization: cutout, mixup, cutmix

How It Works

Image data is represented as an ImageTensor with dimensions (height, width, channels).