"Image" (Net Decoder)

NetDecoder["Image"]

represents a decoder that converts a rank-3 array of pixel values to a 2D image.

NetDecoder[{"Image","param"val,}]

represents a decoder with specific parameters for postprocessing.

Details

  • NetDecoder[][input] applies the decoder to an input to produce an output.
  • NetDecoder[][{input1,input2,}] applies the decoder to a list of inputs to produce a list of outputs.
  • The input to the decoder input is a rank-3 array of numeric values that lie between 0 and 1.
  • A decoder can be attached to an output port of a net by specifying "port"->NetDecoder[] when constructing the net.
  • Parameters
  • The following parameters are supported:
  • ColorSpace"RGB"color space to use for encoding
    InterleavingFalsewhether channels are interleaved
    "MeanImage"Nonemean image to subtract
    "VarianceImage"Nonevariance image to normalize by
  • The parameter "ColorSpace" can take the same values as the ColorSpace option.
  • The input of the decoder is a rank-3 array of dimensions {c,height,width} when InterleavingFalse and {height,width,c} when InterleavingTrue, where c is the number of color channels.
  • The following settings can be used for "MeanImage" and "VarianceImage":
  • Nonedo not subtract anything
    mvalue to subtract from each channel of each pixel
    {m1,m2,}values to be subtracted from different channels
    Image3D[]overall image to subtract
  • Pixels are normalized to lie between 0 and 1 before doing subtraction.
  • Properties
  • NetDecoder[][data,prop] can be used to calculate a specific property for the input data.
  • When an "Image" decoder is attached to a net, net[data,prop] or net[data,"oport"->prop] can be used to calculate a specific property of the decoded output.
  • The "Image" decoder only supports the bypass property. Setting prop to None bypasses decoding and returns the input to the decoder.

Examples

open allclose all

Basic Examples  (1)

Create an RGB image decoder:

Use it to decode a three-channel matrix as an RGB image:

Scope  (1)

Create a grayscale image decoder:

Use it to decode a single-channel matrix as a grayscale image:

The decoder maps across a batch of examples: