FitsQuantizationOptions#

pydantic model lsst.images.fits.FitsQuantizationOptions#

Quantization options for FITS compression.

Parameters:

data (Any)

Fields:

dither (lsst.images.fits._common.FitsDitherAlgorithm)
level (float)
seed (int | None)

field dither: FitsDitherAlgorithm [Required]#: How to add random noise during quantization to reduce biases.

field level: float [Required]#

Quantization level.

When positive, this is the fraction of the measured standard deviation that corresponds to an integer step. When negative, it is -ZSCALE, the scaling to apply directly to the original pixels before quantization.

field seed: int | None = None#

Random number seed to use for dithering.

Values between 1 and 10000 (inclusive) are used directly. 0 will generate a value from the current time, and -1 will generate a value from the checksum of the image.

If None, the compression_seed parameter must be passed to FitsOutputArchive.open if any quantized compression is configured.