hdtoi: Apply DTOI transformation to density image

Package: dtoi

Usage

hdtoi input output database

Parameters

input: List of images to be transformed.

output: List of output image names.

database: Name of text database describing HD curve.

fog = "": Value of fog level, read from database if unspecified.

option = "mean": Option for calculating fog density when fog is a file list, can be either "mean" or "median".

sigma = 3.0: If fog is a file name, and option = "mean", the mean fog density is iteratively calculated using this rejection criteria.

floor = 0.0: Value assigned to levels below fog, can be either 0.0 or -1.0.

ceiling = 30000.: The final intensities are scaled to this value, such that a saturated input density equals ceiling on output.

datatype = "r": Datatype of output image pixels.

verbose = yes: Print log of processing to STDOUT.

Description

Task hdtoi transforms one image to another as described by the database. There is only one HD curve per run; the same transformation is applied to all input images.

The fog value can be obtained in three ways: read from the database, read as a floating point number, or calculated from a list of fog images. If parameter fog is not specified, the fog value is read from database. If fog is specified, it can be entered as either a floating point number or as a list of file names. If the value cannot be read as a number, it is assumed to be a file name. In that case, the density of each file in the fog list is calculated and the average of these values is subtracted from input before processing. The algorithm used to calculate the fog density is selected by the option parameter, and is either a "mean" or "median" calculation. The fog density can be the mean value after pixels more than the specified number of sigma have been rejected, or the median value of all the fog spot pixels.

The fog value is subtracted from the input image before the transformation takes place. It is possible that some density values will fall below the fog level; these values are handled in one of two ways. Values below the fog value are set equal to 0.0 when floor = 0.0. If floor = -1.0, the resulting intensity = -1 * intensity (abs (value)).

A scaling factor is applied to the final intensities, as typically they will be < 1.0. The ceiling parameter is used to specify what value a saturated density is transformed to; all intensities are scaled to this upper limit. The precision of the transformation is unaffected by this parameter, although caution must be used if the output image pixel type is an integer. The user is responsible for choosing a ceiling that avoids the truncation of significant digits.

Examples

Convert three density images to intensity images as described in database db1.

cl> hdtoi denin* intim1,intim2,intim3 db1

Time requirements

Task hdtoi requires 20 cpu seconds to transform a 512 square image, with a 12 bit data range, on a VAX 750