DATAMANIP commands


PROGRAM NAME

kinverse - Inverse Filtering (Restoration) in Fourier Frequency Domain

DESCRIPTION

kinverse performs restoration of data in the spatial domain by inverse filtering in the frequency domain. The idea is to deconvolve the blurring function from the blurred data. A few problems exist in performing this operation, as described in great detail in any good image processing book.

The major difficulty lies in how small the denominator is allowed to be before you call it garbage. The value chosen is supplied as the -t thresh value. Ideally, this will be just above the noise floor in the image, but experimentation is the best guide here. kinverse places zeros in the output data for those data points where the PSF data object magnitude is below the threshold. In this respect, kinverse is actually an implementation of a pseudoinverse filter.

Note that the input data objects will be treated simply as a field of complex data, presumably obtained by application of an FFT, although the data could be generated by many other means. The output data will be either KCOMPLEX if the input data is any type other than KDCOMPLEX. Otherwise, the output will be of type KDCOMPLEX. Data objects with maps attached are not supported.

REQUIRED ARGUMENTS

-i1
type: infile
desc: source data object (FFT)
-i2
type: infile
desc: PSF data object (FFT)
-o
type: outfile
desc: filtered data object (FFT)

OPTIONAL ARGUMENTS

-t
type: double
desc: filter threshold (magnitude)
default: 1
bounds: value >= 0.0

EXAMPLES

SEE ALSO

RESTRICTIONS

REFERENCES

R.C. Gonzalez, P. Wintz, "Digital Image Processing", Addison-Wesley 1987, ISBN 0-201-11026-1, pp. 221-224.

COPYRIGHT

Copyright (C) 1993 - 1997, Khoral Research, Inc. ("KRI") All rights reserved.