MNFNR

Description

Attempts to remove noise in a single image band using a procedure based on the Maximum Noise Fraction transformation.

Parameters

mnfnr(file, dbic, wlenint, valonly, sampint, noischn)

Name	Type	Caption	Length	Value range
FILE *	str	Input file name	1 -
DBIC	List[int]	Input raster channel(s)	0 -
WLENINT	List[float]	Wavelength interval	0 - 2
VALONLY	str	Valid bands only	0 - 3	YES | NO Default: NO
SAMPINT	List[int]	Sampling interval	0 - 2	1 - Default: 1,1
NOISCHN *	List[int]	Channel containing noisy band	1 - 1	1 -

* Required parameter

Parameter descriptions

FILE

Specifies the name of the image file containing the noisy image channel to be replaced.

DBIC

Specifies the input channel(s) to be used for the noise reduction transformation.

WLENINT

Specifies that the selected bands are to be restricted to those whose center wavelength is either inside or outside a closed interval, specified in nanometers. By default, no restriction is applied.

This parameter has no effect if the input file contains no band center wavelength metadata.

VALONLY

Specifies whether the selected bands are to be restricted to those with "plot" or "bmask" (begin mask) quality values. The default is NO.

This parameter has no effect if the input file contains no band-validity metadata.

SAMPINT

Specifies the X and Y sampling interval within the specified input window.

NOISCHN

Specifies the input channel containing the noisy band to correct.

Details

MNFNR is typically used in a case where a set of image bands contains one band that is considered to have significantly more noise than the others, and it is desirable to transform that band such that its noise content is close to that of the other bands. In this case, it is not necessary to know the noise variance in any band in order to define the MNF transformation.

The application of MNFNR is equivalent to replacing the noisy band (NOISCHN) with the linear combination of the other bands that best approximates the noisy band, in the least-squares sense.

MNFNR performs the same computation whether or not the specified noisy channel (NOISCHN) is included in those specified with DBIC.

MNFNR may be run multiple times to attempt noise removal on multiple bands of a data set. In each run, only one of the noisy bands can be specified as input (NOISCHN and DBIC).

Example

This example demonstrates noise-reduction for channels 107 to 113 of a copy of cuprad.pix. Channels 91 to 106 are selected as the basis of the correction because they visually appear to be highly correlated.

from pci.mnfnr import mnfnr

file	=	"cuprad_copy.pix"
dbic	=	list(range(91, 106))
wlenint	=	[]	# no wavelength interval restriction
valonly	=	""	# no band validity restriction
sampint	=	[]	# no subsampling
noischn	=	[107]

mnfnr( file, dbic, wlenint, valonly, sampint, noischn )

Algorithm

The details of the computation used in program MNFNR may be obtained from Section III.A of the following paper:

Green, A. A., M. Berman, P. Switzer, and M. D. Craig, "A Transformation for Ordering Multispectral Data in Terms of Image Quality with Implications for Noise Removal", IEEE Transactions on Geoscience and Remote Sensing, Vol. 26, No. 1 (1988), pp.65-74.

References

Green, A. A., M. Berman, P. Switzer, and M. D. Craig, "A Transformation for Ordering Multispectral Data in Terms of Image Quality with Implications for Noise Removal", IEEE Transactions on Geoscience and Remote Sensing, Vol. 26, No. 1 (1988), pp.65-74.