fixline - Removes bad data by interpolation.

SYNOPSIS

fixline  [ parameter=value ... ]  [ inputfile ... ]

Parameters are: include_varsauto_selectmax_line_gap,   start_linenum_lines,   start_sample, num_samples.

DESCRIPTION

fixline repairs boxes of bad data by interpolating between lines of good data. Typically, boxes are automatically selected by fixline; e.g., fixline looks for rectangular areas, not including the first or last line, containing bad values, no thicker than max_line_gap.

Optionally, a bad data box can be user-specified, using parameters start_line, num_lines, start_sample, num_samples. User-specified boxes can include the first or last data line.

Interpolation is vertical; a value at (line, sample) coordinates (L, S) is computed using the nearest good data with sample S.

Only 2D variables are fixed; non-2D variables are silently ignored.

There can be one or more input TeraScan datasets. Changes are made to the input datasets in place; therefore, there are no output datasets.

PARAMETERS

include_vars

This is a list of variables to be included in the fixline process. If the list is prefixed with a minus sign, the listed variables are excluded from the line fixing process. Wildcards * and ? are allowed.

There is no default.

auto_select

This parameter determines whether bad data boxes are automatically determined, as opposed to user-specified.

Valid responses are [yes, no]. The default is yes.

max_line_gap

This is the maximum height of automatically selected bad data boxes. This parameter is used only if auto_select=yes.

The valid range is [1, 25]. The default is 1.

start_line

This is the first line of the bad data block to be repaired. This parameter is used only if auto_select=no.

The valid range is [> 0]. There is no default.

num_lines

This is the number of lines to be repaired in the bad data block. This parameter is used only if auto_select=no.

The valid range is [1, 25]. The default is 1.

start_sample

This is the first sample of the bad data block to be repaired. This parameter is used only if auto_select=no.

The valid range is [> 0]. The default is 1.

num_samples

This is the number of samples to be repaired in the bad data block. This parameter is used only if auto_select=no. If this number causes the box to extend beyond the sample dimension of the input dataset the box is truncated to the number of samples in the original dataset.

The valid range is [> 0]. The default is 9999.

EXAMPLES

cabo% fixline
input file     : char(255) ? n11.93195.1103
include_vars   : char(255) ? [] avhrr*
auto_select    : char(  3) ? [yes] no
start_line     : int       ? 11
num_lines      : int       ? [1] 3
start_sample   : int       ? [1]
num_samples    : int       ? [9999]
cabo%


SEE ALSO

avfix, outliers, smear


Last Update: $Date: 1998/05/28 18:00:26 $