Level 2 Help for REPAIR

INP

 input data set


OUT

 output data set


SIZE

 image size (SL,SS,NL,NS)


SL

 The first line of the input image to be output.


SS

 The first pixel of each input line to be output.


NL

 The number of lines to be output.


NS

 The number of pixels to be output for each line.


MODE

      The program REPAIR operates in three different modes: CORR (the default),
 MV, and ALL. 
      If ALL is specified, no statistical checking is performed, and all 
 regions specified by the AREA, LINESET, BADLINE and MODULO parameters are 
 replaced by interpolating the nearest two lines.
      In the CORRelation mode, a line is considered to be a good line if the
 correlation coefficient between it and the last good line is greater than the
 threshold value specified by CORR. The line is also considered good if its
 correlation with the average of the last good line and the next line is above
 the threshold.
      If MV is specified, two tests are made, in addition to the correlation
 test. The difference in means must be less than the value specified by MEAN,
 and the difference in variances must be less than the value specified by
 VARIANCE. If a line passes all three tests, it is considered good.


AREA

      The AREA, LINESET, BADLINE and MODULO parameters are used to limit the 
 portion of the image in which lines may be replaced. BADLINE is used to 
 identify individual lines to be checked. LINESET is used to specify sets of 
 lines, in the format (SL1,NL1,SL2,NL2,...), when the general region of concern
 is known, but not the exact location. The MODULO parameter is used to test
 and/or replace bad lines that may occur in a periodic fashion.  For example,
 MODULO=(4,20) could be used to test/replace every 20th line, starting with
 Line 4. Unlike the other parameters, MODULO acts as an additional restriction.
 If more than one of AREA, LINESET, and BADLINE are specified, then a line
 is tested/replaced if it satisfies any of the parameters, but if MODULO and
 another parameter is specified, then the line must meet both criteria to be
 tested/replaced. The AREA parameter allows the user to select a portion of 
 the lines to be examined. The values for AREA must be specified in sets of 
 four, (SL1,SS1,NL1,NS1,SL2,SS2,NL2,NS2,...). 
      If the MODE is ALL, then all pixels in the regions specified are replaced 
 by interpolation of the adjacent two lines. 
      If MODE is CORR or MV, these values describe the region from which 
 statistics are gathered. If a line is found to be  bad, the entire line is 
 replaced. 
      AREA, LINESET, and BADLINE accept up to 100 regions, and all three may
 be used in the same task. However, if two of the regions contain a common
 line, then the regions are combined (and perhaps enlarged) into the 
 rectangular region that contains both of them.


LINESET

      The AREA, LINESET, BADLINE and MODULO parameters are used to limit the 
 portion of the image in which lines may be replaced. BADLINE is used to 
 identify individual lines to be checked. LINESET is used to specify sets of 
 lines, in the format (SL1,NL1,SL2,NL2,...), when the general region of concern
 is known, but not the exact location. The MODULO parameter is used to test
 and/or replace bad lines that may occur in a periodic fashion.  For example,
 MODULO=(4,20) could be used to test/replace every 20th line, starting with
 Line 4. Unlike the other parameters, MODULO acts as an additional restriction.
 If more than one of AREA, LINESET, and BADLINE are specified, then a line
 is tested/replaced if it satisfies any of the parameters, but if MODULO and
 another parameter is specified, then the line must meet both criteria to be
 tested/replaced. The AREA parameter allows the user to select a portion of 
 the lines to be examined. The values for AREA must be specified in sets of 
 four, (SL1,SS1,NL1,NS1,SL2,SS2,NL2,NS2,...). 
      If the MODE is ALL, then all pixels in the regions specified are replaced 
 by interpolation of the adjacent two lines. 
      If MODE is CORR or MV, these values describe the region from which 
 statistics are gathered. If a line is found to be  bad, the entire line is 
 replaced. 
      AREA, LINESET, and BADLINE accept up to 100 regions, and all three may
 be used in the same task. However, if two of the regions contain a common
 line, then the regions are combined (and perhaps enlarged) into the 
 rectangular region that contains both of them.


BADLINE

      The AREA, LINESET, BADLINE and MODULO parameters are used to limit the 
 portion of the image in which lines may be replaced. BADLINE is used to 
 identify individual lines to be checked. LINESET is used to specify sets of 
 lines, in the format (SL1,NL1,SL2,NL2,...), when the general region of concern
 is known, but not the exact location. The MODULO parameter is used to test
 and/or replace bad lines that may occur in a periodic fashion.  For example,
 MODULO=(4,20) could be used to test/replace every 20th line, starting with
 Line 4. Unlike the other parameters, MODULO acts as an additional restriction.
 If more than one of AREA, LINESET, and BADLINE are specified, then a line
 is tested/replaced if it satisfies any of the parameters, but if MODULO and
 another parameter is specified, then the line must meet both criteria to be
 tested/replaced. The AREA parameter allows the user to select a portion of 
 the lines to be examined. The values for AREA must be specified in sets of 
 four, (SL1,SS1,NL1,NS1,SL2,SS2,NL2,NS2,...). 
      If the MODE is ALL, then all pixels in the regions specified are replaced 
 by interpolation of the adjacent two lines. 
      If MODE is CORR or MV, these values describe the region from which 
 statistics are gathered. If a line is found to be  bad, the entire line is 
 replaced. 
      AREA, LINESET, and BADLINE accept up to 100 regions, and all three may
 be used in the same task. However, if two of the regions contain a common
 line, then the regions are combined (and perhaps enlarged) into the 
 rectangular region that contains both of them.


MODULO

      The AREA, LINESET, BADLINE and MODULO parameters are used to limit the 
 portion of the image in which lines may be replaced. BADLINE is used to 
 identify individual lines to be checked. LINESET is used to specify sets of 
 lines, in the format (SL1,NL1,SL2,NL2,...), when the general region of concern
 is known, but not the exact location. The MODULO parameter is used to test
 and/or replace bad lines that may occur in a periodic fashion.  For example,
 MODULO=(4,20) could be used to test/replace every 20th line, starting with
 Line 4. Unlike the other parameters, MODULO acts as an additional restriction.
 If more than one of AREA, LINESET, and BADLINE are specified, then a line
 is tested/replaced if it satisfies any of the parameters, but if MODULO and
 another parameter is specified, then the line must meet both criteria to be
 tested/replaced. The AREA parameter allows the user to select a portion of 
 the lines to be examined. The values for AREA must be specified in sets of 
 four, (SL1,SS1,NL1,NS1,SL2,SS2,NL2,NS2,...). 
      If the MODE is ALL, then all pixels in the regions specified are replaced 
 by interpolation of the adjacent two lines. 
      If MODE is CORR or MV, these values describe the region from which 
 statistics are gathered. If a line is found to be  bad, the entire line is 
 replaced. 
      AREA, LINESET, and BADLINE accept up to 100 regions, and all three may
 be used in the same task. However, if two of the regions contain a common
 line, then the regions are combined (and perhaps enlarged) into the 
 rectangular region that contains both of them.


CORR

      The parameter CORR sets the minimum value of the correlation coefficient
 between lines for a line to be considered good. Correlation coefficients may
 range from 1.0 (two lines that match perfectly) to -1.0 (one line perfectly
 matches the complement of the other).


MEAN

       The value of MEAN is the maximum allowed difference between the means
 of two lines for a line to be considered good.


VARIANCE

       The value of VARIANCE is the maximum allowed difference between the 
 variances of two lines for a line to be considered good. The variance is the
 square of the standard deviation.


ZOK

       The keyword ZOK is used when lines that are entirely 0 DN may be
present in the picture, and the user does not want them replaced.