Ex: INP=A where A is the input image (byte or halfword) in standard VICAR format and of arbitrary size.
Ex: OUT=B where B is the output image. The output image will be the same data format and size as the input image.
SIZE=(sl,ss,nl,ns). is the standard VICAR2 image size field, specifying the area of the input on which processing is to be applied (starting-line, starting-sample, number-of-lines, number-of-samples). The output image size will be nlxns. If defaulted, the entire input image is processed.
Horizontal upper deviation from mean (default=20). Parameters SUTOL,
SLTOL, DSUTOL, and DSLTOL are thresholds applied to the horizontal pixel test:
Let x0 be the pixel to be tested, and let x1 and x2 be its left and right
adjacent neighbors. Let avg=(x1+x2)/2. Then x0 is in error if:
x0 < A or B < x0
where
A = min(avg-SLTOL, x1-DSLTOL, x2-DSLTOL)
B = max(avg+SUTOL, x1+DSUTOL, x2+DSUTOL)
Horizontal lower deviation from mean (default=20). See SUTOL for details.
Horizontal upper deviation from neighbor (default=20). See SUTOL for details.
Horizontal lower deviation from neighbor (default=20). See SUTOL for details.
Vertical upper deviation from mean (default=20). Parameters CUTOL
CLTOL, DCUTOL, and DCLTOL are thresholds applied to the vertical pixel test:
Let x0 be the pixel to be tested, and let x3 and x4 be its upper and lower
adjacent neighbors. Let avg=(x3+x4)/2. Then x0 is in error if:
x0 < C or D < x0
where
C = min(avg-CLTOL, x3-DCLTOL, x4-DCLTOL)
D = max(avg+CUTOL, x3+DCUTOL, x4+DCUTOL)
The thresholds CLTOL, CUTOL, DCLTOL, and DCUTOL are specifiable parameters.
Vertical lower deviation from mean (default=20). See CUTOL for details.
Vertical upper deviation from neighbor (default=20). See CUTOL for details.
Vertical lower deviation from neighbor (default=20). See CUTOL for details.
All eight thresholds (SUTOL,...,DCLTOL) are set to this value.
Ex: DN=n Specifies the DN value for replacement pixels. All pixel errors are replaced by n (rather then the average of adjacent pixels). A negative value is ignored (default of -1), i.e. the average of adjacent pixels is used to replace pixel errors.
This parameter determines number of bits to replace in spikes. The valid range is -1 to 16 (values from 9-16 are meaningful only for HALF format, and are ignored for BYTE). Let x0 be the erroneous pixel, and let avg=(x1+x2)/2 where x1 and x2 are its adjacent neighbors. If BITS>0, then the BITS most significant bits (MSB) of pixel x0 are replaced with the BITS MSBs of avg and the remaining LSBs of x0 are not changed. If BITS=0, then the number of MSBs to be changed when a pixel spike is detected is determined by the mean deviation allowed. If BITS=-1, then the entire pixel DN is changed. (This is equivalent to BITS=8 for BYTE data.) For example (using byte data): If BITS=3, the 3 MSBs of x0 are replaced with the 3 MSBs of avg. If BITS=0, then: If the default tolerances are used (20 DN), then the 4 MSBs will be replaced, because 2**4-1 < 20 < 2**5-1. If all tolerances are 40 DN, then the 3 MSBs will be replaced, because 2**5-1 < 40 < 2**6-1.
Prints all 8 tolerance thresholds. For debugging purposes. Default is NOPRINT.