AIPS HELP file for MAXFIT in 31DEC24
As of Thu Apr 25 20:23:13 2024
MAXFIT: Verb to find coordinates and value of an image extremum
INPUTS
INNAME Image name (name)
INCLASS Image name (class)
INSEQ 0.0 9999.0 Image name (seq. #)
INDISK 0.0 9.0 Disk drive #
PIXXY 0.0 4096.0 Approximate (X,Y,...) pixel
of maximum
IMSIZE 3.0 16.0 Search area for peak value.
DOINVERS -1.0 1.0 > 0 fit peak in negative of
the image
PRTLEV -2.0 10.0 <= -1 => no display results
@ Output adverbs
PIXXY @ Fitted (X,Y,...) pixel of
@ maximum.
PIXVAL @ Peak intensity at maximum
COORDINA @ Coordinates at pixel
FSHIFT @ RA/Dec shift to put on
nearest pixel
ERROR @ > 0 => fit failed, else okay
HELP SECTION
MAXFIT
Type: Verb
Use: MAXFIT fits a quadratic function to a 3x3 map array to determine
the position and strength of an maximum in the image values or
the negative of the image values. It uses PIXXY as an estimate
of the center of the 3x3 array, but searches over an area of up
to 16x16 pixels for the actual peak value before fitting.
Tasks IMFIT and JMFIT do a more elaborate fitting using
Gaussians plus a base level (constant, sloped, or curved).
Adverbs: (inputs)
INNAME......Image name(name). Standard defaults.
INCLASS.....Image name(class). Standard defaults.
INSEQ.......Image name(seq. #). 0 => highest.
INDISK......Disk drive # of image. 0 => any.
PIXXY.......The approximate (X,Y,...) pixel of the maximum.
IMSIZE......(X,Y) search area in pixels. Verb uses 11 if the values
are outside the range 3 to 16.
DOINVERS....<= 0 -> do the fit in the image values, thus a positive
going peak even in an all negative image will be
found and fit
> 0 -> do the fit in the negative of the image values;
this is needed for a negative going bump.
PRTLEV......Print level: if <= -1, then do not put result info in
message file, otherwise do the coordinate and flux
displays.
Adverbs: (outputs)
PIXXY.......Fitted (X,Y,...) position of peak.
PIXVAL......Fitted intensity at the peak.
COORDINA....The X and Y coordinates are found as:
Xpos = abs(CO(1)) + abs(CO(2))/60 + abs(CO(3))/3600
if any of CO(1), CO(2), CO(3) < 0: Xpos = -Xpos
Ypos = abs(CO(4)) + abs(CO(5))/60 + abs(CO(6))/3600
if any of CO(4), CO(5), CO(6) < 0: Ypos = -Ypos
Note that, although these are most suited to RA, Dec in
sexagesimal notation, they can be used for any type of
coordinate. The units are standard FITS units (e.g
degrees, m/sec, Hz, sec, etc.) except that right
ascensions are in hours of time.
FSHIFT......The values that one could put in RASHIFT and DECSHIFT
to move this peak to the nearest integer pixel. These
values include any previous shifts and are corrected
for rotation.
ERROR.......If the verb is not aborted for error, but the fit fails,
ERROR will be TRUE (1), else it will be FALSE (-1).
Used to detect failures without stopping a procedure or
RUN file.
EXPLAIN SECTION
MAXFIT: Verb to find coordinates and value of an image extremum
DOCUMENTOR: E.B.Fomalont (NRAO/VLA)
RELATED PROGRAMS: IMFIT, MOMFT, IMVAL
PURPOSE
MAXFIT is a verb which fits a quadratic function to a specified
3x3 map area in order to determine the location and intensity of an
extremum. An area as large as 16x16 can be specified in IMSIZE to
search for the extremum, although the fit will only be made on the 3x3
area surrounding the extremum.
The verb is useful for communicating the location and intensity
of the extremum to other verbs and tasks. After execution PIXVAL will
contain the intensity and PIXXY = a,b will contain the pixel location
of the extrumum. COORDINA will contain the RA and Dec of the
component in sexagesimal form. SHIFT will contain an ra, dec shift in
arc sec which could be fed back to IMAGR (in RASHIFT and DECSHIFT) to
put the source more exactly on a pixel for improved imaging.
MAXFIT is less sophisticated than IMFIT and JMFIT, which do
linearized least-square solutions on a specified map region for a set
of Gaussians. If you want to determine the component size, determine
the errors of the parameters, or if the component is complex in shape,
the use of IMFIT or JMFIT is required.
MAXFIT performs a simple least-square analysis on a 3x3 array
with the function
a + b*x + c*y + d*x*y + e*x*x + f*y*y
The location and position of the extremum is obtained from the
parameters a,b,c,d,e and f. No error checking or goodness of fit
analysis is made in MAXFIT.
If there is an input specification error (bad PIXVAL, bad image
name), MAXFIT aborts the input line, procedure, RUN file, or whatever
which invoked it. However, if the fit itself is the sole source of
error, then MAXFIT does not abort the command stream. Instead, it
sets the adverb ERROR to TRUE if the fit fails or FALSE if it
succeeds. Procedures which depend on the output of MAXFIT should test
ERROR and do appropriate things if it is TRUE. For example:
PROC FITS (I, J)
SCALAR K
ARRAY VALS(20)
PIXXY = I, J
FOR K = 1:20;
EGETNAME(K)
IF ERROR THEN VALS(K) = 0;
ELSE MAXFIT ;
IF ERROR THEN VALS(K) = -1;
ELSE VALS(K) = PIXVAL;
END
END
END
FOR K = 1:20; PRINT K,VALS(K); END
FINISH
PRTLEV = -1 could be used to eliminate the message file displays of
all the coordinates and fluxes. This speeds the process.