; CONVL ;--------------------------------------------------------------- ;! convolves an image with a gaussian or another image ;# Task Imaging AP ;----------------------------------------------------------------------- ;; Copyright (C) 1995, 2002-2004, 2008-2010 ;; Associated Universities, Inc. Washington DC, USA. ;; ;; This program is free software; you can redistribute it and/or ;; modify it under the terms of the GNU General Public License as ;; published by the Free Software Foundation; either version 2 of ;; the License, or (at your option) any later version. ;; ;; This program is distributed in the hope that it will be useful, ;; but WITHOUT ANY WARRANTY; without even the implied warranty of ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ;; GNU General Public License for more details. ;; ;; You should have received a copy of the GNU General Public ;; License along with this program; if not, write to the Free ;; Software Foundation, Inc., 675 Massachusetts Ave, Cambridge, ;; MA 02139, USA. ;; ;; Correspondence concerning AIPS should be addressed as follows: ;; Internet email: aipsmail@nrao.edu. ;; Postal address: AIPS Project Office ;; National Radio Astronomy Observatory ;; 520 Edgemont Road ;; Charlottesville, VA 22903-2475 USA ;----------------------------------------------------------------------- CONVL LLLLLLLLLLLLUUUUUUUUUUUU CCCCCCCCCCCCCCCCCCCCCCCCCCCCC CONVL: Task to convolve two images. Input image. INNAME Image name (name) INCLASS Image name (class) INSEQ 0.0 9999.0 Image name (seq. #) INDISK 0.0 9.0 Image disk drive # OPCODE Operation desired: GAUS, IMAG, IMAC, DCON, DGAU BLC Bottom left corner TRC Top right corner Convolving image (IMAG, DGAU) IN2NAME Image name (name) IN2CLASS Image name (class) IN2SEQ 0.0 9999.0 Image name (seq. #) IN2DISK 0.0 9.0 Image disk drive # Convolved image. OUTNAME Image name (name) OUTCLASS Image name (class) OUTSEQ -1.0 9999.0 Image name (seq. #) OUTDISK 0.0 9.0 Image disk drive # IMSIZE Size of box to find position of max in output image (IMAC) BMAJ -999.9 FWHM(asec) maj. axis output. beam - required except IMAG BMIN -999.9 FWHM(asec) min. axis output beam. BPA -360.0 360.0 position angle FACTOR Unit conversion factor; clip inv Gaussian level DCON DOBLANK -1.0 1.0 > 0 => reblank output images where input was blanked BADDISK Disks to avoid for scratch. ---------------------------------------------------------------- CONVL Type: Task Use: CONVL convolves your map with a either an eliptical gaussian or a specified convolving image. Strictly, only the inner quarter of the map is properly convolved, but for the Gaussian this problem may not be apparent. CONVL uses a 2D FFT, and so it cannot properly handle images containing blanks. It will (in 31DEC04) replace blanks with 0.0, do the convolution, and then put the blanks back. Nonetheless, the effect of the 0.0 will be felt in non-blank pixels. The second image must not contain blanked pixels. CONVL surrounds images with sizes other than a power of 2 with zeroes up to the next power of 2, convolves using the 2D FFT, and then removes the borders for output. Adverbs: INNAME......The input image name. Standard defaults. INCLASS.....The input image class. Standard defaults. INSEQ.......The input image seq. #. 0 => highest. INDISK......The input image disk drive #. 0 => any. OPCODE......A code indicating the desired operation. 'GAUS' => convolve the image with the specified Gaussian. 'IMAG' => convolve image 1 with the second image. 'IMAC' => crosscorelate the image1 with the second image. 'DCON' => deconvolve the image from specified Gaussian. 'DGAU' => convolve the image with the deconvolution of the Gaussian and the second image (Gaussian divided by image in Fourier space) blank => 'GAUS' BLC.........The bottom left corner of a subimage to be convolved. (0's => 1,1,1,1,1,1,1) TRC.........The top right corner of the subimage to be convolved. (0's => all of image from BLC) IN2NAME.....The convolving image name. Standard defaults. Must be same size and cell spacing as INNAME. IN2CLASS....The convolving image class. Standard defaults. IN2SEQ......The convolving image seq . #. 0 => highest. IN2DISK.....The convolving image disk drive #. 0 => any OUTNAME.....The output image name. Standard defaults. OUTCLASS....The output image class. Standard defaults. OUTSEQ......The output image seq. #. 0 => highest unique. Input file (only) may be overwritten. OUTDISK.....The clean disk drive no. 0 => highest with space IMSIZE......Size of the box used to calculate max position in the output image. Used only if OPCODE='IMAC' Must be odd!!! 0 => 5, 5 BMAJ........Required for all OPCODEs except IMAG: the FWHM (asec) major axis of the desired output beam. 0 -> Clean beam for DCON only, for rest it is an error. Other than IMAG, it will be deconvolved from the clean beam if the input image is a clean image. If the deconvolution fails, the program will terminate. NOTE: as of 29 June 2010, the task will read a CG table if present and determine the Clean beam for each channel from it. The Gaussian used will then vary from channel to channel and will be fully reported in the messages and the history file. BMIN........The FWHM (asec) minor axis of the output beam. If <= 0 BMIN = BMAJ BPA.........The position angle in the unrotated image of BMAJ. FACTOR......All OPCODEs except DCON: the conversion factor for the image units. This value of FACTOR if specified should be the ratio of the new beam area to the old beam area. If FACTOR is > 0.0, it will be used to convert from JY/(old beam) to JY/(new beam). If FACTOR is 0.0 or less, it will be determined from the output beam size and any CLEAN beam stored in the catalog. If FACTOR is not given and cannot be computed it is set to 1.0 and the map units are changed to "UNDEFINE". If flux per pixel output is desired then use FACTOR=1.0. If input image is measured in Jy/pixel and OPCODE=GAUS, then the convolved image is scaled by FACTOR/(NPIXATBEAM)) where NPIXATBEAM is number of pixels at the convolving beam; NPIXATBEAM = BMIN*BMAJ*ALFA BMIN, BMAJ are in pixels, ALFA = PI/(4*ln(2))=1.13309 If FACTOR=0 then the output image is not scaled(multiplied by 1) DCON only: FACTOR is instead the limit in the value of 1/Gaussian. <= 1.01 => 1000. DOBLANK.....> 0 => reblank those pixels that were blanked in the input image. This still leaves pixels around the blanked affected by the use of 0.0 in the convolution. <= 0 => leave previously blanked pixels at whatever the convolution produced from the 0.0 used instead of the blank value. BADDISK.....This array contains the numbers of disks on which it is desired that scratch files not be located. BADDISK has no effect on input and output maps. ---------------------------------------------------------------- A remark about convolving a map which has been CLEANed (added by DCW, 15 October 84) In the help text above under the heading for BMAJ it says: "[BMAJ] will be deconvolved from the clean beam if the image is a clean image." What this really means is that CONVL will attempt to produce a map which has the desired clean beam shape, and in the process it will account for an existing beam as declared in the header. This is done by deriving the parameters of a Gaussian which when convolved with the existing clean beam will produce the desired beam (the parameters of this beam are listed in the history file). Obviously this scheme only works when the desired beam is larger than the existing beam (use the DGAU option if you want to deconvolve). Please note that AIPS has tools for inserting clean beam parameters into headers if they are not already there. A preliminary explanation of the 'DGAU' option (added by DCW, 26Sept84) Suppose that the second image contains the beam pattern of the first image. Then DGAU will convert the effective beam of the input image to the specified Gaussian. If the specified Gaussian is narrower than the original beam this amounts to a simple linear deconvolution scheme. The operation is done by dividing the transform of the specified Gaussian by the transform of the second image, and multiplying the quotient by the input image. The quotient is set to zero when the transform of the second image is zero. An important application of DGAU is "cleaning up" the seeing profile of an optical image. This is done by specifying a circular Gaussian which is a good fit to the observed profile (i.e., the second image). DGAU will then correct for the ellipticity of the profile and its excess intensity in the wings, producing an image which has a nice Gaussian beam, although no resolution enhancement.