[daip] CONVL and blanked images

Eric Perlman perlman at jca.umbc.edu
Thu Jun 27 14:20:05 EDT 2002


Hi,

I need to use CONVL on some images to degrade their resolution - in this case 
from HST resolution down to Chandra resolution.   The images have blanks in 
them because they've been rotated to north-up (which was not their original 
orientation).  When I run them through CONVL with either a Gaussian or a 
simulated Chandra PSF, I get this error:  

daneel> CONVL1: Task CONVL (release of 31DEC01) begins 
daneel> CONVL1: Using factor = 1.00000E+00 to scale image 
daneel> CONVL1: ZERROR: ON FILE DA05:CAD000000.1E6; 
daneel> CONVL1: ZERROR: IN ZDAOPN ERRNO = 5 (Input/output error) 
daneel> CONVL1: ZERROR: ON FILE DA05:CAD000000.1E6; 
daneel> CONVL1: ZERROR: IN ZDAOPN ERRNO = 5 (Input/output error) 
daneel> CONVL1: Create M87_NEW_P .XIPCV .  1 (MA)  on disk 1 cno 96 
daneel> CONVL1: ZERROR: ON FILE DA05:CAD000000.1E6; 
daneel> CONVL1: ZERROR: IN ZDAOPN ERRNO = 5 (Input/output error) 
daneel> CONVL1: ZERROR: ON FILE DA05:CAD000000.1E6; 
daneel> CONVL1: ZERROR: IN ZDAOPN ERRNO = 5 (Input/output error) 
daneel> CONVL1: ZERROR: ON FILE DA05:CAD000000.1E6; 
daneel> CONVL1: ZERROR: IN ZDAOPN ERRNO = 5 (Input/output error) 
daneel> CONVL1: PLNGET: MAP IS BLANKED AND I AM NOT ALLOWED TO HANDLE IT 
daneel> CONVL1: Destroyed MA image file: catno= 96 disk=1 
daneel> CONVL1: Purports to die of UNNATURAL causes 
daneel> CONVL1: daneel 31DEC01 TST: Cpu= 0.2 Real= 53

Now, I've looked at the HELP and EXPLAIN files for CONVL and understand that 
it can't handle blanked images.  So I tried to specify a subimage, within 
which there were no blanked pixels.  Yet even when I do this I still get the 
same error.  I also get the same error in both OPCO 'GAUS' and OPCO 'IMAG'.

Do you have any idea what is going on?  I've verified through 'imstat' that 
the section of the image I'm using has no blanked pixels, so I'm not missing 
any blanked pixels.  Is there a workaround and if so what is it?

Thanks,

Eric
  




More information about the Daip mailing list