Astronomical Institute

  The art of image processing is a topic  with boundaries well
 beyond  astronomy.  Astronomical   image  processing  applies
 a variety of numerical methods  to extract scientific results
 from the  observed data. The  goal is to  provide theoretical
 grounds  and   reliable  methods  for   image  processing  in
 astronomy and astrophysics. The tasks  are aimed at the three
 general   areas  of   image  processing   -  restoration   of
 single-frame  2D  solar  images,  processing  of time-varying
 images and image reconstruction.

        Recent results:

 Solar corona events imaged in soft X-rays are studied and the
 new processing  method based on  a special algorithm  for the
 local optimization of density has been developed. The results
 led, in  time-sequence  images, to  the  internal and spatial
 description  of  the  developing  coronal  arches.  A  method
 evaluation is in progress on solar prominences processing.
   A new type of  adaptive reconstruction method is introduced
 to  reconstruct  failed  and/or  damaged multi-spectral image
 data. The image reconstruction is  based on a regression type
 of   image  destriping  method,  using   a  regression  model
 prediction for  replacing a corrupted  image row or  isolated
 pixels.  A  method  modification  reconstructs  scratches  or
 several neighbouring missing lines, using a multi-dimensional
 regression   model.  Mono-spectral   data  reconstruction  is
 treated as a special case  of the proposed method. The unique
 results  of reconstruction are higher  in order in comparison
 the   other   methods   (interpolation,   masking,   template
 replacement with an error  adjustment). The suggested methods
 can  be successfully  applied also  in other  acquisition and
 processing application.
   In pre-processing of the photospheric and corona images the
 temporal   image   sequences   have   been  transformed  into
 a geometrically  congruent form.  This step  is essential for
 the  image-to-image registration  and it  is required  before
 further quantitative  and qualitative analysis.  A process of
 removing   blurring   effects   caused   partially   by   the
 non-perfect  point  spread  function  of  the  telescope  and
 partially by atmospheric turbulence has been studied. A novel
 method for  a time sequence restoration  of the blurred solar
 images has been introduced.