eLynx SDK v3.3.0
C++ image processing API reference
Public Member Functions

eLynx::Image::IImagePSF Class Reference

Interface for image PSF (Point Spread Function). More...

#include <IImagePSF.h>

Inheritance diagram for eLynx::Image::IImagePSF:
eLynx::Image::KernelPSF

List of all members.

Public Member Functions

virtual ~IImagePSF ()
 Virtual destructor, provided to ensure virtual destructors in descendants.
virtual uint32 GetWidth () const =0
 Returns width of the PSF function in pixels.
virtual uint32 GetHeight () const =0
 Returns height of the PSF function in pixels.
virtual double EvalPSF (uint32 iX, uint32 iY, int32 iDX, int32 iDY) const =0
 Returns the PSF value (floating point value between 0 and 1).

Detailed Description

Interface for image PSF (Point Spread Function).

The PSF represents an image of ideal point source (like star). It has close relation to convolution kernel - convolution kernel is example of simple PSF. However, PSF can be position dependent - it can vary from point to point. PSF is used in the deconvolution process to revert various effects like remove difraction & seeing caused blur, shapren out-of-focus images, remove aberations (like coma) or field rotation. Deconvolution tends to undone the effects of convolving hypothetical ideal image with PSF function. PSF is 2D nonnegative function relative to pixel. Its total integral should be equal to 1. In theory, it can spread from -infinity to infinity in both directions. In praxis, it is however always limited to small region around pixel - rest of PSF is supposed to be zero.


Member Function Documentation

virtual double eLynx::Image::IImagePSF::EvalPSF ( uint32  iX,
uint32  iY,
int32  iDX,
int32  iDY 
) const [pure virtual]

Returns the PSF value (floating point value between 0 and 1).

The PSF is related to pixel iX and iY of the image. Delta values iDX and iDY represents center-relative coordinates in the PSF image. If the PSF is constant over whole image, iX and iY parameters are ignored.

Parameters:
iXthe x coordinate of the image pixel, for which PSF should be evaluated
iYthe y coordinate of the image pixel, for which PSF should be evaluated
iDXthe delta from PSF center in x direction (PSF coordinates)
iDYthe delta from PSF center in y direction (PSF coordinates)
Returns:
the PSF value for given pixel and delta (floating point value between 0 and 1)

Implemented in eLynx::Image::KernelPSF.

virtual uint32 eLynx::Image::IImagePSF::GetHeight (  ) const [pure virtual]

Returns height of the PSF function in pixels.

The PSF can be viewed as luminance image with width and height. The original point source is in the center of the image. PSF pixel values shows the convolved image of te source. Both width and height should be odd numbers.

Returns:
the width of the PSF function

Implemented in eLynx::Image::KernelPSF.

virtual uint32 eLynx::Image::IImagePSF::GetWidth (  ) const [pure virtual]

Returns width of the PSF function in pixels.

The PSF can be viewed as luminance image with width and height. The original point source is in the center of the image. PSF pixel values shows the convolved image of te source. Both width and height should be odd numbers.

Returns:
the width of the PSF function

Implemented in eLynx::Image::KernelPSF.


The documentation for this class was generated from the following file:

Generated on Thu Dec 9 2010 by doxygen 1.7.2