Paper
21 September 2007 Refocusing of defocused images using SAR2 algorithm
Author Affiliations +
Abstract
This paper utilizes the methodology of SAR2 algorithm, a two dimensional variant of the ω-k algorithm, to refocus out-of-focus images. Refocusing of images may be necessary in machine vision as a preprocessing step before edge detection or image segmentation in the imaging and manipulation of 3D objects. The SAR2 algorithm generates a complex amplitude distribution and the corresponding point spread function in a manner similar to Fraunhofer diffraction distribution model and its point spread function as seen in Fourier optics. The matched filter utilized in the SAR2 algorithm has a focus-in-altitude interpretation and may be varied to increase or decrease the radius of out-of-focus blur associated with a particular point spread function of scatterers of various heights. This paper demonstrates focusing of a line object L={1 : x=y;-≤x≤63; -64≤y≤63}. Although a rectangular aperture is used in the refocusing process, other apertures may also be used such as circular or Gaussian.
© (2007) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Jehanzeb Burki and Christopher F. Barnes "Refocusing of defocused images using SAR2 algorithm", Proc. SPIE 6697, Advanced Signal Processing Algorithms, Architectures, and Implementations XVII, 66970D (21 September 2007); https://doi.org/10.1117/12.735343
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Point spread functions

Fermium

Frequency modulation

Synthetic aperture radar

Image processing algorithms and systems

Cameras

Fourier optics

RELATED CONTENT

Real-time smart optical SAR signal processor
Proceedings of SPIE (September 24 1997)
Time-varying higher order spectra
Proceedings of SPIE (December 01 1991)
Metrics for image segmentation
Proceedings of SPIE (July 06 1998)
FOPEN ultrawideband SAR imaging by wavelet interpolation
Proceedings of SPIE (September 12 2003)

Back to Top