Experimental Investigation of the Performance of Image Registration and De-aliasing Algorithms
Abstract
Various image de-aliasing techniques and algorithms have been developed to improve the resolution of sensor-aliased images captured with an undersampled point spread function. In the literature these types of algorithms are sometimes included under the broad umbrella of superresolution. However, in the current paper we consider them to be a form of image restoration, because we aim to restore image resolution lost due to sensor aliasing, but only up to the limit imposed by diffraction. Much of the previous work in this area demonstrates improvement by using simulated imagery, or using imagery obtained where the subpixel shifts are unknown and must be estimated. This paper takes an experimental approach to investigate performance, in which the subpixel shifts are known a priori at some level. We use a CMOS camera, an ISO 12233 resolution target, and two linear translation stages which provide two-axis camera control via RS-232 interface. The translation stages provide a microstepping capability which allows discrete steps of approximately 0.1 micron. The ISO 12233 target includes many features, including those for estimating the line Spread Function and Modulation Transfer Function (MTF) via the tilted-edge response and Fourier transforem, as well as square wave burst patterns for estimating the MTF more directly using lie Contrast Transfer Function method. This target also includes a series of slightly tilted square wave burst patterns intended for measuring the aliasing ratio. We use one of these patterns to visually demonstrate successful de-aliasing.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 2009
- Accession Number
- ADA511829
Entities
People
- Peter N. Crabtree
- Phan D. Dao
- Richard H. Picard
Organizations
- Air Force Research Laboratory