Efficient subpixel image registration algorithms.

@article{GuizarSicairos2008EfficientSI,
  title={Efficient subpixel image registration algorithms.},
  author={Manuel Guizar‐Sicairos and Samuel T. Thurman and James R. Fienup},
  journal={Optics letters},
  year={2008},
  volume={33 2},
  pages={
          156-8
        }
}
Three new algorithms for 2D translation image registration to within a small fraction of a pixel that use nonlinear optimization and matrix-multiply discrete Fourier transforms are compared. These algorithms can achieve registration with an accuracy equivalent to that of the conventional fast Fourier transform upsampling approach in a small fraction of the computation time and with greatly reduced memory requirements. Their accuracy and computation time are compared for the purpose of… 

Figures from this paper

High-Speed Image Registration Algorithm with Subpixel Accuracy

A new, fast and computationally efficient lateral subpixel shift registration algorithm is presented that reduces computation time and memory requirements without sacricing the accuracy associated with the usual FFT approach accuracy.

Analytical approximations of translational subpixel shifts in signal and image registrations

  • Qiang Zhang
  • Computer Science
    Optical Engineering + Applications
  • 2008
Analytical approximations of translational subpixel shifts in both signal and image registrations are derived by setting the derivatives of a normalized cross correlation function to zero and solving them, achieving a complexity of only O (mn), given an image size of m × n.

Efficient Image Registration with Subpixel Accuracy using a Hybrid Fourier-based Approach

A highly accurate and efficient algorithm for subpixel image registration using Fourier-based cross correlation to determine the translation between two images using a coarse to fine strategy.

A subpixel registration algorithm for low PSNR images

The proposed algorithm is based on the measurement of centroid on the cross correlation surface by Modified Moment method, which shows that the accuracies of the algorithm are comparable with other subpixel registration techniques and the processing speed is higher.

Comparison of Subpixel Phase Correlation Methods for Image Registration

Abstract : The popular phase correlation method (PCM) for image registration has been extended to subpixel accuracy by several alternative algorithms, each claiming accuracies on the order of 0.02 -

Interpolation Artifacts in Sub-Pixel Image Registration

This work considers the problem of registering (aligning) two images to sub-pixel accuracy by optimization of objective functions constructed from the images' intensity values and discusses different solutions to address the problem based on high degree B-spline interpolation, low pass filtering the images, and stochastic integration.

Solving sub-pixel image registration problems using phase correlation and Lucas-Kanade optical flow method

This algorithm combines well-known phase correlation technique with the differential methods of the optical flow field, especially the Locus-Kanade technique to register images to enhance the precision of registration and reduce time complexity.

Comparison of subpixel image registration algorithms

This research has endeavored to find a suitable method for registering severely undersampled images by comparing several approaches and only considers global geometric transformations for monochromatic images.

Subpixel accuracy analysis of phase correlation registration methods applied to aliased imagery

Results show that with some additional features, the methods based on the one or two dimensional Gaussian fit of the spatial phase correlation peak are the less sensitive to severe aliased imagery.

A Noise-Insensitive Algorithm for Pixel Level Image Registration Based on Centroid

Pixel level displacement measurement is the first stage of Coarse-to-Fine approach in high-accuracy image registration. However, the measurement accuracies of well-known techniques drop sharply with
...

References

SHOWING 1-10 OF 12 REFERENCES

Numerical Recipes: FORTRAN

The Diskette v 2.04, 3.5'' (720k) for IBM PC, PS/2 and compatibles [DOS] Reference Record created on 2004-09-07, modified on 2016-08-08.

Color online) Computation time with respect to (a) image size ( = 25 for DFT algorithms) and (b) upsampling factor for 512ϫ 512 images

    CVGIP: Graph. Models Image Process

    • CVGIP: Graph. Models Image Process
    • 1991

    Opt. Express

    • Opt. Express
    • 2007

    Proc. SPIE 6233

    • Proc. SPIE 6233
    • 2006

    Image Vis. Comput

    • Image Vis. Comput
    • 2003

    Int. J. Comput. Vis

    • Int. J. Comput. Vis
    • 1997

    J. R. Fienup, Opt. Lett

    • J. R. Fienup, Opt. Lett
    • 1978

    SMC3. image size ( =25 for DFT algorithms) and (b) upsampling

    • Express 3,
    • 1998