Iterative least-squares wavefront estimation for general pupil shapes

• US 7,088,457 B1
• Filed: 06/10/2004
• Issued: 08/08/2006
• Est. Priority Date: 10/01/2003
• Status: Active Grant

• Current Assignee: University of Central Florida Research Foundation Inc. (State University System of Florida)
• Original Assignee: University of Central Florida Research Foundation Inc. (State University System of Florida)
• Inventors: Rolland, Jannick P., Zou, Weiyao
• Primary Examiner(s): Pham, Hoa Q.
• Assistant Examiner(s): CONNOLLY, PATRICK J
• Application Number:
  US10/867,527
• Time in Patent Office: 789 Days
• Field of Search: 356/511, 356/512, 356124-127
• US Class Current: 356/512
• CPC Class Codes: A61B 3/1015 for wavefront analysis
• CPC Class Codes: G01J 9/00 Measuring optical phase difference devices or arrangements for controlling the phase of light beams G02F1/01; Determining degree of coherence; Measuring optical wavelength spectrometry G01J3/00
• CPC Class Codes: G01M 11/0257 by analyzing the image formed by the object to be tested
• CPC Class Codes: G02B 27/0012 Optical design, e.g. procedures, algorithms, optimisation routines

Abstract

Iterative algorithm methods and systems for slope or gradient-type data is presented for wavefront zonal estimation of regular and irregular pupil shapes. The methods and systems bears universal wavefront estimation matrices that are directly applicable to any pupil shape without the need to set up new matrices. The algorithm consists in first extending the sampling pupil to a larger regular square shape and second extrapolating the sampled slope data outside of the sampling pupil employing Gerchberg-type iterations. Unbiased least-squares wavefront estimation is then performed in the square domain. Results show that the RMS deviation error of the estimated wavefront from the original wavefront can be less than λ/150 after about twelve iterations and less than λ/100 (both for λ equal 632.8 nm) within as few as five iterations.

Claims

1. A method for wavefront reconstruction comprising the steps of:

(a) obtaining slope-type data from a wavefront sensor;

(b) embedding an irregular pupil to a regular square domain and setting the slopes outside the pupil to zero;

(c) computing a wavefront estimation for the square domain;

(d) computing slopes of the computed wavefront;

(e) replacing the slopes in original pupil region with original raw slope data; and

(f) iteratively computing for the whole extended square domain until the wavefront difference between two consecutive iterations is less than a set threshold value.

2. The method of claim 1 wherein said wavefront sensor is selected from at least one of a slope sensor such as a Shack-Hartmann sensor, shear interferometer, and a slope-type sensor.

3. The method of claim 1 wherein said slope data is selected from one of gradients data, wavefront differences data, and slope-type data.

4. The method of claim 1 wherein said regular square domain refers to a square grid array that is indexed from one of sequentially, and column by column.

5. The method of claim 1 wherein said regular square domain is bigger than the original pupil so that it can embrace the original pupil.

6. The method of claim 1 wherein said the wavefront computation and slope extraction computations use wavefront reconstruction geometry.

7. The method of claim 6, wherein the wavefront recontstruction geometry is selected from one of:

Southwell, Hudgin, and Fried models.

8. The method of claim 1 wherein the step of computing the wavefront from slope data and slope data from the computed wavefront includes using Gerchberg-type iterations.

9. The method of claim 1 wherein the step of computing the wavefront from slope data and slope data from wavefront and replacing the original slope data back into the original pupil includes using wavefront extrapolations.

10. The method of claim 1 wherein the step of computing the wavefront and slopes includes:

linear sparse normal matrix equations.

11. The method of claim 1 wherein the matrices for computing the wavefront and slopesincludes one of:

full-rank matrices, and rank-deficient matrices depending on reconstruction geometry.

12. The method of claim 11, wherein the reconstruction geometry is selected from one of:

Hudgin, Southwell, and Fried geometries.

13. The method of claim 1 wherein said the wavefront difference includes:

wavefront slope.

14. An iterative system for wavefront reconstruction comprising:

means for obaining slope-type data from a wavefront sensor;

means for embedding an irregular pupil to a regular square domain and setting the slopes outside the pupil to zero;

means for determining a wavefront estimation for the square domain;

means for determining slopes of the wavefront estimation;

means for replacing the slopes in original pupil region with original raw slope data; and

means for iteratively determining for the whole extended square domain until the wavefront difference between two consecutive iterations is less than a set threshold value.