Cross-Correlation Algorithm Based on Speeded-Up Robust Features Parallel Acceleration for Shack–Hartmann Wavefront Sensing

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The authors present a cross-correlation algorithm based on parallel acceleration feature matching preprocessing for wavefront sensing, aimed at improving the accuracy of scene-based Shack-Hartmann wavefront sensor under partially missing of sub-images. This study provides substantial progress in dealing with extended object scenarios such as strong turbulence or close-range transmission, and shows good accuracy in centroid calculation. In addition, this paper delves into the internal working principle of SURF matching preprocessing, innovatively links the partially missing sub-images with the reference images used in cross-correlation algorithms, and provides accelerated processing for the preprocessing method to meet the real-time performance of AO system, which provides a new idea for extended object sensing under extreme conditions.

 

The manuscript is well-structured and the results are convincingly presented. I commend the authors for their innovative approach and believe it has great potential for future applications. Nevertheless, there are a few aspects that need to be clarified before the manuscript can be accepted for publication in Photonics:

 

1. While the manuscript mentions the limitations of conventional algorithms, a more detailed comparison highlighting the specific advantages of the proposed preprocessing approach would be beneficial. But where is its greatest performance? If the degree of sub-image missingness is too large, can it be processed, and how can it be solved?

 

2. The scene-based Hartmann wavefront sensing of different traditional point source objects has been mentioned many times in the manuscript. Can the author explain the boundary and difference of wavefront sensing between the two? And explain why different centroid calculation methods are used?

 

3. The Hartmann sensor has a high demand for the timeliness of wavefront sensing. Can the acceleration method in this manuscript meet the frequency requirements of Adaptive Optics for atmospheric turbulence correction?

 

In conclusion, the manuscript presents a promising approach for centroid calculation under partial missing of sub-images for AO wavefront sensing, which is a valuable addition to this field. I recommend its publication after addressing the aforementioned points.

Comments on the Quality of English Language

I think the English Language is well, because I can understand what the author wants to express.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Good paper; please see comments in the attached file. The main problem in this paper is the treatment of partial images as compared to partial apertures. The first can occur when looking at the sun and seeing different parts of the same patch because of the different lenslet tilts. The second is due to the edge lenslets which are not full because of the circular aperture blocking their light. This second problem was dealt with by theory by Carmon and Ribak (JOSA A 26, 85, 2009) and dealt with practically by Talmi and Ribak (JOSA A 21, 832, 2004) by extending the last subapertures outside the aperture. It might be that the solution in this paper can apply to both cases (subapertures and images), but the authors need to show it. In addition, I could not find Reference 19, which might be the basis for this paper (it might be in Chinese, but Goggle Scholar did not show it). So some of the explanations in this paper need to be extended because of this missing reference.

Comments for author File: Comments.pdf

Comments on the Quality of English Language

No comments

Author Response

Please see the attachment.

Author Response File: Author Response.pdf