Current Status of the Community Sensor Model Standard for the Generation of Planetary Digital Terrain Models

Round 1

Reviewer 1 Report

The present paper endeavors to apply the recently prominent CSM to planetary exploration data. By incorporating the usgscsm, ALE, and knoten libraries, the study enables a novel application of SOCET GXP, with a focus on the creation of HiRISE Digital Terrain Models (DTMs) as the most prolific illustrative example. While the majority of the content is meticulously described, there are certain sections that require additional clarification, prompting the request for necessary revisions. Moreover, there appears to be a discrepancy in the level of extrapolation between the title, abstract, and highlights sections in comparison to the main body of the text, suggesting a need for reconsideration in these areas. Please refer to the comments provided in the attached PDF for detailed feedback and comments.

Comments for author File: Comments.pdf

Author Response

Great comments. I should have addressed them all. Comments in the original comments.

Author Response File: Author Response.pdf

Reviewer 2 Report

Minor revision

Currently, the available photogrammetric software tools to process the planetary images are still limited. The open source USGS ISIS and NASA ASP exhibit their advantages in some aspects (e.g., official camera geometric parameters for dozens of planetary scientific instruments, rigorous sensor model for pushbroom images, and powerful bundle adjustment capabilities). NASA ASP can automatically extract DTM based on dense image matching. However, both USGS ISIS and NASA ASP do not provide software tools to edit DTM. But DTM edition feature is still very useful for generating high-quality DTM. Thus, the transition to the Community Sensor Model (CSM) is valuable for planetary DTM generation. Then, photogrammetric engineers can use SOCET GXP to generate planetary DTMs using manual edit or dense image matching. Currently, in planetary mapping community, the software interoperability is achieved mainly among USGS ISIS, NASA ASP and SOCET SET. Since millions of planetary images (e.g. LRO NAC and MRO HiRISE) need to be geometrically processed, we require more software tools to process the massive planetary images. In my opinion, the transition to the CSM also facilitates other software to support planetary images, encompassing conventional photogrammetric software such as Leica Photogrammetry Suite and VirtuoZo or computer vision software such as Pix4D and Photoscan (i.e., Agisoft Metashape). Thus, considering the high value of the CSM in planetary mapping community, I suggest minor revision for this manuscript.  

1.     I am very interested in the efficiency of the CSM especially for line scan sensors. One of the camera model’s main objectives is to conduct the image-to-ground and ground-to-image (back-projection) transformation operations. As we all know, for line scan sensors, the ground-to-image operation requires iterative calculations to determine the best scan line corresponding to the ground point, which is time consuming. The ground-to-image operation is used for both DOM and DTM generation. Thus, I suggest the authors add some materials for evaluating the efficiency of the CSM.

2.     In Table 1., the camera type of TGO CaSSIS is push frame instead of frame.

3.     Line 204~213, the instruments have been listed in Table 1. Thus, I suggest that the authors remove these texts and the references can be merged into Table 1.

4.     Line 230, “It was missing an important method…”, may be grammatical errors. “It missed an ***” may be appropriate.

5.     Line 284-285, “can feature match” is also a grammatical error.

6.     “DTM extract” in Figure 4 caption may be “DTM extraction”.

need more editing

Author Response

Thanks for the comments. All grammar issues have been updated and I believe I addressed 1. also.

1.   I am very interested in the efficiency of the CSM especially for line scan sensors. One of the camera model’s main objectives is to conduct the image-to-ground and ground-to-image (back-projection) transformation operations. As we all know, for line scan sensors, the ground-to-image operation requires iterative calculations to determine the best scan line corresponding to the ground point, which is time consuming. The ground-to-image operation is used for both DOM and DTM generation. Thus, I suggest the authors add some materials for evaluating the efficiency of the CSM.

We do plan a more technical follow-on paper to assess the updated CSM code and new stereo methods (ASM) offered in SOCET GXP. The original "usgscsm" paper from 2020, as referenced, does take a stab at reporting the g2i and ig2 calculations using usgscsm. The paper is here - https://doi.org/10.1029/2019EA000713.

2.     In Table 1., the camera type of TGO CaSSIS is push frame instead of frame.

good catch. Updated

3.     Line 204~213, the instruments have been listed in Table 1. Thus, I suggest that the authors remove these texts and the references can be merged into Table 1.

Good recommendation. I pushed the references into the table.

4.     Line 230, “It was missing an important method…”, may be grammatical errors. “It missed an ***” may be appropriate.

corrected

5.     Line 284-285, “can feature match” is also a grammatical error.

corrected

6.     “DTM extract” in Figure 4 caption may be “DTM extraction”.

Updated

Reviewer 3 Report

This paper shows the usgscsm library is ready for DTM production in SOCET GXP, tested the new capabilities of SOCET GXP, first implement a CSM-based sensor model over an ISIS-based sensor model, which will be very useful planetary community. I recommend this paper that can be published as soon as possible.

Comments for author File: Comments.pdf

Author Response

Thank you for your feedback. I have fix the issues you have noted.

Regards,
Trent