Technical Note: Blue and White Light RTI for Imaging Micro-Features on Glass Surfaces

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This is interesting research that presents a practical application of Reflectance Transformation Imaging (RTI) for the study of delicate features on glass, which includes crizzling and toolmarks. It is methodologically sound, and the writing is very clear, and the methods are based on conservation practice.

However, there are some areas that need to be clarified and expanded and/or improved. With that said, the paper also makes a useful contribution to the field of conservation, especially when it comes to dealing with fragile or transparent surfaces.

Some of the strengths are relevance and applicability, innovation – such as the application of blue light RTI and the use of image subtraction for degradation monitoring, documentation clarity and reproducibility.

Here are the places where there is some work needs to be done.

1. There is no quantitative analysis:

The authors acknowledge that their findings are qualitative, but no attempts were made for quantitative surface difference (like RTI normal analysis, texture metrics)

2. The limitations have not been discussed much:

The challenge with focusing and demo-camera-object geometry constraints are well-noted but the trade-off between the resolution and reproducibility could be better delved into.

3. RTIViewer Modes are not explored much:

The specular enhancement and normal visualization are briefly described but no technical discussion is presented on what those transformations imply for surface analysis.

4. There is a limited comparison to prior studies:

There is a good literature review but there is no critical comparison.

5. Clarity on blue light wavelength:

The paper vaguely references “about 450 nm” for blue, but the LED spectral profile can vary significantly.

6. Ethical considerations

The historical significance of the Haitian buttons is mentioned briefly, but it would be good to consider adding a sentence handling ethics, access limitations, or risk mitigation, even in technical notes.

 

Please also ensure all images are properly captioned in the final layout, as a few of them are floating and not even referenced in the body.

Maybe also a more informative title, to also include the glass and the white and blue light.

 

Author Response

Reviewer 1

This is interesting research that presents a practical application of Reflectance Transformation Imaging (RTI) for the study of delicate features on glass, which includes crizzling and toolmarks. It is methodologically sound, and the writing is very clear, and the methods are based on conservation practice.

However, there are some areas that need to be clarified and expanded and/or improved. With that said, the paper also makes a useful contribution to the field of conservation, especially when it comes to dealing with fragile or transparent surfaces.

Some of the strengths are relevance and applicability, innovation – such as the application of blue light RTI and the use of image subtraction for degradation monitoring, documentation clarity and reproducibility.

Here are the places where there is some work needs to be done.

There is no quantitative analysis:

· The authors acknowledge that their findings are qualitative, but no attempts were made for quantitative surface difference (like RTI normal analysis, texture metrics)

· Response: This is a good point, thank you. We acknowledge that future plans for the study include quantitative analysis in the conclusion, citing a recent paper published by two of the co-authors that used Machine Learning to assess surface changes on transparent media. The current paper goal is to share the initial study as a technical note given the usefulness of the technique and the acknowledgement that not all conservators are comfortable using advanced computational analysis to assess images.

The limitations have not been discussed much:

· The challenge with focusing and demo-camera-object geometry constraints are well-noted but the trade-off between the resolution and reproducibility could be better delved into.

· Response: We have made additions that hopefully speak more to this trade-off between the level of detail that can be resolved and the acquisition time and reproducibility.

RTIViewer Modes are not explored much:

· The specular enhancement and normal visualization are briefly described but no technical discussion is presented on what those transformations imply for surface analysis.

· Response: Additions have been made to provide more technical discussion of what is observed with the different rendering modes.

There is a limited comparison to prior studies:

· There is a good literature review but there is no critical comparison.

· Response: We feel that an extended discussion is beyond the scope of a brief technical note given the short word count. Please advise if this requested addition is obligatory as we feel that our review of related publications is sufficient.

Clarity on blue light wavelength:

· The paper vaguely references “about 450 nm” for blue, but the LED spectral profile can vary significantly

· Response: We added more specific details for the LEDs.

Ethical considerations

· The historical significance of the Haitian buttons is mentioned briefly, but it would be good to consider adding a sentence handling ethics, access limitations, or risk mitigation, even in technical notes.

· Response: A note has been added about access and ongoing preservation.

Please also ensure all images are properly captioned in the final layout, as a few of them are floating and not even referenced in the body.

· Response: We confirmed that all images are captioned and referenced.

Maybe also a more informative title, to also include the glass and the white and blue light.

· Response: The title has been updated to be more informative and include RTI along with white and blue light.

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript is clearly written, and the methodology is particularly well detailed and articulated. However, the results are somewhat limited, as the study lacks a convincing demonstration of the method’s effectiveness in tracking image changes over time in a realistic or semi-realistic context. To strengthen the validation of the proposed approach, I recommend conducting tests on mock-up glass samples with unstable compositions, subjected to artificial ageing to simulate crizzling. This would allow for the monitoring of crack formation and its progression over time, thereby providing more robust evidence of the method’s applicability in practical scenarios.

Author Response

Reviewer 2:

The manuscript is clearly written, and the methodology is particularly well detailed and articulated. However, the results are somewhat limited, as the study lacks a convincing demonstration of the method’s effectiveness in tracking image changes over time in a realistic or semi-realistic context.

To strengthen the validation of the proposed approach, I recommend conducting tests on mock-up glass samples with unstable compositions, subjected to artificial ageing to simulate crizzling. This would allow for the monitoring of crack formation and its progression over time, thereby providing more robust evidence of the method’s applicability in practical scenarios.

· Response: Thank you for the suggestion. Now that we have shown, and shared, the usefulness of this technique to capture and study micro-features on glass, future work could entail mock-up glasses and artificial aging as per this suggestion. At the time of the study, these steps were outside of the scope and budget for the project.

Reviewer 3 Report

Comments and Suggestions for Authors

The Technical Note "Imaging Micro-features on Glass Surfaces" presents a test of using RTI to document and monitor defects on glass surfaces. The authors show the initial test using a clock glass and images from ivory buttons. The manuscript is very well written, and the structure is clear and well organized. The results are promising and merit further development. Based on these observations, I suggest accepting this technical note after minor revisions. I explain them in detail below. 

• The word "caption" is not needed in the caption of the Figures.
• In the introduction, when referring to weeping. I suggest adding a few references on the possible objects where this phenomenon has been encountered to give the reader a quick overview of the frequency of this degradation in cultural heritage collections. Mainly thinking about the applicability of this methodology, which, if I understand correctly, can be applied to relatively small objects.
• Only if it is not a requirement from the owner of the buttons, I suggest removing from the captions the phrase "Gift of R. Keith Kane from the Estate of Mrs. Robert B. Noyes; 1949-94-1/18."  Mentioning it in Section 2.1, which I suggest to rename as Mock-up and Case Study. 
• In the same section, you mention the name of the author. I suggest moving the information "(Italian, ca. 1730–1796)" to the text, since it is currently only present in the Figure caption.
• Painting the clock glass might not be a practical solution for real artefacts. Please elaborate on this and the potential solutions to this issue. 

As a more general comment for future research I suggest you to develop a methodology that uses Open source software (for example by substituting Photoshop with Gimp) to make it more accessible to potential users. 

 

 

 

Author Response

Reviewer 3

The Technical Note "Imaging Micro-features on Glass Surfaces" presents a test of using RTI to document and monitor defects on glass surfaces. The authors show the initial test using a clock glass and images from ivory buttons. The manuscript is very well written, and the structure is clear and well organized. The results are promising and merit further development. Based on these observations, I suggest accepting this technical note after minor revisions. I explain them in detail below.

· The word "caption" is not needed in the caption of the Figures.

· Response: This has been corrected.

· In the introduction, when referring to weeping. I suggest adding a few references on the possible objects where this phenomenon has been encountered to give the reader a quick overview of the frequency of this degradation in cultural heritage collections. Mainly thinking about the applicability of this methodology, which, if I understand correctly, can be applied to relatively small objects.

· Response: We added a citation that discusses both published occurrences of the phenomenon but also a Smithsonian project focused on capturing a rough estimate of the percentage of unstable glass in the collections.

· Only if it is not a requirement from the owner of the buttons, I suggest removing from the captions the phrase "Gift of R. Keith Kane from the Estate of Mrs. Robert B. Noyes; 1949-94-1/18." Mentioning it in Section 2.1, which I suggest to rename as Mock-up and Case Study.

· Response: Gift is part of the required image credits for the object. Proxy changed to mock-up.

· In the same section, you mention the name of the author. I suggest moving the information "(Italian, ca. 1730–1796)" to the text, since it is currently only present in the Figure caption.

· Response: The suggested move has been taken care of.

· Painting the clock glass might not be a practical solution for real artefacts. Please elaborate on this and the potential solutions to this issue.

· Response: This sentence has been expanded to provide more information and to remove a reference. There is also a footnote that addresses the reviewer’s comment.

· As a more general comment for future research I suggest you to develop a methodology that uses Open source software (for example by substituting Photoshop with Gimp) to make it more accessible to potential users.

· Response: While Photoshop was used for this investigation, the processing is based on straightforward image math and open-source software could be used in place of Photoshop for the processing. We have added language to acknowledge this.

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

I appreciate the revisions made to the manuscript. I believe this study provides a valuable foundation, and it would certainly merit further research aimed at developing and implementing the quantitative aspects of the proposed method.