Verification of Characteristics of Spline Filter Series Robust Filters for Surface Roughness and Proposal of Filter Selection Guidelines

Round 1

Reviewer 1 Report

In this manuscript, the authors clarified the characteristics of SF-based robust filters, and which filters can deal with spike and shape components, and verified the cases in which problems arise. It is better if the authors consider the following mentioned remarks and further improve the manuscript before submitting the final version.

1. Experiments: "Figure 1(b)shows the deviation between the mean line calcu- lated using SF and the mean line calculated using each filter.Figure 1(b)depicts that the mean lines calculated using SF and FMSF are equal.In contrast,the mean lines calculated using SF and RL2-RSF are different,confirming that these filters are incompatible." The picture does not correspond to the described content.

2. No deviation curve of SF is observed in Figure 3.

3. Experiments: "Figure 3 showsthat the mean line calculated using SF is affected by the spike and is affected depending on the height of the spike.The region between the spikes is also affected in the direction opposite to the spikes,and the influence of the spikes can be observed throughout.In contrast,the mean lines calculated using RL2-RSF and FMSF are less affected by the spikes." This is an explanation of Figure 2.

4. For the arc profile, Figure 7 does not show the characteristics of each filter.

5. The author should consider when spikes, steps and arcs are included in the main profile to verify the characteristics of each filter, which helps to establish a new RSF standard.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The authors show how different spline filters to measure surface roughness behave in the presence of outliers and step-shaped protuberances. They also suggest a workflow to decide which one to use depending on the situation.

 

I find the work interesting and worth of publishing, after minor revisions indicated below.

 

1) It is not clear to me why results from the standardized Gaussian filters (GFs) are not presented. As GFs are standardized, it is important to show how the spline filters (SFs) behave with respect to the standard, in particular given that the authors seem to argue about including SFs in ISO standards. I'd like the authors to add GFs measurements in the plots that include SFs measurements to compare them.

 

2) While the SFs have been introduced in other papers that are properly referenced by the authors, it is hard to follow explanations of their behaviors without a formula. I'd like the authors to briefly re-introduce the formulations for the different SFs and how the different quantities are estimated with each filter (e.g. Ra). They can be reported either in the main text or in an appendix.

 

3) The introduction is redundant. The same concept of one filter being excluded by ISO and one being included but having limitations is repeated without new information.

 

4) The text is confusing at times, e.g. sentences like "In this way, standards are sometimes established, including fatal problems" and "The target primary profiles are follows" do not make sense to me. Same for sentences like "Therefore, because the previous ISO/TS 166110-32 has already

been rejected, because the specification of the new ISO 16610-32 is still unknown and is not included in the target filters in this paper" - I don't understand what the authors mean.

Same here "Second, re-quantize in the height z-direction and vote the weight to the cell corresponding to the height value" maybe the authors mean 'choose' instead of 'vote'? I'm not sure.

 

5)Fig 1b: no SF line in the plot.

 

6)Not necessary but it would be interesting to see a comparison between Ra of the filters already presented, and Ra of the raw profile data (e.g. assuming an horizontal average profile).

 

7)Fig 8: not clear which figure is which filter

 

8)"when the width is less than 0.2λc one" why the word 'one'?

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf