Review Reports

Round 1

Reviewer 1 Report

Following the geometry of the equipment presented in Figs. 1 and 8, laser line projected from a particular angle with respect to the camera can be used for measurement of the projection (calibration) board profile characteristics, roughness or waviness (using laser profilometry principle). For example, see the articles "Laser Profilometry of Polymeric Materials", "Application of Laser Profilometry to Evaluation of the Surface of the Workpiece Machined by Abrasive Waterjet Technology", "Following the geometry of the equipment presented in Figs. 1 and 8, laser line projected from a particular angle with respect to the camera can be used for measurement of the projection (calibration) board profile characteristics, roughness or waviness (using laser profilometry principle). For example, see the articles "Laser Profilometry of Polymeric Materials", "Application of Laser Profilometry to Evaluation of the Surface of the Workpiece Machined by Abrasive Waterjet Technology", "Non-contact study of surfaces created using the AWJ technology" etc.
So, did the authors measure the profile characteristics of the board? 
For the calibration purpose, at best, the laser line should be projected and also viewed from above, perpendicularly to the board. What are the angle and distance of the laser source and CCD camera from the board? 
The laser line was projected to various places on the board? Taking Fig. 8 into account, the angle and the place where the laser line was projected were varying. 
What are the optical properties of the camera in combination with the lens, namely depth of field? 
Calibration with resulting accuracy is valid for the distance to the board as well for the distance to the gauge block which is relatively thick (see Fig. 9)?
Could authors calculate line/pixel measurement uncertainty with regard to the accuracy of the calibration board (0.01 mm) with respect to the used measurement distances and angles?" etc.
So, did the authors measure the profile characteristics of the board? 
For the calibration purpose, at best, the laser line should be projected and also viewed from above, perpendicularly to the board. What are the angle and distance of the laser source and CCD camera from the board? 
The laser line was projected to various places on the board? Taking Fig. 8 into account, the angle and the place where the laser line was projected were varying. 
What are the optical properties of the camera in combination with the lens, namely depth of field? 
Calibration with resulting accuracy is valid for the distance to the board as well for the distance to the gauge block which is relatively thick (see Fig. 9)?
Could authors calculate line/pixel measurement uncertainty with regard to the accuracy of the calibration board (0.01 mm) with respect to the used measurement distances and angles?

Author Response

Reviewer 1:

 

Dear reviewer, we appreciate your comments on improving the quality of this manuscript. Our responses to the comments are as follows.

 

Question 1:

"Following the geometry of the equipment presented in Figs. 1 and 8, laser line projected from a particular angle with respect to the camera can be used for measurement of the projection (calibration) board profile characteristics, roughness or waviness (using laser profilometry principle).

For example, see the articles "Laser Profilometry of Polymeric Materials", "Application of Laser Profilometry to Evaluation of the Surface of the Workpiece Machined by Abrasive Waterjet Technology", "Non-contact study of surfaces created using the AWJ technology" etc.

So, did the authors measure the profile characteristics of the board? 

Response:

      We thank you for such a good comment. We have downloaded the three articles above and read them in detail following your suggestion. Because of the excellent contribution of these articles in non-contact measurement, we have added these three articles to the review of non-contact measurement technology in our manuscript, which can be seen on line 33 of the revised manuscript.

However, we did not measure the profile characteristics of the calibration board in the manuscript. The calibration board is only used as the standard world coordinate for the camera calibration and the light plane calibration.

 

Question 2:

For the calibration purpose, at best, the laser line should be projected and also viewed from above, perpendicularly to the board. What are the angle and distance of the laser source and CCD camera from the board? 

Response:

     We think this is a very professional question. We appreciate your good comment. It is very sorry that we did not describe the calibration process clearly in Fig. 1.

     When calibration is in progress, we admit that the position of the camera in Fig. 1 relative to the calibration board was wrong, and the relationship between the two should be as shown in Fig. 8, where the camera is vertically placed above the calibration board.

We also admit that it is better to project the laser vertically onto the calibration board when the light plane is calibrated. When the laser is vertically projected on the object, the light intensity presents a Gaussian distribution. When the laser is projected non vertically on the object, that is, there is a certain angle between the laser and the normal of the object surface, the light intensity presents a non-Gaussian distribution. In this case, it is difficult to extract the light stripe center effectively by traditional light stripe center extraction algorithms.

In our manuscript, we innovatively proposed an algorithm to extract the center of light stripe with uneven light intensity. So, the three-dimensional model that the laser generator is at a certain angle with the calibration board is adopted. The angle between the laser generator and the normal of the calibration board is 45 degrees. The distance between the CCD camera and the calibration board is adjusted to make the calibration board fall in the field of view of the CCD camera, and the distance should also meet the requirements of clear light stripe and complete measured object when the light stripe is projected on the measured object. Therefore, the distance between the CCD camera and the calibration board is 300 mm in our study.

We have modified Figure 1 and added a detailed overview of calibration and measurement in our revised version, which can be seen from line 143 to line 148 in our revised manuscript.

 

Question 3:

The laser line was projected to various places on the board? Taking Fig. 8 into account, the angle and the place where the laser line was projected were varying. 

Response:

We thank you for such a detailed question. When the camera is calibrated, the camera position does not change, and the calibration board changes position relative to the camera. When the light plane is calibrated, the laser generator position does not change, and the calibration board changes position relative to the laser generator. In 3D model and actual measurement, the projection angle and position of laser line are invariable.

 

Question 4:

What are the optical properties of the camera in combination with the lens, namely depth of field? 

Response:

Thank you for your professional question. We used a laser and a color CCD industrial camera and a lens to form a line structured light camera. The optical properties of our line structured light camera are that FOV is H 53°x V 28° and the working range is 30 mm to 2000 mm, which has been added in our revised manuscript from line 368 to line 370.

 

Question 5:

Calibration with resulting accuracy is valid for the distance to the board as well for the distance to the gauge block which is relatively thick (see Fig. 9)?

Response:

     Think you for your good comment. In fact, we took into account the question you asked when we were doing experiments. We repeatedly adjusted the distance between the camera and the calibration board, so that the calibration could get accurate results at this distance, while the accuracy of measurement is not affected by the thickness of the gauge block. Therefore, the model combined with the proposed algorithm could get accurate measurement values of the surface parameters of the gauge block.

 

Question 6:

Could authors calculate line/pixel measurement uncertainty with regard to the accuracy of the calibration board (0.01 mm) with respect to the used measurement distances and angles?" etc.

 

Response:

     We think this is an excellent comment. We appreciate your good comment. In this study, we focus on the extraction algorithm of light stripe center when the light stripe does not have regular light intensity distribution. The accuracy of the algorithm for extracting the light stripe center is verified by comparing the pixel deviation of the corresponding position of the proposed algorithm and the standard method. The validity of the proposed algorithm is verified by experiments with standard gauge blocks and calibrated model. It will be the focus of our future research that calibration error modeling and analysis of the line structured light system, which will include pixel measurement uncertainty analysis and so on.

Author Response File: Author Response.docx

Reviewer 2 Report

This manuscript proposes to use light intensity equalization and adaptive algorithm to find light stripe center more accurately in order to improve the accuracy of line structure light based measurement system. The experimental result has shown the effectiveness of the proposed method with the comparison between skeleton method and gray centroid method. The proposed concept is a good reference for the improvement of line structure light system and  can be considered for the publication in Processes. There is a   minor comment for the authors' reference to enhance the manuscript on the discussion of the experimental result.

Based on Fig. 7 of the manuscript, the best result looks like the one from standard method in terms of line center identification. But that line stripe looks a bit tilted. The proposed method shows a good contrast, but there is local wiggle of line center, i.e. the line center does not look quite straight. The average maximum deviation of 0.021 pixel is based on relative deviation from 50 pictures, but not absolute value with respect to a reference line. A visible wiggle in the picture should imply that the deviation is already more than at least half pixel. It is suggested to put more discussion on the above issues.

Author Response

Dear reviewer,

Because our response includes some images, please check out the attachment where we have uploaded our response file.

Sincerely,

Authors

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The authors have responded to the questions and comments in a suitable and professionally competent manner and significantly improved the paper. In my opinion, the paper publishing may interest readers of the Processes journal and enhance the field of research as well as practical applications on the topic.

My last proposal is to adjust fonts in Figs. 2 and 3 to make them consistent with the paragraph fonts.

Author Response

Reviewer 1

The authors have responded to the questions and comments in a suitable and professionally competent manner and significantly improved the paper. In my opinion, the paper publishing may interest readers of the Processes journal and enhance the field of research as well as practical applications on the topic.

Question:

My last proposal is to adjust fonts in Figs. 2 and 3 to make them consistent with the paragraph fonts.

Response:

    We appreciate your recognition for our first revision. And we have adjusted the fonts in Figs. 2 and 3, which can be seen in line 182 and line 235 of the second revision.

Author Response File: Author Response.docx