Color and Texture Analysis of Textiles Using Image Acquisition and Spectral Analysis in Calibrated Sphere Imaging System-I

Round 1

Reviewer 1 Report (Previous Reviewer 4)

(1)Please check the manuscript carefully to remove the typos, improve the language and format. E.g.

-The reference numbers should be in square brackets rather than round brackets.

-Some spaces are missing, such as “Funding:Some” Line 342.

-The line space of Para. 1, Section 1 is larger.

-The reference formats are not uniform.

...

(2)Many references were published too early (e.g. 1949). Is it necessary to cite so early works?

(3)It is difficult to find the originality of this paper. Sections 2~4 are simply. The experiments are insufficient.

Author Response

(1)Please check the manuscript carefully to remove the typos, improve the language and format. E.g.

-The reference numbers should be in square brackets rather than round brackets.

-Some spaces are missing, such as “Funding: Some” Line 342.

-The line space of Para. 1, Section 1 is larger.

-The reference formats are not uniform.

Ans: Thanks, we have corrected it.

(2)Many references were published too early (e.g. 1949). Is it necessary to cite so early works?

Ans: We have added 15 recent literatures with explanation.  We restructured manuscript also with 3 paragraphs.

1. Liu, L., Xie, X., Zhang, Y., Cao, F., Liang, J., & Liao, N. Colorimetric characterization of color imaging systems using a multi‐input PSO‐BP neural network. Color Research & Application. 2022. doi: 10.1002/col.22772
2. Wang, H., Yang, J., Xue, B., Yan, X., & Tao, J. A novel color calibration method of multi-spectral camera based on normalized RGB color model. Results in Physics, 2020, 19,
3. Bondi, L., Baroffio, L., Güera, D., Bestagini, P., Delp, E. J., & Tubaro, S. First steps toward camera model identification with convolutional neural networks. IEEE Signal Processing Letters, 2016, 24(3), 259-263.
4. Lukac, R. (Ed.). (2018). Single-sensor imaging: methods and applications for digital cameras. CRC Press.
5. Nayak, S., Khandual, A., & Mishra, J. Ground truth study on fractal dimension of color images of similar texture. The journal of the Textile Institute, 2018, 109(9), 1159-1167.
6. Menna, F., Nocerino, E., & Remondino, F. Flat versus hemispherical dome ports in underwater photogrammetry. The International Archives of Photogrammetry, Remote Sensing and Spatial Information Sciences, 2017, 42, 481.
7. Khandual,A.,Baciu & Rout,N. Colorimetric processing of digital color image!, J. of Adv.

Res. In Comp. Sc. And Soft.Engg., 2013, 3(7),pp: 103-107.

8. Yao, P. Advanced textile image analysis based on multispectral color reproduction. 2022, PhD dissertation.HKPolyU
9. Liu, T., Siegel, E., & Shen, D. Deep Learning and Medical Image Analysis for COVID-19 Diagnosis and Prediction. Annual Review of Biomedical Engineering, 2022, 24.
10. Gomez-Gonzalez, E., Barriga-Rivera, A., Fernandez-Muñoz, B., Navas-Garcia, J. M., et. al. Optical imaging spectroscopy for rapid, primary screening of SARS-CoV-2: a proof of concept. Scientific reports, 2022, 12(1), 1-18.
11. Nguyen, K., Proença, H., & Alonso-Fernandez, F.. Deep Learning for Iris Recognition: A Survey. 2022, arXiv preprint arXiv:2210.05866.
12. Zhao, J., Kumar, A., Banoth, B. N., Marathi, B., Rajalakshmi, P., Rewald, B., et. al.. Deep-Learning-Based Multispectral Image Reconstruction from Single Natural Color RGB Image—Enhancing UAV-Based Phenotyping. Remote Sensing,2022, 14(5),
13. Kakarla, S. C., Zhu, Z., Ampatzidis, Y., Fountas, S., Ehsani, R., & Pardalos, P. Artificial Intelligence Applications in Specialty Crops. Frontiers in Plant Science, 2022, 13, 866724-866724.
14. Arad, B., Timofte, R., Yahel, R., Morag, N., Bernat, A., Cai, Y., & Roomi, S. NTIRE 2022 Spectral Recovery Challenge and Data Set. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition 2022, (pp. 863-881).
15. Zhang, J., Su, R., Fu, Q., Ren, W., Heide, F., & Nie, Y. A survey on computational spectral reconstruction methods from RGB to hyperspectral imaging. Scientific reports, 2022, 12(1), 1-17.

(3)It is difficult to find the originality of this paper. Sections 2~4 are simply. The experiments are insufficient.

Ans: Thanks, we have revised sincerely and highlighted the changes and included spectral and imaging experimental data in table 3 and 4. Hope you will accept

Author Response File: Author Response.pdf

Reviewer 2 Report (Previous Reviewer 3)

The authors have proposed a calibration technique and described how unique digital color signatures can be derived from calibrated RGB derivatives to extract the best features for color and texture. 

The article is well-structured in terms of pictures and tables . However in the section related work :

  104  utilising complex models like polynomial regression and neural networks[], even some 


.    105  recent research have reported mediocre colour accuracy (61–67). In particular, a recent 


.    106  thesis by Yao, P. (2022) demonstrated the typical colour difference mistakes for various 


.    107  textile materials (68). 


[] I suggest to cite the following articles:

Sciuto, G. L., Susi, G., Cammarata, G., & Capizzi, G. (2016, June). A spiking neural network-based model for anaerobic digestion process. In 2016 International Symposium on Power Electronics, Electrical Drives, Automation and Motion (SPEEDAM) (pp. 996-1003). IEEE.

Tuccitto, N., Bombace, A., Torrisi, A., Licciardello, A., Sciuto, G. L., Capizzi, G., & Woźniak, M. (2019). Probabilistic neural network-based classifier of ToF-SIMS single-pixel spectra. Chemometrics and Intelligent Laboratory Systems, 191, 138-142.

Author Response

Ans : Many Thanks for accepting .

The authors have proposed a calibration technique and described how unique digital color signatures can be derived from calibrated RGB derivatives to extract the best features for color and texture. 

The article is well-structured in terms of pictures and tables. However in the section related work:

Ans: Thanks for your appreciation.

 104 utilising complex models like polynomial regression and neural networks [], even some 
.

 105 recent research have reported mediocre colour accuracy (61–67). In particular, a recent 


106 thesis by Yao, P. (2022) demonstrated the typical colour difference mistakes for various 


107 textile materials (68). 


[] I suggest to cite the following articles:

Sciuto, G. L., Susi, G., Cammarata, G., & Capizzi, G. (2016, June). A spiking neural network-based model for anaerobic digestion process. In 2016 International Symposium on Power Electronics, Electrical Drives, Automation and Motion (SPEEDAM) (pp. 996-1003). IEEE.

Tuccitto, N., Bombace, A., Torrisi, A., Licciardello, A., Sciuto, G. L., Capizzi, G., & Woźniak, M. (2019). Probabilistic neural network-based classifier of ToF-SIMS single-pixel spectra. Chemometrics and Intelligent Laboratory Systems, 191, 138-142.

Ans: Thanks , we have corrected and added the suggested reference.

  Complex models with use of spiking neural network-based model [59] and probabilistic neural network-based classifier [60] were reported for reasonable spectral accuracy.

Author Response File: Author Response.pdf

Reviewer 3 Report (Previous Reviewer 1)

In the manuscript “Color and texture analysis by image acquisition and spectral analysis of textiles in calibrated sphere imaging system-I” is proposed an integrated sphere imaging system with a simple calibration technique. The authors describe how unique digital color signatures can be derived from calibrated RGB derivatives to extract the best features for color and texture.

Although, the authors made some improvements, in my opinion, they are not enough to make the manuscript suitable for publication.

My comments are as follows:

The authors have maintain the structure of the document, which, as I referred in my first revision is adequate.

Related work should be a section, instead a subsection of the Introduction.

Lines 103-105 - The sentence "By utilizing..." is not understandable.

Line 109 - "This study applied a methodical approach to the issue..." What issue?

Lines 109-123 - The paragraph should be in the Introduction section instead the Related Work subsection.

The dataset used in the experiments must be presented and characterized in the manuscript.

The description of the proposed method still lacks detail.

The Conclusion section shouldn’t begin with “These experimental results…”. This section must be rewritten.

Author Response

My comments are as follows:

The authors have maintain the structure of the document, which, as I referred in my first revision is adequate.

Related work should be a section, instead a subsection of the Introduction.

Lines 103-105 - The sentence "By utilizing..." is not understandable.

Line 109 - "This study applied a methodical approach to the issue..." What issue?

Lines 109-123 - The paragraph should be in the Introduction section instead the Related Work subsection.

The dataset used in the experiments must be presented and characterized in the manuscript.

The description of the proposed method still lacks detail.

The Conclusion section shouldn’t begin with “These experimental results…”. This section must be rewritten.

Ans: Many thanks for your sincere comments for improvement. We have revised it thoroughly and rewritten the conclusion. We added all experimental data as suggested.

This study applied a methodical approach to for improving the colorometric accuracy by creating a unique system with a highly diffused coating on the integrating sphere, controlled illumination, viewing, and calibration for image acquisition

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report (Previous Reviewer 4)

(1)The length of this paper can be shortened. Please do not list too long data tables. Some unnecessary data can be removed. The authors should highlight and clearly explain the regulars, phenomena and conclusions from the results. Some too early (before 2010) references can be removed unless they are necessary.

(2)The caption of Fig. 5 is unclear. How many colors? 20 colors? What do the different colors denote? There are only 7 colors in the caption, including r, g, b, rg, rb, gb and rgb. The same to Fig. 6.

Author Response

(1)The length of this paper can be shortened. Please do not list too long data tables. Some unnecessary data can be removed. The authors should highlight and clearly explain the regulars, phenomena and conclusions from the results. Some too early (before 2010) references can be removed unless they are necessary.

Ans- Thanks, we removed those and revised fully.

(2)The caption of Fig. 5 is unclear. How many colors? 20 colors? What do the different colors denote? There are only 7 colors in the caption, including r, g, b, rg, rb, gb and rgb. The same to Fig. 6.

Ans: , Sorry, these are 20 samples with closer textures. We added to explain.  Figure 5. SCE reflectance   and   r, g, b, rg, rb, gb and rgb responses of 20 fabrics. 

The supplementary file contained the image details, together with spectral and digital RGB polynomial data, of 20 different fabrics made from the same cotton fibres (only texture is varied in close proximity by spinning for making yarn and weaving for the fabric development). The aim was to verify the unique presence of colors with respect to their spectra in visible wavelengths, which could lead further to ensure quantitative and qualitative analysis of color.

Author Response File: Author Response.pdf

Reviewer 3 Report (Previous Reviewer 1)

In the manuscript “Color and texture analysis by image acquisition and spectral analysis of textiles in calibrated sphere imaging system-I” is proposed an integrated sphere imaging system with a simple calibration technique. The authors describe how unique digital color signatures can be derived from calibrated RGB derivatives to extract the best features for color and texture.

Although, the authors made considerable and important improvements, covering most of my comments and suggestions, in my opinion, there is still an important issue that has not been addressed correctly.

To address my previous comment “The dataset used in the experiments must be presented and characterized in the manuscript”, the authors chose to include two large tables in the manuscript (RGB data and Spectral Data). In my opinion, these tables do not add any value to the manuscript.

I think that the discussion section should be included in the Results section. What I mean is that the authors should discuss the results as they are presented (not in a separate section).

Please, revise the numbering of the section… There are a section 5, after a section 6.

Remove square brackets from figure 4 caption.

Please, revise the caption text of table 2.

Table 2 is unformatted, not realizing which parameter each value corresponds to. This table should be referred in the text.

Author Response

Although, the authors made considerable and important improvements, covering most of my comments and suggestions, in my opinion, there is still an important issue that has not been addressed correctly.

To address my previous comment “The dataset used in the experiments must be presented and characterized in the manuscript”, the authors chose to include two large tables in the manuscript (RGB data and Spectral Data). In my opinion, these tables do not add any value to the manuscript.

Ans: Thanks for your suggestion. We have revised accordingly. The complete data set with images  used in the experiments are presented in the supplementary file and explanations were added  in the manuscript. We propose this novel method to avoid the errors from theoretical characterization and sufficient references on it has been described. We will discuss further investigation on  characterization, validated the responses in various color spaces, quantitative color evaluation in the next next paper. It would be too long to discuss in this paper.

Kindly Consider

I think that the discussion section should be included in the Results section. What I mean is that the authors should discuss the results as they are presented (not in a separate section).

Ans : We Included in results section now.

Please, revise the numbering of the section… There are a section 5, after a section 6.

Ans : We corrected.

Remove square brackets from figure 4 caption.

Ans : Sorry, we removed [] from fig 3 and 4.

Please, revise the caption text of table 2.

Ans : we revised.

Table 2 is unformatted, not realizing which parameter each value corresponds to. This table should be referred in the text.

Ans : Sorry, we formatted table 2.

Author Response File: Author Response.pdf

Round 3

Reviewer 1 Report (Previous Reviewer 4)

Accept.

Author Response

Many Thanks for accepting our revision.

Reviewer 3 Report (Previous Reviewer 1)

Citation formatting should be uniform. For instance, in Page 2 – Line 93, the citation format is different from the one used in Page 3 – Line 98, and different from Page 3 – Line 105.

The manuscript should be revised carefully. There are still some typos. For instance, Page 2 – Lines 96-97.

Author Response

Citation formatting should be uniform. For instance, in Page 2 – Line 93, the citation format is different from the one used in Page 3 – Line 98, and different from Page 3 – Line 105.

The manuscript should be revised carefully. There are still some typos. For instance, Page 2 – Lines 96-97.

Ans: We sincerely thank you for noticing it. We are sorry and  have corrected it now.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

In the manuscript “Color and texture analysis by image acquisition and spectral analysis of textiles in calibrated sphere imaging system-I” is proposed an integrated sphere imaging system with a simple calibration technique. The authors describes how unique digital color signatures can be derived from calibrated RGB derivatives to extract the best features for color and texture.

Although, in my opinion, the authors present a work where they propose an interesting approach to the analysis of color and texture in textile materials, whose results seem promising, the manuscript has several flaws.

My comments are as follows:

The manuscript is poorly structured and not easy to follow.

In the Introduction section, the authors make, although superficially, a contextualization of the work developed, as well as a literature review, also presenting part of the developed method. This section should be properly divided into different sections. For instance, it should be included a “Related work” section.

Regarding the Introduction, in addition to the contextualization of the work, should its importance be highlighted and how does the proposed method contribute to scientific advancement in the area? What's the novelty of the work presented? What are the limitations of other approaches that this method overcomes?

Lines 33-39 – The sentence is not understandable. It must be rewritten.

The state of the art must be complemented with reference to recent works. The works cited are not current... The most recent reference I could find (only one) is from a decade ago (2012).

Lines 55-56 – The sentence makes no sense.

Lines 57-59 – It seems that some words are missing for the sentence to make sense.

Line 83 – In the sentence “The way they…” Who does "they" refer to?

The description of the proposed method lacks detail.

The dataset used in the experiments must be presented and characterized.

A discussion of the results is made, but no conclusions of the work are presented.

Reviewer 2 Report

The authors of the text deal with the issue of colour image capture and its interpretation. Many imaging applications, analyses, and evaluations require human perception (color) needs to transform the three-band color interpretation (RGB) color space to a CIE color space standard that is independent of the sensing device.
There are a number of factors that affect this transformation, in particular the image acquisition environment, theoretical conversion errors, display geometry, precisely defined illumination uniformity and calibration protocol limit the accuracy of their applicability. It is detrimental to the process that in most image processing applications, spectral data are either unavailable or unmeasurable.
The present study is based on the development of a new integrated sphere imaging system and on experiments with controlled variation of texture and color of textiles. It proposes a simple calibration technique and describes how unique digital color signatures can be obtained from calibrated RGB derivations. This is the way to obtain the best properties for color and texture. In addition, a reflectance evaluation function has been proposed, which is lacking in the imaging field and could be useful for visualizing, identifying and applying qualitative and quantitative analysis of color and texture.
Tests were performed on examples and the method was shown to have promising colorimetric results when validated in different color spaces, color combinations over three textures, and color characterization.
The thesis contains elements of novelty and is written on scientific foundations and procedures. It is of content of interest to multiple disciplines.
From an outside perspective, I have no major comments.
No plagiarism has been found.

Reviewer 3 Report

The authors present  a novel integrated sphere imaging system and experimentation with the textiles' controlled variation of texture and color. 

I think that the authors have to describe "very well" the technology and novelty in the field of electronics.

Reviewer 4 Report

The references were published too early, most were published before 2012. Are there any systems developed recently?

The authors did not compare their system with some SOTA methods. Are there none similar or related systems that can be compared with the authors’ system?

This work has somewhat application value, but academic value is low.