Fruit Powder Analysis Using Machine Learning Based on Color and FTIR-ATR Spectroscopy—Case Study: Blackcurrant Powders

Round 1

Reviewer 1 Report

The research highlights the potential of machine learning techniques combined with color space analysis and FTIR spectroscopy as valuable tools for characterizing and distinguishing currant powders based on their qualities. The findings of this research could contribute to the food industry, providing insights into the quality and homogeneity of currant powders, which may have implications for product development, quality control, and consumer satisfaction.

The introduction is clear and logical.

This work includes 4 Tables and 4 figures

Some specific comments:

Lines 22/26; 69-71: In general, the aim of a study is only one.....in this sentence appears twice, even if the second one comes in addition to the first one, I recommend rewording it so that the aim remains only one.

 

 Lines 135: It is recommended to specify the origin (company/country) of the equipment under study (where not specified).

Regarding the 3.3. Color analysis/ In Table 1, I recommend a comparison of the results with other studies from the literature (if available)

 For the fragment Lines 324-333, I suggest moving to Result and Discussions (between tables 3 and 4) and rewarding.  Conclusions to remain clear and precise (related to the purpose of the study)

Comments for author File: Comments.doc

Author Response

Dear Reviewer 1,

thank you very much for your valuable comments. I have made the changes as you recommended. This certainly improved and strengthened the substantive aspects of my study.

Response to Reviewer 1 Comments:

 

Review Manuscript ID review applsci-2543339. entitled “Fruit powders analysis using machine learning based on color 2 and FTIR spectroscopy. Case study: black currant powders.”  

The research highlights the potential of machine learning techniques combined with color space analysis and FTIR spectroscopy as valuable tools for characterizing and distinguishing currant powders based on their qualities. The findings of this research could contribute to the food industry, providing insights into the quality and homogeneity of currant powders, which may have implications for product development, quality control, and consumer satisfaction.

The introduction is clear and logical.

This work includes 4 Tables and 4 figures

Some specific comments:

Lines 22/26; 69-71: In general, the aim of a study is only one.....in this sentence appears twice, even if the second one comes in addition to the first one, I recommend rewording it so that the aim remains only one.

We corrected into the text. We agree with the reviewer's opinion. We removed the thread in 69-71 and merged it in 22-26 giving clarity and transparency to the aim of the study.

 

 Lines 135: It is recommended to specify the origin (company/country) of the equipment under study (where not specified).

Spectrum Two with ATR Diamond Accesory, made by PerkinElmer Ltd.  (USA). We corrected into the text.

 

Regarding the 3.3. Color analysis/ In Table 1, I recommend a comparison of the results with other studies from the literature (if available):

The authors did not notice that this section for Color Lab lacked the references that were present in the bibliography for 36-38. It is already corrected. Below the mentioned comparative citations:

36. Kulapichitr, F.; Borompichaichartkul, C.; Fang, M.; Suppavorasatit, I.; Cadwallader, K.R. Effect of Post-Harvest Drying Process on Chlorogenic Acids, Antioxidant Activities and CIE-Lab Color of Thai Arabica Green Coffee Beans. Food Chem 2022, 366, 130504, doi:10.1016/J.FOODCHEM.2021.130504.
37. Pawlak, T.; Pilarska, A.A.; Przybył, K.; Stangierski, J.; Ryniecki, A.; Cais-Sokolińska, D.; Pilarski, K.; Peplińska, B. Application of Machine Learning Using Color and Texture Analysis to Recognize Microwave Vacuum Puffed Pork Snacks. Applied Sciences 2022, 12, 5071, doi:10.3390/app12105071.
38. Przybył, K.; Wawrzyniak, J.; Samborska, K.; Gierz, Ł.; Koszela, K.; Szychta, M. Application of Artificial Neural Networks in Recognizing Carrier Based on the Color of Raspberry Powders Obtained in the Spray-Drying Process. In Proceedings of the Fourteenth International Conference on Digital Image Processing (ICDIP 2022); Xie, Y., Jiang, X., Tao, W., Zeng, D., Eds.; SPIE, October 12 2022; Vol. 12342, p. 172.

We corrected into the text.

 

The fragment from Lines 324-333, I suges  to move to Result and discusions (between table 3 and 4) and rewarding.  Conclusions to remain clear and precise (related to the purpose of the study)

We corrected into the text. We moved the paragraph as suggested by the reviewer.

 

 

Kind regards,

Krzysztof Przybył

 

 

Author Response File: Author Response.docx

Reviewer 2 Report

1. Title: The term FTIR spectroscopy could be revised as FTIR-ATR spectroscopy

2. Introduction. The authors did not mention well the main reason why focus on fruits black currants. What are the important properties of this fruit? Why needs preservation for black currant? Why drying?

3. Materials and Methods. The reason for carrier selection and specific proportion in solution (30 and 50%) is not clear. Please consider adding more detailed information.

4. Please revise.

2.2. Moisture content (It should be 2.2. Water activity).

5. Please revisit.

2.4. Fourier-transformation infrared spectroscopy (It should be 2.4. Fourier-transform infrared spectroscopy).

6. How many grams are put on the surface of ATR? Only one spectrum for one sample? Any replication for spectral acquisition?

7. Please revise Figure 2. It is not easy to read the graph. Is the moisture content data continuous? I think your moisture content data is discrete data.

8. Please revise Figure 4. Put additional information on the x-axis (wavenumber?) and y-axis (absorbance?)

Author Response

Dear Reviewer 2,

thank you very much for your valuable comments. I have made the changes as you recommended. This certainly improved and strengthened the substantive aspects of my study.

Response to Reviewer 2 Comments:

Comments and Suggestions for Authors

1. Title: The term FTIR spectroscopy could be revised as FTIR-ATR spectroscopy

We corrected into the text. We changed the text in the title as suggested by the reviewer.

2. Introduction. The authors did not mention well the main reason why focus on fruits black currants. What are the important properties of this fruit? Why needs preservation for black currant? Why drying?

In the Introduction part, the authors have included the issues raised by the reviewer 2. In the opinion of reviewer 1, they also claim that it is transparently and clearly presented how important these fruits and fruit drying are, exactly it presented in lines: 34-72. However, they improved the target taking into account the cost-effectiveness aspect of powder production.

3. Materials and Methods. The reason for carrier selection and specific proportion in solution (30 and 50%) is not clear. Please consider adding more detailed information.

The study concentrates observation on the extreme proportions of the carrier contained in the currant solution, i.e. the smallest (30% d.s. content named as: 30) and the largest (50% d.s. content named as: 50) proportion of them. This will identify the point at which the best physicochemical properties of currant powders are obtained. As a result, studying currant solutions with extreme proportions of the chosen carrier can contribute to more effective scaling of the drying process under industrial conditions while maintaining other (intermediate) proportions.

We corrected into the text.

4. Please revise.

               2.2. Moisture content (It should be 2.2. Water activity).

We corrected into the text. It our mistake it should be water activity. We change the text as suggested by the reviewer.

5. Please revisit.

               2.4. Fourier-transformation infrared spectroscopy (It should be 2.4. Fourier-transform infrared spectroscopy).

We corrected into the text. It our mistake it should be water activity. We change the text as suggested by the reviewer.

6. How many grams are put on the surface of ATR? Only one spectrum for one sample? Any replication for spectral acquisition?

The spectra of the test samples of 0.005±0.005g were recorded as a function of absorbance in the wave number range of 4000-500 cm^-1, with a resolution of 0.9 cm^-1. Each spectrum is the result of the superimposition of five replicates

7. Please revise Figure 2. It is not easy to read the graph. Is the moisture content data continuous? I think your moisture content data is discrete data.

The authors agree that moisture content data is discrete data. The authors corrected this graph. As part of clarity, data visualization using Python was presented. We corrected into the text.

8. Please revise Figure 4. Put additional information on the x-axis (wavenumber?) and y-axis (absorbance?)

We revise Figure 4. We corrected into the text.

 

Kind regards,

Krzysztof Przybył

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

I consider this form suitable for publication

Reviewer 2 Report

The authors have replied well all the comments.