Intelligent Alignment Monitoring Method for Tortilla Processing Based on Machine Vision

Round 1

Reviewer 1 Report

1. The Abstract is long-winded and contains many insignificant sentences. It should be shortened, stating the purpose that was assumed in the preparation of this manuscript and the experiments performed, what was done, and the most important conclusions of the research. When the manuscript is the result of a literature review also the most important data and findings of the study. The abstract should briefly inform the reader about the paper's content so that the reader is informed to what extent the article is the information the reader is looking for.  

2 The content of the rest of the manuscript also needs serious repair. The authors did not clearly state whether it is the result of a research paper or a literature review. After reading the conclusions, one realizes there is a rather research article, but it is not so much about research as it is about training a method to perform research. I propose introducing the classic formal style of a research article into the manuscript by introducing specific names of subsections and filling them with appropriate content.

 In the "Methods" section, I propose to describe the specific method chosen and justify the choice. Describe the construction of the test bed, specifically the one that was built, and not explain the theory of identifying objects by the two-eye method and the idea of the construction of the test bed and how it was designed. Please show pictures of the things evaluated or used to train the system and not a picture of the chessboard. 

In the chapter "Experimental and analysis," you should describe the specific test trials, the number of repetitions and all the results, and the effects of statistical processing of the results obtained, not just examples.

The next chapter that should be introduced is "Discussion."  In it, you should provide critical or positive opinions concerning the results obtained, considering the citation of at least some items from the previously cited literature. If this chapter cannot refer to the previously mentioned literature, it means that the review of this literature was poorly done.

The last chapter: "Conclusion," should be improved completely. In this manuscript, the "Conclusion" chapter includes a abstract of the manuscript, and this should be in the "Abstract" and not in the "Conclusion". This chapter should only contain the specific conclusions of the study.

If this manuscript is purely a review, it also needs a lot of additions and is unsuitable for publication in its present form.

Author Response

Comments and Suggestions for Authors

1. The Abstract is long-winded and contains many insignificant sentences. It should be shortened, stating the purpose that was assumed in the preparation of this manuscript and the experiments performed, what was done, and the most important conclusions of the research. When the manuscript is the result of a literature review also the most important data and findings of the study. The abstract should briefly inform the reader about the paper's content so that the reader is informed to what extent the article is the information the reader is looking for.  

Response: Thank you so much for your efforts on improving the quality of this paper. We have revised the abstract to highlight the motivation and contribution of this article.

2 The content of the rest of the manuscript also needs serious repair. The authors did not clearly state whether it is the result of a research paper or a literature review. After reading the conclusions, one realizes there is a rather research article, but it is not so much about research as it is about training a method to perform research. I propose introducing the classic formal style of a research article into the manuscript by introducing specific names of subsections and filling them with appropriate content.

Response: Thank you so much for your efforts on improving the quality of this paper. We have addressed this issue.

 In the "Methods" section, I propose to describe the specific method chosen and justify the choice. Describe the construction of the test bed, specifically the one that was built, and not explain the theory of identifying objects by the two-eye method and the idea of the construction of the test bed and how it was designed. Please show pictures of the things evaluated or used to train the system and not a picture of the chessboard. 

Response: Thank you so much for your efforts on improving the quality of this paper. We have addressed this issue.

In the chapter "Experimental and analysis," you should describe the specific test trials, the number of repetitions and all the results, and the effects of statistical processing of the results obtained, not just examples.

Response: Thank you so much for your efforts on improving the quality of this paper. We have revised the relevant description of the experiment.

The next chapter that should be introduced is "Discussion."  In it, you should provide critical or positive opinions concerning the results obtained, considering the citation of at least some items from the previously cited literature. If this chapter cannot refer to the previously mentioned literature, it means that the review of this literature was poorly done.

Response: Thank you so much for your efforts on improving the quality of this paper. We have replaced “Conclusion” to “Discussion” and added positive opinions concerning the results obtained.

The last chapter: "Conclusion," should be improved completely. In this manuscript, the "Conclusion" chapter includes a abstract of the manuscript, and this should be in the "Abstract" and not in the "Conclusion". This chapter should only contain the specific conclusions of the study.

Response: Thank you so much for your efforts on improving the quality of this paper. We have revised the Conclusion section and added future work.

 

 

Reviewer 2 Report

This paper studies the problem of alignment monitoring on noodles and a machine vision-based monitoring method is proposed. The matter is of interest. However, the motivation of the article is not clear enough, and the proposed algorithm does not seem to be special. If noodles are replaced with other products, this vision-based monitoring system may also play the same role. I have the following comments on this manuscript:

1. In the Abstract, the authors mentioned “it is necessary to improve the physical properties of corn flour and combine some improved formulations to change the hysicochemical and processing properties of corn flour”. It seems that this is the motivation of this article, however, what is the relationship between “physical properties” and “noodle positioning monitoring”?

2. In the Introduction, if the authors could introduce some existing work and their drawbacks, that would be better.

3. In Sec. 2, It is difficult to understand why the “Corn pasta” is a related work. Does it have anything to do with the monitoring system or method?

4. In Sec.3, “binocular stereo vision” is a typical and traditional way and “Displacement measurement algorithm” is an old method as well. The whole section mostly explains the traditional methods and lacks innovation. Since the author mentioned "We designed a binary vision-based vision monitoring system" in Sec. 1, the authors should at least introduce how to design each part of the system.

5. In Sec.4, concerning training and testing, the author did not mention what network model was used. In addition, what is “Traditional manual monitoring” and what is the difference from your work?

Furthermore, its presentation and use of English should be improved. There are too many errors in English writing and formula formats throughout the entire article. The manuscript is hard to follow.

Author Response

Comments and Suggestions for Authors

This paper studies the problem of alignment monitoring on noodles and a machine vision-based monitoring method is proposed. The matter is of interest. However, the motivation of the article is not clear enough, and the proposed algorithm does not seem to be special. If noodles are replaced with other products, this vision-based monitoring system may also play the same role. I have the following comments on this manuscript:

1. In the Abstract, the authors mentioned “it is necessary to improve the physical properties of corn flour and combine some improved formulations to change the hysicochemical and processing properties of corn flour”. It seems that this is the motivation of this article, however, what is the relationship between “physical properties” and “noodle positioning monitoring”?

Response: Thank you so much for your efforts on improving the quality of this paper. We have revised the Abstract section and highlighted the motivation of this paper.

2. In the Introduction, if the authors could introduce some existing work and their drawbacks, that would be better.

Response: Thank you so much for your efforts on improving the quality of this paper. We have revised the Introduction section.

3. In Sec. 2, It is difficult to understand why the “Corn pasta” is a related work. Does it have anything to do with the monitoring system or method?

Response: Thank you so much for your efforts on improving the quality of this paper. We have revised the Related Work section.

4. In Sec.3, “binocular stereo vision” is a typical and traditional way and “Displacement measurement algorithm” is an old method as well. The whole section mostly explains the traditional methods and lacks innovation. Since the author mentioned "We designed a binary vision-based vision monitoring system" in Sec. 1, the authors should at least introduce how to design each part of the system.

Response: Thank you so much for your efforts on improving the quality of this paper. We have added related description.

 

5. In Sec.4, concerning training and testing, the author did not mention what network model was used. In addition, what is “Traditional manual monitoring” and what is the difference from your work?

Response: Thank you so much for your efforts on improving the quality of this paper. We have added the network model and the difference between Traditional manual monitoring and our work.

Furthermore, its presentation and use of English should be improved. There are too many errors in English writing and formula formats throughout the entire article. The manuscript is hard to follow.

Response: Thank you so much for your efforts on improving the quality of this paper. We have polished the English of our paper.

Round 2

Reviewer 1 Report

The authors have made major changes to the text of the manuscript, improving the understanding of the work as a whole. Nevertheless, there are still drone deficiencies that make it difficult to understand :

- (in section 4.3 Monitoring accuracy result) the word "instances" seems to have been used unhappily in the context of the data presented in Table 1. It would be appropriate for the authors to say what it means in this case. Is it the number of tests performed or the number of noodles tested? How does the number of "instances" relate to manual monitoring of noodles?

- Table 1 also needs improvement. Regardless of whether the text, above and below the table, tried to explain what the first column in this table means, its name should still be included in the table.

- Since the inference in this paper about the superiority of the described method of evaluating noodles is based on the results that Table 1 contains, there should be a thorough description in the Methods section of the manuscript of how the experiment was conducted and at least a mention of how the results were statistically evaluated.

Author Response

Comments and Suggestions for Authors

The authors have made major changes to the text of the manuscript, improving the understanding of the work as a whole. Nevertheless, there are still drone deficiencies that make it difficult to understand :

 

- (in section 4.3 Monitoring accuracy result) the word "instances" seems to have been used unhappily in the context of the data presented in Table 1. It would be appropriate for the authors to say what it means in this case. Is it the number of tests performed or the number of noodles tested? How does the number of "instances" relate to manual monitoring of noodles?

Response: We have clarified the meaning of "instances" in the text and explained it in detail to facilitate the reader's understanding of section 4.3.

- Table 1 also needs improvement. Regardless of whether the text, above and below the table, tried to explain what the first column in this table means, its name should still be included in the table.

Response: We have revised Table 1.

- Since the inference in this paper about the superiority of the described method of evaluating noodles is based on the results that Table 1 contains, there should be a thorough description in the Methods section of the manuscript of how the experiment was conducted and at least a mention of how the results were statistically evaluated.

Response: We have added a description of how to statistically evaluate the results in Table 1.

Reviewer 2 Report

I appreciate the author’s efforts and their response to the previous version’s comments. However, some responses are not clear.

1. The English of the article needs a lot of editing work. A large number of grammatical errors make the article difficult to read. For example, figure 9 or Figure 9 on Page 10, “In order to” or “in order to” on Page 1, the coordinate system is O-xyz or o-xyz, make it consistent.

2.  The author still hasn't explained how the monitoring system is designed. Sec. 3 has sections 3.1 and 3.2. What is the relationship between these two sub-sections, and what is the purpose of the mathematical analysis work in section 3.1? It seems that it is only discussing the concept of binary stereo vision rather than the vision monitoring system.

3. The structure of the neural network is not clear since the authors only mentioned 128*128*1. It is better to use a figure to show the network used in this paper.

 

4. Considering the comparative experiment, the authors compared a method 45 years ago. This makes this article lose its competitiveness. At the same time, the results of the traditional method are doubtful (the last column of Table 1). If the authors could explain how to calculate these results by using the traditional monitoring method, that would be better. 

Author Response

Comments and Suggestions for Authors

I appreciate the author’s efforts and their response to the previous version’s comments. However, some responses are not clear.

 

1. The English of the article needs a lot of editing work. A large number of grammatical errors make the article difficult to read. For example, figure 9 or Figure 9 on Page 10, “In order to” or “in order to” on Page 1, the coordinate system is O-xyz or o-xyz, make it consistent.

Response: We have addressed the grammatical errors issue and improved the English of this article.

 

2.  The author still hasn't explained how the monitoring system is designed. Sec. 3 has sections 3.1 and 3.2. What is the relationship between these two sub-sections, and what is the purpose of the mathematical analysis work in section 3.1? It seems that it is only discussing the concept of binary stereo vision rather than the vision monitoring system.

Response: We have explained how to design the monitoring system.

3. The structure of the neural network is not clear since the authors only mentioned 128*128*1. It is better to use a figure to show the network used in this paper.

Response: We have added a figure to show the network used in this paper.

 

4. Considering the comparative experiment, the authors compared a method 45 years ago. This makes this article lose its competitiveness. At the same time, the results of the traditional method are doubtful (the last column of Table 1). If the authors could explain how to calculate these results by using the traditional monitoring method, that would be better.

Response: We have added a description of how to statistically evaluate the results in Table 1.