A Study on the Non-Contact Artificial Intelligence Elevator System Due to the Effect of COVID-19

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

 Referee report

This paper introduces a smart elevator system (EVAI) that utilizes artificial intelligence for non-contact operation, which is particularly relevant during the COVID-19 pandemic. The system is designed to reduce physical contact with elevator buttons, thereby minimizing the risk of virus transmission. The EVAI system incorporates AI facial recognition technology to identify residents and automatically select their desired floor based on the time of day and the resident's previous usage history. The experimental results of the proposed system showed a more accurate facial recognition function and a reduction in waiting time and travel time. Therefore, I think that the paper makes a contribution and has the potential to be published.  However, I summarize in the GENERAL COMMENTS as follows:

 

GENERAL COMMENTS

 

1. In the introduction section, this section could benefit from a brief literature review of existing elevator technologies and previous attempts to improve efficiency in elevator systems. This would situate the study within the broader context of the field and demonstrate how the proposed system advances or differs from current practices.

2. In the Proposed EVAI System section, while the system's high-level operation is described, there is a potential lack of detailed technical specifications. For instance, the manuscript could provide more in-depth information on the hardware and software requirements, system architecture, and the specific algorithms used for facial recognition and elevator control logic.

3. In the Proposed EVAI System section, the manuscript could provide more information on how the proposed system integrates with existing elevator and building management systems. Details on compatibility, installation processes, and potential challenges or costs associated with system integration would be valuable.

4. In the Experiment and Results section, the experiments primarily demonstrate the performance of the proposed system in isolation. There is a lack of comparative data against traditional elevator systems or other non-contact alternatives, which makes it difficult to assess the relative improvements and benefits of the new system.

 

5. In some formulas, such as formulas 1-4, mathematical variables should be italicized.

 

 

Comments for author File: Comments.pdf

Author Response

Dear
I would like to express my sincere gratitude to the reviewers who reviewed my submitted paper and provided advice to improve its quality. I have reflected the points as much as possible and added contents and figures are marked in red. 

This paper introduces a smart elevator system (EVAI) that utilizes artificial intelligence for non-contact operation, which is particularly relevant during the COVID-19 pandemic. The system is designed to reduce physical contact with elevator buttons, thereby minimizing the risk of virus transmission. The EVAI system incorporates AI facial recognition technology to identify residents and automatically select their desired floor based on the time of day and the resident's previous usage history. The experimental results of the proposed system showed a more accurate facial recognition function and a reduction in waiting time and travel time. Therefore, I think that the paper makes a contribution and has the potential to be published. However, I summarize in the GENERAL COMMENTS as follows:

Q 1 - In the introduction section, this section could benefit from a brief literature review of existing elevator technologies and previous attempts to improve efficiency in elevator systems. This would situate the study within the broader context of the field and demonstrate how the proposed system advances or differs from current practices.

A 1 - Revised and reflected. 

Q 2- In the Proposed EVAI System section, while the system's high-level operation is described, there is a potential lack of detailed technical specifications. For instance, the manuscript could provide more in-depth information on the hardware and software requirements, system architecture, and the specific algorithms used for facial recognition and elevator control logic.

A 2 - Revised and reflected. 

Q 3 - In the Proposed EVAI System section, the manuscript could provide more information on how the proposed system integrates with existing elevator and building management systems. Details on compatibility, installation processes, and potential challenges or costs associated with system integration would be valuable.

A 3 - I agree. We will reflect this in future research regarding the actual implementation and commercialization of the proposed system.

Q 4 - In the Experiment and Results section, the experiments primarily demonstrate the performance of the proposed system in isolation. There is a lack of comparative data against traditional elevator systems or other non-contact alternatives, which makes it difficult to assess the relative improvements and benefits of the new system.

A 4 - Revised and reflected. 

Q 5 - In some formulas, such as formulas 1-4, mathematical variables should be italicized.

A 5 - Revised and reflected. 

Thanks again to the reviewers.

Reviewer 2 Report

Comments and Suggestions for Authors

In this paper, Sun-Kuk Noh studied the non-contact artificial intelligence elevator system due to the impact of COVID-19. Generally, this paper is well-written and can contribute significantly to the scientific community. Therefore, I recommend publishing the paper after several minor modifications. Below are my suggestions:

 

Line 33: The authors cited five papers [1-5] to support one sentence, including three papers from the same author. I suggest citing a maximum of three papers to support the statement.

Line 61: The authors may delete this sentence: "This study consists of 2. Related Work, 3. Proposed System, 4. Test and Analysis, and 5. Conclusion," since it doesn't provide meaningful information. Every readers know this point already, if the author doesn't write this sentence.

Line 104, Table 1: What is the meaning of "Error rate" showing "doesn't exist"? Does it mean the number is 0, or that the authors didn't provide the error rate number? Please clarify.

Line 105, Figure 1 & Line 127, Figure 3: The figures are squeezed. The authors should avoid squeezing the figures and may change the font size.

Line 169, Table 3: "2.8m" and "5m/s" have no space between the number and unit. In contrast, Line 77 shows "5 m/s to 20 m/s" with spaces between the number and unit. Please make it consistent.

Line 188, Figure 9: Please provide information for the vertical axis and the unit.

Line 222: If applicable, the authors may add the fund number to identify the specific funding from Chosun University.

Comments on the Quality of English Language

To my knowledge, English Language is OK.

Author Response

Dear
I would like to express my sincere gratitude to the reviewers who reviewed my submitted paper and provided advice to improve its quality. I have reflected the points as much as possible and added contents and figures are marked in red. 

In this paper, Sun-Kuk Noh studied the non-contact artificial intelligence elevator system due to the impact of COVID-19. Generally, this paper is well-written and can contribute significantly to the scientific community. Therefore, I recommend publishing the paper after several minor modifications. Below are my suggestions:

Q 1 - Line 33: The authors cited five papers [1-5] to support one sentence, including three papers from the same author. I suggest citing a maximum of three papers to support the statement.

A 1 - Revised and reflected

Q 2- Line 61: The authors may delete this sentence: "This study consists of 2. Related Work, 3. Proposed System, 4. Test and Analysis, and 5. Conclusion," since it doesn't provide meaningful information. Every readers know this point already, if the author doesn't write this sentence.

A 2 - Deleted

Q 3 - Line 104, Table 1: What is the meaning of "Error rate" showing "doesn't exist"? Does it mean the number is 0, or that the authors didn't provide the error rate number? Please clarify.

A 3 - Revised and reflected

Q 4 - Line 105, Figure 1 & Line 127, Figure 3: The figures are squeezed. The authors should avoid squeezing the figures and may change the font size.

A 4 - Revised and reflected

Q 5 - Line 169, Table 3: "2.8m" and "5m/s" have no space between the number and unit. In contrast, Line 77 shows "5 m/s to 20 m/s" with spaces between the number and unit. Please make it consistent.

A 5- Revised and reflected

Q 6 - Line 188, Figure 9: Please provide information for the vertical axis and the unit.

A 6- Revised and reflected

Q 7 - Line 222: If applicable, the authors may add the fund number to identify the specific funding from Chosun University.

A 7- The number is 2023.

Thanks again to the reviewers.

Reviewer 3 Report

Comments and Suggestions for Authors

The paper proposes a method based on image recognition to manage the movement of an elevator. The goal is to shorten the time of the travel and to have contactless management.

To my knowledge, the idea is new and interesting. The methods of face-recognition is not novel, but the application is of practical importance. The author provides the results of experiments.

The paper is well-structured and within the scope of the journal Electronics.

I think, the paper merits publication. I have the following remarks for improvements:

1) About the method - it is good to elaborate the following cases:

- What will happen in case of individuals using the elevator who are new for the system? 

- How the case of cyber security glitches (such as the one we observed recently) will be handled?

b) It is also good to elaborate the privacy principles in this case.

2) About the content:

a) The English needs to be improved. I recommend proofreading by a native English speaker.

 

 

Comments on the Quality of English Language

The English needs stylistic improvement. There are some word repetitions in the sentences.

Author Response

Dear
I would like to express my sincere gratitude to the reviewers who reviewed my submitted paper and provided advice to improve its quality. I have reflected the points as much as possible and added contents and figures are marked in red. 

The paper proposes a method based on image recognition to manage the movement of an elevator. The goal is to shorten the time of the travel and to have contactless management. 

To my knowledge, the idea is new and interesting. The methods of face-recognition is not novel, but the application is of practical importance. The author provides the results of experiments.

The paper is well-structured and within the scope of the journal Electronics.

I think, the paper merits publication. I have the following remarks for improvements:

Q 1 - About the method - it is good to elaborate the following cases:

• What will happen in case of individuals using the elevator who are new for the system?

• How the case of cyber security glitches (such as the one we observed recently) will be handled?

• It is also good to elaborate the privacy principles in this case.

A 1 - Since it is intended for residents, individuals using it for the first time can use it as usual.

         In this study, cyber security was considered a future research topic.

Q 2- About the content:

The English needs to be improved. I recommend proofreading by a native English speaker.

The English needs to be improved. I recommend proofreading by a native English speaker.

A 2 - Revised and reflected. I will proofread it again if necessary.

Thanks again to the reviewers.

Reviewer 4 Report

Comments and Suggestions for Authors

High-rise apartments are very popular in big cities. Thus, there are big markets for the lift and its related equipment and technologies.

 

In the paper, the authors presented their proposed smart elevator system using artificial intelligence that allows residents in high-rise apartments to conveniently shorten waiting and use times and use the elevator non-contact during rush hours. The results are encouraging.

 

In the result section, the authors show the data to support that their proposed system can work and achieved their claim. The authors just said that the experiment used OpenCV for face recognition, and to capture the video, create an image file, and use the image to check whether a human face exists through a face detection algorithm. OpenCV's Haar feature-based cascade classifiers was deployed to identify residents' faces in real time from images from the Pi camera. The sample size, i.e., number of residents, is 240 persons. The results are shown in the Figure 9, consisting of just three simple columns.  For conference publication, this very brief result is good enough. Nevertheless, for journal publication, much more comprehensive evaluations are needed to show the advantages of the proposed AI methods over other methods. For example, there is no comparison with other methods in this version of the paper. The sample size is also very small. It may not cover many real-world situation for its real-world application. More details about the deployed AI methods are needed as well to show the advantage of using this existing AI technology. Otherwise, the content of this paper is more like a simple implementation of the existing AI technology to some human face images taken from the lift. The authors need to show their choice of this existing AI technology is superior to other AI technologies for this specific task.

Comments on the Quality of English Language

The English is easy to read.

Author Response

Dear
I would like to express my sincere gratitude to the reviewers who reviewed my submitted paper and provided advice to improve its quality. I have reflected the points as much as possible and added contents and figures are marked in red. 

High-rise apartments are very popular in big cities. Thus, there are big markets for the lift and its related equipment and technologies.

In the paper, the authors presented their proposed smart elevator system using artificial intelligence that allows residents in high-rise apartments to conveniently shorten waiting and use times and use the elevator non-contact during rush hours. The results are encouraging.

In the paper, the authors presented their proposed smart elevator system using artificial intelligence that allows residents in high-rise apartments to conveniently shorten waiting and use times and use the elevator non-contact during rush hours. The results are encouraging.

For conference publication, this very brief result is good enough. Nevertheless, for journal publication, much more comprehensive evaluations are needed to show the advantages of the proposed AI methods over other methods. For example, there is no comparison with other methods in this version of the paper. The sample size is also very small. It may not cover many real-world situation for its real-world application. More details about the deployed AI methods are needed as well to show the advantage of using this existing AI technology. Otherwise, the content of this paper is more like a simple implementation of the existing AI technology to some human face images taken from the lift. The authors need to show their choice of this existing AI technology is superior to other AI technologies for this specific task.

A 1 -I agree with your good opinion. I have revised and supplemented it, and will reflect it in future follow-up research related to the actual implementation and commercialization of the proposed system.

Thanks again to the reviewers.

Reviewer 5 Report

Comments and Suggestions for Authors

There's a significant overlap between the abstract and introduction: the former should be made more concise and focused, while the latter could be more detailed and technical. The conclusion, on the other hand, is almost a copy of the introduction. The literature is referenced somewhat carelessly - 11 articles at one reference on page 2; Fig. 1 and Fig. 9 are of poor quality.

Yet, the biggest issue is the lack of sufficient details about the experiment: how many iterations it takes the algorithm to be trained, why was this particular method chosen, what are the confidence intervals for the average times provided in the manuscript? These are just some of the questions that need to be answered before the text begins to look like a scientific paper. 

Comments on the Quality of English Language

There are plenty of repetitions (e.g. lines 15, 48, 50) and misplaced commas. Other than that, the English is okay.

Author Response

Dear
I would like to express my sincere gratitude to the reviewers who reviewed my submitted paper and provided advice to improve its quality. I have reflected the points as much as possible and added contents and figures are marked in red. 

Q 1 - There's a significant overlap between the abstract and introduction: the former should be made more concise and focused, while the latter could be more detailed and technical. The conclusion, on the other hand, is almost a copy of the introduction. The literature is referenced somewhat carelessly - 11 articles at one reference on page 2; Fig. 1 and Fig. 9 are of poor quality.

A 1 - Revised and reflected. 

Q 2- Yet, the biggest issue is the lack of sufficient details about the experiment: how many iterations it takes the algorithm to be trained, why was this particular method chosen, what are the confidence intervals for the average times provided in the manuscript? These are just some of the questions that need to be answered before the text begins to look like a scientific paper.

A 2 - I agree with your good opinion. Revised and reflected. 

Thanks again to the reviewers.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Referee report

 

This paper introduces a smart elevator system (EVAI) that utilizes artificial intelligence for non-contact operation during the COVID-19 pandemic. The system is designed to reduce physical contact with elevator buttons, thereby minimizing the risk of virus transmission. The EVAI system incorporates AI facial recognition technology to identify residents and automatically select their desired floor based on the time of day and the resident's previous usage history. The experimental results of the proposed system showed a more accurate facial recognition function and a reduction in waiting time and travel time. The authors carefully revised the manuscript and made some changes to the version according to the comments of the reviews. Therefore, I think that the paper makes a contribution and has the potential to be published.

Comments for author File: Comments.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

The revised version is much better than the original version.

Reviewer 5 Report

Comments and Suggestions for Authors

All major issues in the old version seem to have been addressed properly. I see a significant improvement in the manuscript and therefore recommend that it is published either in its present form or with minor changes the authors might decide to make in the last moment.