Research on Digital Meter Reading Method of Inspection Robot Based on Deep Learning
Round 1
Reviewer 1 Report
Comments for author File: 
Comments.docx
适当改进
Author Response
We would like to thank the reviewers for their valuable comments on our manuscript. We have revised the manuscript according to your comments. For details, please refer to the revised manuscript and attachments.
Author Response File: Author Response.docx
Reviewer 2 Report
It is an interesting research. However, the following issues need to be considered.
1. In the article, it is mentioned that using various deep learning methods such as Polygon YOLOV5, YOLOV5s, CRNN. which are your main tasks, the contributions of this study should also be given more clearly.
2. The state of art should be strengthened. Some latest object detection approaches for related to the research of this paper, such as: “Apple grading method design and implementation for automatic grader based on improved YOLOv5” and “Real-time detection of underwater river crab based on multi-scale pyramid fusion image enhancement and MobileCenterNet model”, may be referred for the author.
3. Figure.4 is unclear and duplicated with Figure.5, suggest deletion it.
4. In line 220，what does (x'/z', y'/z') mean and is there an error? Please confirm it.
5. In line 267，“The simulated generated digital composite images are shown in Figure.11”. “Figure 11” should be “Figure.12”, right?
6. The English writing of the article should also be improved.
The English writing of the article should also be improved.
Author Response
We would like to thank the reviewers for their valuable comments on our manuscript. We have revised the manuscript according to your comments. For details, please refer to the revised manuscript and attachments.
Author Response File: Author Response.docx
Reviewer 3 Report
The manuscript entitled "Research On Digital Meter Reading Method of Inspection Robot Based on 2 Deep Learning" can be published after a proper revision.
1. The figures have low qualities. For example, Fig. 4, has much lower quality than Fig. 3.
2. The introduction part should be enhanced. The physics-informed neural network is one of the most popular machine learning schemes to significantly reduce the calculation resource, training data set, and learning time by informing the existing physical knowledge to the algorism. I would like to see more discussion in terms of this form, see. Journal of Computational Physics 378 (2019): 686-707; Laser & Photonics Reviews 16.10 (2022): 2100658; ACS Applied Materials & Interfaces 14.23 (2022): 27397-27404.
3. If the authors can provide a comparison between this method and other groups' methods using a Table, it would be better for a broader audience to understand the significance of this work.
4. English should be improved.
Author Response
We would like to thank the reviewers for their valuable comments on our manuscript. We have revised the manuscript according to your comments. For details, please refer to the revised manuscript and attachments.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
I think the revision has been perfected
Reviewer 2 Report
All my concerns have been resolved in the revision. There are no more comments there.
Minor editing of English language required
Reviewer 3 Report
I would like to endorse the publication of this manuscript.
