All reviews of published articles are made public. This includes manuscript files, peer review comments, author rebuttals and revised materials. Note: This was optional for articles submitted before 13 February 2023.
Peer reviewers are encouraged (but not required) to provide their names to the authors when submitting their peer review. If they agree to provide their name, then their personal profile page will reflect a public acknowledgment that they performed a review (even if the article is rejected). If the article is accepted, then reviewers who provided their name will be associated with the article itself.
Based on the recommendation of the experts, I am happy to let you know that your paper has been recommended for publication. Thank you for your fine contribution to our esteemed journal.
[# PeerJ Staff Note - this decision was reviewed and approved by Jyotismita Chaki, a 'PeerJ Computer Science' Section Editor covering this Section #]
accept as it is.
No amendments required.
Agreed
No
No comments
None
No comments
I accept this work. Author Did a great job and carefully address all comments. I have no further comments.
Best of luck
Dear authors,
Your paper requires a couple of comments to be incorporated, therefore, please revise and re-submit the updated version.
[# PeerJ Staff Note: The review process has identified that the English language must be improved. PeerJ can provide language editing services - please contact us at copyediting@peerj.com for pricing (be sure to provide your manuscript number and title) #]
[# PeerJ Staff Note: Please ensure that all review comments are addressed in a rebuttal letter and any edits or clarifications mentioned in the letter are also inserted into the revised manuscript where appropriate. It is a common mistake to address reviewer questions in the rebuttal letter but not in the revised manuscript. If a reviewer raised a question then your readers will probably have the same question so you should ensure that the manuscript can stand alone without the rebuttal letter. Directions on how to prepare a rebuttal letter can be found at: https://peerj.com/benefits/academic-rebuttal-letters/ #]
This paper proposes a segmentation network based on G-lite-Deeplabv3+, and deploits it to Cyber Physical System to improve the accuracy and efficiency of product packaging image segmentation. Firstly, Mobilenetv2 is used to replace Deeplabv3+ feature extraction network, and packet convolution and attention mechanisms are introduced to process high-level semantic features, so as to improve the sensitivity of the network to useful features. Secondly, the image acquisition module, image processing module and output execution module are set respectively in CPS, and G-lite-Deeplabv3+ network is integrated into the image processing module in CPS to realize remote and real-time product packaging image segmentation in Cyber space. The following suggestions can help the author improve the manuscript.
Before submitting a revision be sure that your material is properly prepared and formatted;
If you are unsure, please consult the formatting instructions to authors that are given under the "Instructions and Forms" button in the upper right-hand corner of the screen;
In the second paragraph of the introduction, the author lists a large number of existing studies, but the logical relationship between them is not clear. Some transitional statements can help readers better understand these works;
Uncommon abbreviations (such as FPS, mPA) should be spelled out at first use. Do not include footnotes or references;
Some references are too old and need to be replaced. The author had better use the latest literature in the last three years for citation;
What is the optimization method of Lite-bottleneck?
Add more description to Figure 2;
The language expression of the conclusion part needs to be optimized, and the content of this part needs to supplement the limitations specifically;
The author should give specific results relevant to the purpose; Avoid outcomes that are irrelevant to the purpose.
Please see report above
Please see report above
no comment
no comments
no comments
Aiming at the problem that traditional segmentation methods can not meet the real time and accuracy of product packaging image segmentation in practical application, a lightweight method G-lite-DeepLabv3+ is proposed. Experimental results show that the G-lite-DeepLabv3 + network proposed in this paper is superior to the control network in terms of accuracy and efficiency.
This study can provide some basis for improving the application of segmentation network in product packaging image segmentation task. This paper has some innovation, but it needs to be modified.
1. Try to set the problem discussed in this paper in more clear, write one section to define the problem;
2. I have not seen more details about how to determine the parameters of model network training (especially the selection of super parameters), but it is very important;
3. Add more analysis to Table 5, focusing on the comparison of data between different models and the underlying principles;
4. Legend description should be spontaneous, therefore, some description is not necessary, for example “The purple curve represents the method of this paper, under the same number of training rounds”;
5. The language of the conclusion needs to be further strengthened;
6. The description of the conclusion coincides with the abstract, and the author should focus on the outstanding contribution of the research and the application direction;
7. Most sentences contain grammatical and/or spelling mistakes or are not complete sentences;
8. The authors must have their work reviewed by a proper translation/reviewing service before submission; only then can a proper review be performed.
All text and materials provided via this peer-review history page are made available under a Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.