Meta Classification Model of Surface Appearance for Small Dataset Using Parallel Processing
Round 1
Reviewer 1 Report
In this paper, the authors propose an approach for innovative accurate and efficient fabric protrusion detection. The proposed approach is based on machine learning and aims to improve model training with a small dataset.
The topic considered by the authors is extremely interesting and the solution they propose appears technically sound although not very innovative. The authors propose lots of experiments to demonstrate the soundness of their approach.
My main concern about this paper concerns the innovativeness of the proposed solution. The authors should enrich the related literature. Also, after the experiments, they should add a session "Discussion" in which they indicate what are the main novelties of their approach compared to past ones already proposed in the literature.
Author Response
We really appreciate your kind and detailed answer.
Please find the attached PDF for our response,
Regards,
Authors.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
1. Please well check the format.
2. In section 5.2, does the author really konw dense network? What is the "kernel(60*60)" mean, is there so big kernel size?
......
Please well reconsider the work.
Author Response
Thank you a lot for your time and your kind response.
You are correct that the representation of the Neural Network can cause a lot of misunderstanding.
The meaning of (axb) is the input and the output of each hidden layer. Therefore the meaning of 60x60 is that this hidden layer receive an input size of 60 and the output of this hidden layer to the next one is also 60.
I will change this format, and once again thank you for you consideration.
Reviewer 3 Report
Comments for author File: Comments.pdf
Author Response
Dear reviewer,
Thank you very much for your kind and detailed feedback, we really appreciate your time and contribution.
Please find the attached PDF answer.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The authors have performed a (limited) effort to comply with my suggestions. They have slightly improved the part about the innovativeness of their approach; instead they did not put any "Discussion" section.
As a consequence, the paper is slightly improved but need major revisions again.
Author Response
Dear Reviewer,
Thank you very much for you kind and detailed answer, we really apricate your time and help.
We took to our attention all your suggestions and you’ll be able to see that on the fixed paper. The below text is where we would like to clarify and to know your thoughts about.
We've added a "Discussion" paragraph according to your suggestion.
Author Response File: Author Response.pdf
Reviewer 2 Report
The authors have addressed all the queriesAuthor Response
Dear reviewer,
Thank you very much for you kind and detailed answer, we really apricate your time and help.
Reviewer 3 Report
Authors have to provide the figures with high resolution
Author Response
Dear reviewer,
Thank you very much for you kind and detailed answer, we really apricate your time and help.
Round 3
Reviewer 1 Report
The authors have added the section "Discussion" I had required, albeit it is very short. Therefore, in my opinion, the paper can ben published.
