A Lightweight Residual Model for Corrosion Segmentation with Local Contextual Information

Round 1

Reviewer 1 Report

The work is novel and very interesting. The authors demonstrate a new method of corrosion detection and the method is shown superior to the existing methods. 

a) I would suggest a few grammatical error corrections for the paper.

b) What is the limitation of the existing methods?

c) How can the method be practically applied today for sensitive devices such as aircraft? 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

A Lightweight Residual Model for Corrosion Segmentation with Local Contextual Information

Dear Editor,

The manuscript by Huang et al reports the lightweight residual model for corrosion segmentation. I appreciate the opportunity to review this interesting manuscript (minor revision). The authors should carefully revise and check the manuscript for any typos or mistakes.  More specific comments are found below:

1-    Please correct grammar mistakes and spelling errors.

For example:

To address these issues, a lightweight residual deep-learning model based on encoder-decoder structure is proposed in this paper. We design small and large kernels to extract local detailed information and capture distant dependencies respectively at all stages in encoder.

Please check the reference writing and style. 

In recent years, image processing technology based on deep learning methods has improved the detection accuracy obviously, it is widely applied for industrial inspection [3].

2- The first time you use an abbreviation, it's important to spell out the full term and put the abbreviation in parentheses.

global GDP, the traditional FCN[8] framework, etc.

3- In the introduction section, the novelty of the study and its superiority over other studies should be explained.

4- Conclusion re-written emphasizing on results and future directions.

Best regards.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

The paper is a good attempt; however, the following is suggested to improve the quality of the paper:

1.     The organization of the paper is bit confusing and can be improved in a sequential manner.

1.     Provide more numerical values as properties in the Abstract. It looks general. Also, the abstract is incomplete and does not tell the story of the whole article in a concise way. Revise the abstract to provide (i) significance of the study, (ii) the aim of the study, (iii) the research methodology, (iv) major conclusion of the study.

2.     The literature review is very descriptive and doesn't clearly highlight the current gaps in the field, which underpins the current investigation. More critical and comprehensive literature is needed.

3.     As there is lot of similar work available, the novelty of the work is not explained clearly.

4.     Please indicate that the significant findings in references [15], [16], and [18] will be beneficial for readers.

5.     The results should be analyzed and discussed in greater depth in this paper.

6.     Conclusions - Important new results and knowledge along with their potential use should be listed. Being as quantitative as possible. Do not just summarize what work was conducted in the manuscript.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

The authors have revised this paper according to the  reviewers' comments.  It can be accepted.