Recognition Method of Knob Gear in Substation Based on YOLOv4 and Darknet53-DUC-DSNT

Round 1

Reviewer 1 Report

The paper proposes a convolutional neural network YOLOv4 (You Only Look Once) as the main structure of the knob area detector.

 

The authors did not indicate what the contributions of this paper clearly. The novelty of the article is not compared to the state-of-the-art sufficiently in the introduction section.

why do you use YOLOv4 if there is YOLOv5/s?

Figure 2 seems to be a very standard flow for YOLOv4. I do not see any novelty.

And there is a lack of novelty subsistence for a journal.

Comparison with other networks are not available. It is hard to see the merits of the proposed method.

Figure 4 needs to explain in details. it is hard to follow the flows.  Hard to see the contributions/novelty of the block diagrams.

 

Author Response

Please see the attachment. Thanks!

Author Response File: Author Response.pdf

Reviewer 2 Report

This study developed a method for recognition of substation knob gear position based on YOLOv4 and Darknet53-DUC-DSNT models.

The manuscript is well prepare and easy to follow.

My comments are as follows.

1. The motivation of this study is unclear. The author needs to show a situation that needs to use the method in the article to solve.

2. Figure 1 should be explained in more detail. 

The rest is well prepared, I have no further comments.

Author Response

Please see the attachment. Thanks!

Author Response File: Author Response.pdf

Reviewer 3 Report

The authors propose a methodology for identifying substation knob gear position based on the combination of a convolutional neural network and darknet53.

Next, there are some comments:

a)      The introduction is very brief, and it is necessary that the authors do a better investigation of state of the art and that it be included in this section because only four articles are considered. The authors said there are few researches on the recognition of knob gear, but some applications of used methods could be included. The section on related work could be included in the introduction.

b)     The proposed method has some lacks compared to other proposed methodologies as the authors consider, but the best novelty is that the knob gear recognition model improves the detection performance, and the best characteristic is that job is achieved in real-time.

Author Response

Please see the attachment. Thanks!

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

I have no further comments

Author Response

Please see the attachment. Thanks!

Author Response File: Author Response.docx

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.