Application of Image Processing Techniques for UAV Detection Using Deep Learning and Distance-Wise Analysis

Round 1

Reviewer 1 Report

This paper seems to be a comparative analysis work including incomplete data and author contributions. The paper needs major revision by reconsidering all sections.

Author Response

Dear Reviewer, Please see the attachment.

Thank you

Author Response File: Author Response.docx

Reviewer 2 Report

This paper uses images containing drones and uses the Yolo model to successfully identify drones and obtain good results. This shows the progress of science and the power of software applications. However, there are still some deficiencies in this paper that need to be corrected, such as the following suggestions.

1. There is only one UAV in the identified image, it is recommended to identify multiple UAVs/types at the same time.

2. It is recommended to have images of drones and animals (such as birds) for identification. This approach can strengthen the identification ability of this study. This is possible in the YoloV5 model.

3. The advantages and disadvantages of the Yolov5 and Yolov7 models should be listed in order to increase readability.

4. Does not account for nighttime or complex images, suggesting that study limitations should be stated.

5. Forest terrain can use image enhancement (red, blue) to improve recognition, which should not be a research limitation.

 

6. It is recommended to add relevant references to improve relevant research skills.

Author Response

Dear Reviewer, Please see the attachment.

Thank you

Author Response File: Author Response.docx

Reviewer 3 Report

Line 19: “a number of”?

Line 37: why is there “(UAV)” here?

Line 74: “Aker et al.”?

Section 2: It would be better for the authors to highlight the issues with existing methods and give the motivation of this research.

Line 106-107: I think it would be better to say that “identifying things of interest in digital photos or videos”

Line 107: No need of “The objects found are people, vehicles, chairs, stones, structures, and animals.”

Section 3.2: It would be better to briefly introduce the evolution of YOLO. What is the difference between each version of YOLO?

Line 173: How is the hue augmentation applied? It would be better to explain the “-50 degree” and the “+50 degree” in Figure 2. More details are needed.

Table 1: The bold highlight is not obvious.

Table 1: Could you give some explanation about why the experiments on Hue have the highest accuracy for YOLOV5 but the experiments on original RGB have the highest accuracy for YOLOV7?

Line 280-323: is there specific rule for the authors to determine the ranges? For the far-range analysis, it seems to me that the drone is still very close when it is 8 meters away.

 Line 280-323: It would be better to organize the distance-wise analysis as a table.

Author Response

Dear Reviewer, Please see the attachment.

Thank you

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

I have reviewed the paper again and satisfactory work done by the authors. All the sections has been revised greatly. Now the paper look ok for the publication after minor grammatical corrections. It is recommended that the paper is well suitable for publication in mdpi.

Author Response

Respected Reviewer,

We appreciate your insightful advice.

Reviewer 2 Report

The authors address most of the doubts, but some practices still need to be clarified.

1. It is recommended to have images of drones and animals or aircraft (such as birds, airplanes, and helicopters) for identification. This approach can strengthen the identification ability of this study. This is possible in the YoloV5 model. It's just drone form recognition, which doesn't make much sense.

2. You need to expand your literature review. A typical academic paper should have around 25-30 academic references. To this end, your manuscript MUST include at least 5 additional relevant references published in ACADEMIC JOURNALS in the past three years (2020, 2021, and 2022).

Author Response

Respected Reviewer,

We appreciate your insightful advice. Please see the attachment

Author Response File: Author Response.docx

Reviewer 3 Report

Thank you for addressing all my comments. The manuscript can be accepted in present form.

Author Response

Respected Reviewer,

We appreciate your insightful advice.