A Review of Practical AI for Remote Sensing in Earth Sciences

Round 1

Reviewer 1 Report

This review paper evaluates the use of AI techniques in remote sensing, analyzing methodologies, outcomes, and limitations. It explores various applications such as image classification, land cover mapping, and object detection, discussing challenges and potential solutions. The paper provides valuable insights for researchers, practitioners, and decision-makers in this field. Herein, some comments improve the quality of the article:

*Section 1 (Introduction): The review paper lacks a clear introduction that outlines the motivation behind the study. Additionally, the last paragraph should present the contributions in bullet points, and the paper's organization needs to be explicitly stated. Addressing these areas will enhance the overall clarity and structure of the paper, making it more reader-friendly for researchers, practitioners, and decision-makers in the field.

*Section 2 of the review paper should present the research methodology employed to conduct this review. It is important to include details about the methodology used to collect, evaluate, and synthesize the existing literature on AI applications in remote sensing. This will provide transparency and credibility to the review process, enabling readers to understand the rigor and validity of the findings presented in the paper.

*Section 2.1 (Brief recap of remote sensing technologies ), should be including a comparison table at the end of the subsection. This table should highlight the advantages, limitations, and applications of each technology for easy reference and comparison.

*In Section 2.2, the discussion on AI and ML techniques is inadequate and needs improvement. It is recommended to include a summary comparison table at the end of this subsection, highlighting the advantages, limitations, and applications of each technique. This will provide readers with a quick and accessible overview for easy comparison.

fine

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

This paper conducts a comprehensive review of AI applications on remote sensing. This survey is interesting and has some potential. Some minor modifications should be made before publication:

1) As AI technology relies on large size of data, how to obtain remote sensing data is pretty important. Based on the platform of UAV, UGV, aircraft, and satellite, several sensors would be used such as GPS, IMU, LiDAR, and camera would be utilized. Thus, some related work should be included in the introduction from the perspective of data acquisition: autonomous vehicle kinematics and dynamics synthesis for sideslip angle estimation based on consensus kalman filter, vehicle sideslip angle estimation by fusing inertial measurement unit and global navigation satellite system with heading alignment, automated vehicle sideslip angle estimation considering signal measurement characteristic, imu-based automated vehicle body sideslip angle and attitude estimation aided by gnss using parallel adaptive kalman filters, vision-aided intelligent vehicle sideslip angle estimation based on a dynamic model. In this way, it would be better for readers to follow the knowledge of data acquisition.

2) For the LiDAR-based semantic segmentation, the Unet network is a popular network. It would be meaningful to include the work: an automated driving systems data acquisition and analytics platform. It is from another perspective of surrounding environment perception.

3) Furthermore, the precision agriculture is also a meaningful research direction: YOLOv5-Tassel: Detecting tassels in RGB UAV imagery with improved YOLOv5 based on transfer learning. Discuss this work would help to illustrate the broad application of remote sensing.

4) In Figure 12, please make sure the font size is consistent.

Overall, this survey is meaningful and can promote the rapid development of this field

 

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

The manuscript provides a comprehensive overview of practical AI for remote sensing, but limited to Earth sciences.

The presented material is too popular and scientific and, in its current form, is not useful for students or science adepts, perhaps as a presentation for seniors.

Comments:

1. According to the content, the manuscript should be titled "A Review of Practical AI for Remote Sensing in Earth Sciences".

2. Since the manuscript in its current form is neither an original research paper nor does it contain unique scientific insights, the thesis of the work should be formulated in the Introduction - what the authors want to achieve, in what research scope and for whom.

3. Too many, as many as 124 cited references, are not supposed to show how many authors can find different sources of knowledge, but instead of quantity, decide on the quality and relevance to the topic of the paper.

4. About 25% of the references are not related to AI or have no substantive content or are old.

5. I suggest numbering references and limiting them to half.

6. Terrible low-quality graphics of the manuscript.

Drawings should be made in the same style and size as the paper text.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

I am satisfied with the latest revision

Readable

Reviewer 3 Report

Since the authors have revised the manuscript in accordance with the reviewer's comments, I propose its publication in Remote Sensing Journal.