Accurate Extraction of Ground Objects from Remote Sensing Image Based on Mark Clustering Point Process

Round 1

Reviewer 1 Report

In this paper authors present a new method for extraction of ground objects from Remote Sensing Image Based on Mark Clustering Point Process. The paper is well-structured, however I have some major concerns regarding its content, such as:

- Introduction sections text needs to be structured into meaningful paragraphs.

- Add point-by-point novelties in regards state-of-the-art.

- Replace Fig 6 with a table, as the figure is not really meaningful.

- Why 4000 iterations, and not more or less?

- Results should include runtimes of the given algorithm, as well as comparison to other state-of-the-art satellite imagery clustering algorithms in terms of accuracy.

- How does the algorithm compare to deep learning methods such as R-CNN ?

- Conclusion should be extended to include drawbacks and future work.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report

The authors proposed the mark clustering point process to extract the geometric features of ground12 object accurately. However, the organization of the manuscript has important errors. The main issues are presented below:

1. The abstract is confused, it needs to be more direct to inform the reader about some background, purpose of the study, methods implements, main results, and brief conclusion;
2. The introduction must have more than one paragraph, presenting background, related work, purpose of the study, scope of the study, and the structure of the manuscript. Now, it is also very confuse;
3. The proposed method is well-presented and very detailed, but a linguistic revision is needed;
4. The results and conclusions are very detailed and compared with other algorithms, but a linguistic revision is needed;
5. Some references are incomplete, and the format of the references is not correct.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 3 Report

This paper proposed the mark clustering point process to extract the geometric features of ground object. The experimental results have verified the effectiveness of proposed method. Some specific comments are as follows.

(1) The title stated that the proposed method is aimed at remote sensing images. However, I think this method can’t be applied to SAR images. It is necessary to explain the scope of application of the proposed method.

(2) In my opinion, the scenarios in synthetic images and real images (island and lake experimental images) are simple. They may not be enough to verify the effectiveness of this method.

(3) In the experiment, the comparison methods need to be described clearly.

(4) The writing and typesetting of this article need to be carefully checked.

(5) What’s the performance of deep-learning-based methods on getting ground objects?

Author Response

Please see the attachment

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

This revised version can be accepted.