Next Article in Journal
Multi-Scale Object Detection in Remote Sensing Images Based on Feature Interaction and Gaussian Distribution
Previous Article in Journal
Near-Surface Dispersion and Current Observations Using Dye, Drifters, and HF Radar in Coastal Waters
 
 
Article
Peer-Review Record

Geocomplexity Statistical Indicator to Enhance Multiclass Semantic Segmentation of Remotely Sensed Data with Less Sampling Bias

Remote Sens. 2024, 16(11), 1987; https://doi.org/10.3390/rs16111987
by Wei He 1,2,†, Lianfa Li 1,2,*,† and Xilin Gao 1,2
Reviewer 1:
Reviewer 2: Anonymous
Reviewer 3: Anonymous
Remote Sens. 2024, 16(11), 1987; https://doi.org/10.3390/rs16111987
Submission received: 16 April 2024 / Revised: 22 May 2024 / Accepted: 23 May 2024 / Published: 31 May 2024
(This article belongs to the Section AI Remote Sensing)

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Very well designed and organized article, great job!!!

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The structure and logic of this paper are clear and reasonable. It discusses based on two indicators and achieves commendable results. However, one question remains: Why did the authors choose these two specific indicators over others? It would be beneficial if the authors could provide a detailed explanation for their selection of these indicators within the text.

 

Additionally, there appear to be some inaccuracies in the citation format within the references. It would be appreciated if the authors could carefully verify and correct these.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript entitled “Geocomplexity statistical indicator to enhance multiclass semantic segmentation of remotely sensed data with less sampling bias” focuses on the general description of analyses relying on indicator geocomplexity statistical. Unfortunately, based on information given by the Authors, one can only state that the analyzes carried out were using deep learning models, such as UNet, SegNet, Global CNN, DeepLab V3, FCN-ResNet, UperNet, and SegFormer. The goal of the study was to explore the effectiveness of utilizing complexity-related indicators for optimal sampling, with the goal of reducing sampling bias in remote sensing. And t wo indicators were used to achieve this goal:information entropy and Moran ‘I used to quantify multiclass complexity in remote sensing images.The visualization of the complexity scores showed that the entropy-based indicator was sensitive to the boundaries of different classes and the contours of geographical objects, and the gray-based indicator (Moran ‘I) better extracted spatial structure information of geographical objects in remote sensing images.

In the absence of detailed analyses, it is difficult to determine the degree of novelty of this manuscript. From a scientific point of view, the manuscript is of low value, it has only cognitive significance as in the current form this manuscript just presents an information about the validity of introducing the proposed indicators for remote sensing research. The attached drawings 8 to 11 are difficult to read and do not provide the expected information.

I have two editorial comments: I would place the text from lines 110 to 153 in the "Methods" section, and the text from 566 to 587 in the "Conclusion" section. In the current version, the conclusions are a summary.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Back to TopTop