Next Article in Journal
Application of LiDAR Sensors for Crop and Working Environment Recognition in Agriculture: A Review
Previous Article in Journal
Perceptual Quality Assessment for Pansharpened Images Based on Deep Feature Similarity Measure
 
 
Communication
Peer-Review Record

Multi-Sensor Image Classification Using the Random Forest Algorithm in Google Earth Engine with KOMPSAT-3/5 and CAS500-1 Images

Remote Sens. 2024, 16(24), 4622; https://doi.org/10.3390/rs16244622
by Jeonghee Lee 1, Kwangseob Kim 2 and Kiwon Lee 1,*
Reviewer 1: Anonymous
Reviewer 2:
Remote Sens. 2024, 16(24), 4622; https://doi.org/10.3390/rs16244622
Submission received: 5 November 2024 / Revised: 6 December 2024 / Accepted: 7 December 2024 / Published: 10 December 2024

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you to the authors for presenting a well-written manuscript. Below are a few minor point for your consideration: 

- While the authors highlight the use of external satellite images in Google Earth Engine (GEE) and the combination of different image sources as the novelty of the work, both approaches have already been explored by other researchers. For example, in a paper authored by the same team, as cited in Ref. 17.

- In line 126, please replace "by [21]-[23]" with the authors' names to provide clarity and proper attribution.  

- The study combines sensors with significantly different spatial and spectral resolutions. Although resampling is used to address alignment issues, it may introduce noise or artifacts that could affect classification accuracy. However, the paper does not address how these discrepancies might influence the results. (Please elaborate the methodology and potential challenges encountered when introducing the external satellite images to GEE and performing image resampling) 

- I recommend discussing the limitations of the method in the conclusion section. For example, the study utilizes only six land cover classes, including an "unclassified" category, which may oversimplify the complexity of land cover in the study area. Subdividing urban or agricultural areas further could have provided a more detailed analysis. The conclusion should address how effective this method would be in more complex scenarios.

- Furthermore, the scalability of the approach could be limited when applied to larger study areas or regions where similar datasets are unavailable. These aspects should be considered and addressed to provide a more comprehensive evaluation of the method's broader applicability.  

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

See the attached review. 

Comments for author File: Comments.pdf

Comments on the Quality of English Language

See review attached for authors. 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Back to TopTop