Extraction of Canal Distribution Information Based on UAV Remote Sensing System and Object-Oriented Method

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The authors analyzed a way of identifying canal systems using high-resolution images. The justification for the study is clear and contributes to the area of ​​interest. The introduction and methodology are well explained and structured. However, I have some suggestions for improving the article:

- If possible, improve the resolution of Figure 3 as it makes it easier to understand and read;

- Section 5.2 on metrics would be better in the methodology. The equations in the methodology and only the results later;

- The results and discussion could be improved. Clearly state the limitations of the study and tests. Show what the next steps would be to continue the research and obtain better results...

 

- In the conclusion, review the terms “will be”.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Abstract: In general, the abstract needs to be rewritten. First, the objective of the article should be more direct. For example, "The objective of this study is to evaluate, analyze...". Second, the methodology should be transparent, such as "The methodology applied to the study was the use of the method...". Regarding the results, the authors have already presented the main findings, which is appropriate for me. Also, I would suggest that the authors include an introductory sentence and another to conclude the abstract.

1. Introduction:

·         In line 36, the authors should rewrite it to "orbital, sub-orbital, and terrestrial."

·         In line 53, the term should be corrected. The appropriate phrasing would be "small canals," "minor canals," or something similar, depending on the type of canal being described. "Hairy" is not suitable for describing irrigation canals.

·         In line 64, the sentence should start with a capital letter.

·         In line 92, the authors should clearly state the article's objective, as is done in the abstract.

·         Lines 93 to 95 can be omitted, as discussing the methodology at this point is unnecessary. It will be detailed in the dedicated section.

 

2. Materials and Methods

2.1 Overview of the Study Area and Basic Information on the Drainage System

·         The subsection title should be "Study Area" only.

·         In the study area description, it is essential to mention the name of the location, city, and state.

·         Marginal information should be included on the map, such as the coordinate system (e.g., Geographic), datum (WGS 84?), and the source used. Additionally, use "km" instead of "kilometre" and place it next to the scale. A border should also be added around the location map.

 

3.3 Mask Processing:

·         The first sentence in lines 152 and 153 needs to be rewritten as it is unclear. The source and meaning of the value [0.4920] should be explained. These values need a clear explanation.

·         In line 168, the term "salt and pepper phenomenon" must be explained, as it is disconnected, and the reader will not understand it without proper context.

4.3.1 Classification Based on Spectral Mean

 

The authors should not refer to Figure 7 as a "map." Please refer to it only as a "figure."

 5.1 Drainage System Extraction Results

·         In the results description, it is necessary to cite figures 9A, 9B, 9C, 9D, 9E, and 9F. The absence of these references makes it difficult to understand the results. This section should be reviewed.

 5.2 Precision Evaluation and Analysis

·         Section 5.2, from lines 285 to 304, refers to methodology and should be relocated to the corresponding section.

·         The results of this section are in lines 305 to 314. Additionally, the position of Table 3 should be reversed: start with the paragraph and then present the table.

 6. Discussion:

·         The authors should explain the phenomenon presented in line 327 in the methodology section, as it is still unclear in the discussion.

·         When the authors state that OCNN had the best performance in Zhao's study, it is essential to explain why this model was not used, given its better performance.

·         In line 351, the authors discuss a model applied to the extraction of large landslide edges, which does not seem aligned with the article’s theme. I recommend reviewing this part of the discussion.

 6.2 Effect of Canal Type on Extraction Accuracy

When I started reading this section, I expected to discuss the article's results, but most of the section is a literature review. Only a small portion deals with the article’s findings. I suggest the authors connect the discussion with the results on the canal top width presented in this study. Begin with the second paragraph, then link it with the literature.

 7. Conclusions

·         The conclusions begin by stating that the authors "will process," indicating future actions. However, the conclusions should reflect what has already been done, i.e., in the past tense.

·         The first paragraph should be removed, as it repeats unnecessary information.

·         The conclusion needs to be rewritten. Do not repeat the results; instead, highlight the article’s main conclusions, answering the objective proposed at the beginning.

·         I suggest adding a paragraph with recommendations for future research.

 

·         The section's title should be changed to "Conclusions and Recommendations."

Comments on the Quality of English Language

The article needs a major English revision.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The paper focuses on the extraction of canal system information in irrigation districts and conducts research on unmanned aerial vehicle remote sensing image information processing methods, which has good practical value.

The paper uses multispectral drone for image data acquisition. Using rule-based object-oriented methods to process high-resolution multispectral images. The segmentation parameters of remote sensing images have been determined. A rule for extracting channel distribution information was established by utilizing the differences in spectral and spatial features between channels and other ground objects. The paper conducted statistical analysis on the identification results of the canal system.

There are still some shortcomings in the paper:

1. The paper lacks sufficient explanation of the innovative methods and main contributions. The innovation of the methods used in the paper lacks clear and concise explanations and summaries; The main contribution of the paper lacks clear explanation.

2. The method introduction in the paper is not clear and specific enough. The explanation of methods often uses descriptive language, lacks formal expression, lacks clear organization, and cannot highlight the differences from existing methods.

3. The results and analysis explanation of the paper are not sufficient: the correct information of the canal system should be given. and illustrations, classification statistics according to Table 1, and other data, images, and comparative recognition effects should be provided; A comparative explanation of the impact of different factors or parameters on recognition performance should be provided. Clear pictures and statistical data for classification and recognition should be provided. In addition, the hardware performance, running time, and resource consumption used should be explained.

4. The data and methods in the paper lack comparative statistical analysis with existing methods. Why can water channels below 1m be identified compared with reference 24? Lack of comparative analysis with existing methods in terms of accuracy and precision.

5. There are some formal and normative issues with the paper. Some images are not clear, such as Figure 3, Figure 5, and Figure 6; Some figure captions have inconsistent formats, with some captions placed above the images, such as in Figures 4 and 5; Some of the image content lacks description or explanation, such as Figure 7 and Figure 8.

Comments on the Quality of English Language

Minor editing of English language required.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

Dear Authors,

The article has improved significantly; however, some improvements are still necessary. Please see my comments below:

The authors mention an interval in line 185 but do not present it. Please include the corresponding interval.

The discussion needs improvement, as some sentences are unclear. The authors should reorganize their ideas for better clarity.

Sincerely,

Comments on the Quality of English Language

The authors need to make a minor revision to the English.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The previous questions have been well addressed and there are no new issues.

Author Response

Please see the attachment

Author Response File: Author Response.pdf