Remote Sensing and Spatial Analysis for Land-Take Assessment in Basilicata Region (Southern Italy)

Round 1

Reviewer 1 Report

Dear Authors,

   Thanks a lot for your manuscript submission to MDPI Journal of Remote Sensing.  After careful review, my comprehensive evaluation on this paper is a very good set of work. The study is fair, complete and convincing, the literal writing is also moderately acceptable. Hence, this research article, can be recommended as "Acceptance with Minor Edits", after fixing a few major and minor problematic issues, which I specify each of them as follows (may not limited to these as mentioned below):

   Major problematic issues suggested for improvement in your revision:

   a) Abstract: Please shorten the 4 sentences in Lines 11-16. Be a bit more specific on the quantitative results in Line 26. Meanwhile, I think after editing, the perfect abstract session should be of length in 150~180 words.

   b) Introduction: a few problems need to be addressed. The first sentence at the first paragraph might have grammatical errors, consider editing. The historical review had better include more specific details, pros and cons, then address the research problem to be solved. Meanwhile, the last two passages are a bit too generic. Some suggested edits for you: condense the two parts into a main summary of your research contributions, and then summarize 3-4 manifolds of your technical work. Add a 5th paragraph to introduce the remainder of this paper (which is organized as follows: ...), for your reference.

   c) Section 2 (Materials and Methods): the current version taks up 5 page, which is too long. Please condense this section, especially the narrations in Subsection 2.1. Besides, the resolution of sub-diagrams in Fig. 2, should be enhanced. This figure contains some blurs, and I don't think it is necesssary to capitalize each letter in the rectangle frames. Please update. Thanks.

   d) Approach: one transparent issue is that this research article lacks any performance metric (along with the related formulas for machine learning tools and classifiers, no derivation of your mathematical models). Please either explain or supplement some necessary components. 

   e) Figures and Tables: I feel that Figs. 6-14 seems repetitive, there should be some alternative ways to display similar results. Also, except for Table 3, the experimental section has the potential defects of lacking (tabulated) quantitative results. Regarding the other figures, I don't think that simply displaying some visual results may factorize adding the importance or strengthening the significance of your approach. Please explain and update.

   f) If possible, can the authors enumerate 2-3 actual innovative aspects of your study? What are the advantages of your approach when combining the major classical machine learning classifiers such as SVM, kNN, etc?

   g) Discussions: the current shape is a bit too generic. In addtion to one single limitation of your whole paper, I think discussions should include your justification on the limitations of study (with specific statements) along with some of the quantitative results. Also, if any ablation study or sensitivity analysis were proposed, those can be included in the expanded section.

   h) Conclusions: the structure of current version looks fine. While I think the emphasis of summarized work should address keynote concluding remarks (along with major quantitative scores), Besides, the last paragraph can be further polished by supplementing some statements on summary of research challenges and a few specific orientations on your future study.

   i) References: Citations for most journal publications are acceptable. Suggestions for edits: (i) The authors cited quite a few Google scholar research materials, if those had got published, please consider citing the most professional version (published in a journal or highly visible conference proceedings). (ii) Regarding the online sources, I think the authors need to check the MDPI template and calibrate each of them for the current citation format. (iii) Besides, it is recommended to cite more published works within the latest 3 years (i.e., March 2020 to March 2022) in the related IEEE journals (i.e., Tans. GRS, GRS Letter) and a few closely related MDPI publications (i.e., Sustainability, Remote Sensing, Geosciences, etc.), which also need to comply with the MDPI template requirement for citations. 

   Minor issues to be calibrated in the revised version:
   1) If you use MS word or Latex, please avoid hyphenating a word (which currently appears multiple time at the end of some lines to cross-over two adjacent lines). The  MDPI online template has the options to adjust that. 
   2) While the literal quality of English writing is quite acceptable, there is still room for improvement in the updated version. Be sure the half-space interval are fine among words between sentences. Some redundant space should be removed. Please proof reading the related context carefully to eliminate any grammatical errors and improper use of English phrases.
   3) Fix some minor issues, i.e, the missed full-stops at the end of Figs. 9, 11 and 13, the foggy appearance in Figs. 12 and 14, the number of valid digits (after decimal) are expected to be uniform in Table 3, and the out-of-right boundary issues in Figs. 4, 5, and 16.   

   Once again, we wish you the best of luck for your research article coming into acceptance. Thanks for your interests on publishing at MDPI affiliated Journals. We expect your success in the near future. Take care!

Best wishes,

Yours faithfully,

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

This research concerns a practical study on land take time series, carried out through remote sensing data by the Landsat Mission, aiming to provide accurate information for selected sample area in Basilicata Region from 1994 to 2014.

Unfortunately, I do not see anything new in this manuscript. I encourage the authors to find new methods on this topic.

Author Response

Dear Reviewer,

thank you for your review. The purpose of our work and its innovation lies in having created a quick methodology for the historical quantification of the waterproofed soil in the indicated areas, and it is described in the text. The innovation lies in having integrated the analysis and classification of land cover and land use of the Landsat data through the SVM classification algorithm and with the use of ancillary data from the geotopographic database of the Basilicata region. This database contains information layers relating to urbanized soil.
Sincerely,

the authors

Reviewer 3 Report

Reviewer’s Report on the manuscript entitled:

Remote sensing and spatial analysis for land take assessment in Basilicata Region (Southern Italy)

The authors proposed a classification method to detect land cover information and to identify impermeable soils in the Basilicata region in Italy. I found the topic, methodology, and results interesting and useful. However, the presentation needs further improvement. Please see below my comments.

Line 19. Grammar issue: “it is crucial” not “is crucial”

Line 63. Please also add here the following most recent articles for deep learning and deep transfer learning for LULC classification:

Deep learning: https://doi.org/10.3390/rs12122000

Deep transfer learning: https://doi.org/10.3390/s21238083

Line 66. Please also add the following article for the use of Landsat and MODIS for LULC delineation: https://doi.org/10.3390/s19224891

Lines 62-64. Please use a consistent format when you define the acronyms (e.g., capitalize the first letters). Please also ensure that all the acronyms are defined and add an acronym table at the end of the manuscript listing all the acronyms used in the manuscript.

Line 80. Support Vector Machine (SVM) is already defined in line 62. So you can simply say here SVM.

Line 83. Please describe how the rest of the paper is organized. For example, something like:

“The rest of the paper is organized as follows. Section 2 describes the study region, datasets, and methodology. Section 3 demonstrates the results, etc.”

Line 92. Please don’t start a sentence with a number. You can write “Thirty percent” instead.

Lines 95-104. Please merge all these three sentences into one paragraph. Please note that a paragraph should usually contain at least three sentences. Similarly, join lines 217-231, join lines 308-317, join lines 353-361, join lines 443-462etc.

Line 11. Please fix the parenthesis “)” after Potenza.

Figure 1. Please enlarge the font size of the text in the bottom left map. Please also insert the latitude and longitude grids and scale bar in Kilometer and the sign for the north direction.

Line 126. Grammar issue. Please re-write this sentence.

Section 2.3 also describes the dataset. Please join this subsection to the previous subsection and have the title written as 2.2 Datasets and pre-processing.

Line 208. Please replace “non-supervised” with “unsupervised”. Please check and correct elsewhere.

Figure 2. Please enlarge the font size of the texts in the flowchart. Also, in the Caption. Please note that “flowchart” is one word.

Lines 241, 263. The acronym ROI is already defined.  

Line 377. Please write it as (Figures 7 and 8)

Figure 8. Optional. It would be nicer if you use the same colors for the bar chart as in Figure 7, i.e., red, pink, blue. Similarly, for Figures 10 and 12, and 14.

Line 391.  Typo: “a rte-rial roads”?

Thank you for your contribution

Regards,

Author Response

Dear Reviewer,

Thank you for the suggestions you have given us to improve our paper. We have made all the changes you suggested and supplemented our work with the scientific articles you suggested. Changes have been made to the images and caption of Figures 7 - 8 -10 - 12 - 14. Suggested references have been included in the text and grammatical and syntax changes suggested in its revision have been made.

Yours faithfully,

The authors

Reviewer 4 Report

Authors contributions:

According to the authors, remote sensing is the primary technique for extracting land use/land cover (LULC) data. Also, many approaches have been introduced and explored to optimize the resulting classification maps by using several spectral indices and algorithms.

In this work, the authors have been created an expeditious methodology for classifying multitemporal Landsat imagery, using the semi-automatic classification algorithm in a Geographical information system (GIS) environment for mapping land use. The method developed has the potential of Support Vector Machine (SVM) based on machine learning theory to produce a synthetic map of land take provides valuable and detailed information to improve the accuracy of land cover mapping in complex landscapes and environments such as urban peri-urban areas.

 

I have some reviewer notes:

Abstract. In the beginning of this part, you have to show what is the problem with the known solutions in this study area. With one short sentence.

Introduction. At the end of this part, it will be good to describe your contributions.

2.1 Study Area. It will be good to show geographical coordinates of the study area. For example, in WGS84.

Figure 1. What is the measurement unit of the altitude?

Line 217. QGIS software. You have to show manufacturer and country of origin for this software.

2.4 Support vector machine (SVM) and classification methodology. SVM is supervised method. How did you choose the volume of training, validation and test data sets? It is not clear.

Figure 2. What will have happened if the “Accuracy” is equal to 90%? You can change the block with rhomb, and check <=90% or >90%. There must be a case when accuracy is equal to 90%.

Figure 6. Use grid lines, if it is possible.

Figures 8 and 10. If there is a Standard deviation in the measurement data, you have to show it on the chart.

Discussion part. You have to compare your results with those from other authors. Not simply put the citations at the end of this part.

 

I have some suggestions:

Reduce the using of Internet sources. It is scientific paper and you have to cite more scientific works. Improve the presentation of your algorithms. Make more comparative analyses. The corrections of these notes will increase the contribution of your work.

Author Response

Dear Reviewer,

thank you for carefully reviewing our work. Changes and suggestions indicated by your review have been made within the text. Below are reported the major  improvements made at the same time as your observations.

The reference system used in the work is WGS84 / UTM Zone 33 N. The coordinates of the Basilicata region are shown in figure 2.1. The measurement unit of the altitude is meter and I have been corrected the figure 2.1.

QGIS software is an open source spatial analysis software. A citation has been included within the article that links to the software's website where all information can be found.

Based on multiple tests performed, trying to minimize classification errors, we chose 5% of the total area as the volume of training areas. The value of 5% guarantees us the best classification result, with an overall accuracy greater than 92%.

The results have been validated using the layers of the geotopographic database of the Basilicata Region (DBGT) regarding the impermeable areas (roads, railways, buildings, factories, etc.).

A further check of the results obtained has been made using the orthophotos present on the national geoportal.

The discussion chapter has been improved and supplemented. All other suggestions and corrections have been made in the text of the article.

 

Yours faithfully,

the authors.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

The quality of the manuscript is really improved. 

Author Response

Dear Reviewer,

thank you for carefully reviewing our work.

Yours faithfully,

The authors

Reviewer 3 Report

I would like to thank the authors for addressing my comments. However, I still have some suggestions:

Line 15, there are two commas.

Line 17. Shouldn't the abbreviation for geotopographic database be GTDB instead of DBGT? Please check and correct elsewhere.

Unfortunately, the authors seem to carelessly describe the literature. For example, line 63 should be "Heydary et al. [32] ...". Reference [33] talks about deep learning for LULC classification while Reference [34] talks about deep transfer learning for LULC classification. Similarly for line 65 and 66 and reference [35]. Please Carefully check all the references and describe them properly.

Reference [33], line 621. Please add the year of publication, volume, and article number for this article.

Reference [34], line 623. Please insert the volume (21) and article number (8083) here.

Please check all other references for such issues.

Figure 1. Please have the map of study region shown as a projection (e.g., WGS84) with latitude and longitude grids instead. Also, the top left map could show the full map of Italy instead of part of it.

Line 155. Grammar issue: It should be "has been developed"

Please merge lines 218-220 to the previous paragraph.

Please merge lines 296-297 to the previous paragraph. Line 452 and 458 to the previous paragraph. As I mentioned before, authors should not use less than three sentences in a paragraph.

Please check all other references for such issues.

Please very carefully proofread the manuscript and improve the quality of the figures (resolutions minimum 300dpi, the text and font size should be enlarged in some of the figure for better readability, etc.)

Thank you for your contribution

Author Response

"Please see the attachment."

Author Response File: Author Response.docx

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.