Red Palm Weevil Detection in Date Palm Using Temporal UAV Imagery

Round 1

Reviewer 1 Report

The authors evaluate the potential of a remote sensing approach for timely and reliable detection of the physiological consequences of Red Palm Weevil (RPW) infestation on the palm canopy. I find the topic interesting, but it need to address some concerns, listed as follows: 

1- Please separate the material and methods into three subsections: 1- Study area, 2- Dataset, and 3- methodology. 

2- please list the characteristics of the camera mounted on the UAV, that the authors considered for this work in a separate table. I and readers may want to see spectral, spatial, and radiometric resolutions of the UAV images. 

3- How the authors generate the vegetation and other indices in this paper. please list them in a table by formula and citations. Readers may want to generate solely the indices from the images. Moreover, If authors present spectral bands of images in previous comments this comment can be easily presented. 

5- please present the formula of GLCM features that the authors used in this study.

4- I can't see some parts of figure 4 (workflow diagram). Please represent again in a true way. 

5- please present key contributions(s) of the study in the last paragraph of the introduction in summary. 

6- please summarize the discussion parts and highlight their importance. 

7- is it possible to do this study with the deep learning method? please discuss it in a paragraph in the discussion, and please present this in conclusion section for the future works. 

Author Response

Dear reviewer, thanks a lot for reviewing our manuscript! We have addressed your comments and suggestions in the attached review report and in the updated version of the manuscript. 

Author Response File: Author Response.docx

Reviewer 2 Report

Dear authors, I had the pleasure to read and review your scientific work. Topics like yours contribute a lot to science, economy and ecology, which is the special value of your work. After analyzing your work, I have a couple of suggestions that, it seems to me, would contribute to the quality of the paper.

Abstract - the content of the abstract does not reflect the content of the paper. I suggest you shorten it, edit it and add more information about the research results.

Introduction - in the introduction, attention should be paid and more information should be given about similar research that used similar or the same methods. Also, show what were the results and conclusions of those investigations and how your research differs from previous ones.

Materials and Methods - when I first read your paper I had the impression that the small series of data and sample trees was the main problem of your research. In addition, further validation of your method is also problematic. However, as a reviewer, I accept that this is an experiment that still needs to be proven and I accept that your results are certainly interesting to be published as an incentive for further research in this field.

I have no objections or suggestions for changes in the Results and Conclusion sections.

Best Regards.

Author Response

Dear reviewer, thanks a lot for reviewing our manuscript. We have addressed your comments and suggestions in the attached review report and in the updated version of the manuscript. 

Author Response File: Author Response.docx

Reviewer 3 Report

 

To limit the palm trees’ loss and control of Red Palm Weevils, non-destructive detection of this infestation like remote sensing is of great importance. This paper investigated the capacity of RGB, multispectral and thermal UAV imagery to predict the presence of RPW infestations. Authors found maximum temperature and standard deviations of the temperature within the crown revealed good discriminatory performance between infested and non-infested palms. These findings are interesting for both management practice and related studies. I suggest add detailed information for some key vegetation indices used in this paper, discuss further why RGB and multispectral data performed not as well as thermal data, and improve the quality of figures before publication. More comments are listed below.

 

1)      L84-85, delete ‘were defined. The main aim of this work ’.

2)      L124, change ‘method of infestation’ to ‘method for infestation’.

3)      L158, ‘UAS flight’ means ‘UAV flight’?

4)      L171-172, what is the empirical line model for correction? And where is the model from?

5)      L176, change ‘from’ to ‘for’?

6)      L203-205, delete the blank space.

7)      L226-229, correct the layout of Figure 4.

8)      L242-244, and L251-252, correct the layout of Figure 5.

9)      L248 and hereafter, what is palm n°37? No.37?

10)   L342-345, correct the direction of the legend NDRE to upward from downward, and adjust the scale location.

11)    As show in L183-185,there are many VIs, so why this paper focus on NDRE, and how to calculate it, please show clearly in state of offering references..

12)   L350-351, the model selected the RCC index as the best indicator …,how you selected?

13)   L350-351, If the index value was equal to or higher than 0.38…,give a figure or table to support this.

14)   What is the spatial resolution of the thermal data? Generally how many thermal pixels covers a canopy? Is the number enough for statistics analysis?

 

Author Response

Dear reviewer, thanks a lot for reviewing our manuscript! We have addressed your comments and suggestions in the attached review report and in the updated version of the manuscript. 

Author Response File: Author Response.docx

Reviewer 4 Report

1. Please highlight this paper’s contribution in the last paragraph of Sec. 1
2. In Table 1, I think the authors should replace “x” with “√”
3. Please resize Fig. 4 and reorganized Fig. 5
4. Please explain why you select these vegetation indices introduced in Line 180.
5. Please give an explanation of Wilcoxon Signed-Rank Test
6. Please explain why you give “*” in Table 3.
7. The related works lack of some literature about palm tree detection and the method review. Here we list some related palm tree detection papers.

(1) ISPRS21-Growing status observation for oil palm trees using Unmanned Aerial Vehicle (UAV) images. (UAV image in agriculture & growing status monitoring & palm tree detection)

(2) ISPRS21-Automatic large scale detection of red palm weevil infestation using street view images. (red palm tree detection)

(3) ISPRS20-Cross-regional oil palm tree counting and detection via a multi-level attention domain adaptation network. (palm tree detection)

(4) GRSL21-Multisource-domain generalization-based oil palm tree detection using very-high-resolution (vhr) satellite images. (palm tree detection)

Author Response

Dear reviewer, thanks a lot for reviewing our manuscript! We have addressed your comments and suggestion in the attached review report and in the updated version of the manuscript.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

All of my comments were addressed by the authors. As for me this work well done. 

Author Response

Thank you for reviewing this manuscript again. 

Reviewer 4 Report

The authors have addressed most of issues. However, I still think these two papers are really related to your work and I recommend you to cite them:

(1) ISPRS21-Growing status observation for oil palm trees using Unmanned Aerial Vehicle (UAV) images. (UAV image in agriculture & growing status monitoring & palm tree detection)

(2) RSE23-Surveying coconut trees using high-resolution satellite imagery in remote atolls of the Pacific Ocean. (palm tree detection)

Author Response

Thanks a lot for the interesting references. I have added them to the manuscript.