Winter Wheat Take-All Disease Index Estimation Model Based on Hyperspectral Data

Round 1

Reviewer 1 Report

Title: Winter Wheat Take-All Disease Index Estimation Model Based 2 on Hyperspectral Data

Journal: MDPI Applied Sciences

Authors: Guo Wei, Yang Yifeng, Zhao Hengqian, Song Rui, Dong Ping, Jin Qian, Liu Zelong and Yang Zihan

Summary:
The authors develop a hyperspectral identification method for the take-all disease of wheat based on the normalized difference spectral index of ground spectrometer data from a winter wheat field.

Comments:

- In the introduction the authors state that take-all is a kind of soil-dwelling bacteria.
Even a brief web search reveals that take-all is a fungus. 

- page 2, line 78: '...by hyperspectral' Please state what hyperspectral thing you mean here.

- page 3: Please provide a reference for standard GB-T-17980.109-2004

- page 3: How was the root disease area determined in table 1? Did you use image processing or did you estimate the area with the naked eye?

Section 2 is well written.

- Section 3, Figure 1: There is a monotonous decreasing trend for the reflectance of diseased wheat between 750 and 950 nm.
However, the reflectance between 450 and 630 nm shows the reflectance to be sometimes higher, sometimes lower than the healthy plant.
Please explain.

- page 6, Figure 2: Please replot figure 2 without interpolation.

- Figre 3: the ordinate label of the graph is not visible.

General comment: 

The number of measured datasets is quite low.

Please make your full dataset publicly available.

Please add photos of:
- the wheat field where the data were measured.
- a sample wheat plant for each of the four grades of the disease index.
- the spectral radiometer you used.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

The manuscript “Winter Wheat Take-All Disease Index Estimation Model Based on Hyperspectral Data” by Guo Wei et al. is devoted to developing disease evaluation techniques based on hyperspectral data. The work has some merits (e.g., Fig 2 is an excellent visualization of the data). However, it has substantial drawbacks, which prevent it from publishing in the current form. First, I would say that the poorly written text prevents a full assessment of the article. Furthermore, based on my understanding, the methods are insufficient (see critique below).

Major shortcomings

1. Line 116: What do the authors meant by “According to different degrees of disease, 30 samples were randomly selected”? It is either random or “according to different degrees.”
2. Line 188: “30 sample points of measured data, among which the first 15 sample points were selected for modeling and the last 15 sample points for verification”. In line 136, the authors stated, “In this data, there were 7 healthy samples, 16 mild samples, 5 moderate samples, and 2 severe samples”. So, how the authors selected 15 training and 15 test points from these data? The described method (line 188) is unacceptable
3. Was the model trained on 15 spectra collected on one day on one field? Again, I have severe doubts about the extrapolations of these data.
4. The details of the correlation analysis (disease index vs. spectral index) need to be better explained.
5. Line 66: How this work is different from the previous work of the same author (Guo, W.; Zhu, Y.; Wang, H. Study on winter wheat take-all monitoring model based on UAV hyperspectral image. 359 Transactions of the Chinese Society for Agricultural Machinery. 2019,50,162-169. ). It is stated that Guo Wei et al. 66 [11] built a winter wheat take-all monitoring model by using the relationship between spectral index and disease index and successfully analyzed the degree of wheat disease, which also provided a solution for monitoring winter wheat and other crop diseases on the plot scale. “What is the contribution to the field?
6. There is no comparison with other results, e.g. [11] by the same authors
7. There is no future work
8. References: a lot of citations from obscure university journals. Are they peer-reviewed?

Minor shortcomings (not the complete list. I stopped at some point)

9. Line 34: Latin name of the “take all” bacteria would be helpful.
10. Line 35-36: What is the “diseased body” the authors refer to? Dead animal or something else? Some examples in brackets would be helpful
11. Line 124: It is unclear what the authors meant by “the incidence of root disease was investigated under a white back ground.”. Is it a visual inspection? By a trained specialist or a random person?
12. Line 127 Table 1: Are grades consistent with descriptions? Grades 2,4,6,8 are missing.
13. Line 134: What is the “×Disease progression” in Eq.2? It is not defined
14. Line 138: Please refer to the manufacturer in a standard way, the name and country. What is the resolution bandwidth of the spectrometer?
15. Line 143: The picture or schematics of the experimental measurements would be helpful. It is not clear how data were gathered—E.g. sampling area.
16. Line 147: “at the same time”??? Did the authors mean sequentially?
17. Line 186: What are x and y in Eq. 4 and 5? It seems that they were copied and pasted from something else
18. It is not clear how Table 3 was produced. Some information can be derived from Table 4 (fitting formulas). However, the text in the body of the article is uncomprehensible and needs to be rewritten
19. Figure 3: What s the y axis?

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

The paper is concise. Perhaps the grammar needs a little bit of attention. The informal writing style allows ease of reading but occasionally becomes more casual than what you would expect.

Take all disease is caused by a fungus, not a bacteria, maybe give the name?

The paper would benefit from replication, as the 30 samples were from one location at a single point in time.

Were there any other diseases, pests or environmental stress? I don't think the paper currently shows conclusive evidence that the approach can be used to identify the disease from amongst other causes of plant stress (which can have similar appearances).

45 and 46 – need a reference for the claim

Test area section-  could you include the climate of the area

144-146 – it currently reads that there were no shadows in the captured images, what was the look area/field of view of the camera? How is it possible that no shadow from the canopy was captured?

There is only one reference in the discussion, it would be better to see the results discussed in the context of other reports.

The statement at the end of the paper claiming that the method is suitable for use with UAV imaging is unsupported.

There are a few inconsistencies in the references

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

The authors improved the article significantly. My primary concerns have been addressed.

The article can be published if the minor deficiencies were addressed:

Line 138: Disease grades cited in Table 1 are not of common knowledge. The authors referred to “The national standard: GB/T 17980.109-2004” in the rebuttals. I recommend including it in the description

Line 151: Picture label is merged with the following paragraph

Fig 2 is helpful. However, it is not referenced in the text.

Line 249: Fig 2? I guess it should be Fig 3

Line 268-271: The paragraph is merged with the Table label.

Line 274-279: Text is merged with the table label

Line 348: The in-line formula is not legible. I recommend separating it in equation 6

Line 352-363: The text is merged with the Author contribution

Author Response

Response to Reviewer 2 Comments

Point 1:

Line 138: Disease grades cited in Table 1 are not of common knowledge. The authors referred to “The national standard: GB/T 17980.109-2004” in the rebuttals. I recommend including it in the description

Response 1:

We have added the English name of the national standard into the article, and it is mentioned that the assessment of the disease grade was carried out according to this national standard.

 

 

Point 2:

Line 151: Picture label is merged with the following paragraph

Response 2:

Sorry, this is a format problem and has been adjusted.

 

 

Point 3:

Fig 2 is helpful. However, it is not referenced in the text.

Response 3:

Thank you for your carefulness, we have added the description of Table 2 in the manuscript.

 

Point 4:

Line 249: Fig 2? I guess it should be Fig 3

Response 4:

Thanks for your suggestion. It is indeed Fig 3.

 

Point 5:

Line 268-271: The paragraph is merged with the Table label.

Response 5:

Format problem, corrected.

 

Point 6:

Line 274-279: Text is merged with the table label

Response 6:

Format problem, corrected.

 

Point 7:

Line 348: The in-line formula is not legible. I recommend separating it in equation 6

Response 7:

The formula on Line 348 has been listed separately as equation 6.

 

Point 8:

Line 352-363: The text is merged with the Author contribution

Response 8:

Format problem, corrected.