Hyperspectral Image Analysis and Machine Learning Techniques for Crop Disease Detection and Identification: A Review

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript addresses an innovative integration of hyperspectral imaging and machine learning algorithms for precision agriculture, a timely topic given the increasing needs for sustainable agricultural practices. The novelty lies in the comprehensive review of current technologies and the exploration of both the potentials and challenges of these technologies in agriculture.

Limitations:

- While the review provides an overview of hyperspectral imaging and machine learning applications in agriculture, it lacks depth in explaining the machine learning models. For example, differences between algorithms like SVM, RF, and deep learning models like CNN are not explored in detail.

- The manuscript lacks a detailed methodology section that explains how literature was selected and analyzed. This omission may affect the reproducibility of the review and the perceived reliability of the conclusions drawn.

- The paper discusses theoretical aspects and general findings from various studies but does not provide concrete examples or case studies where these technologies have been applied.

- The manuscript is generally well-written but could benefit from proofreading to correct minor grammatical errors and improve sentence structure for enhanced readability.

- The paper effectively highlights the integration of hyperspectral imaging and machine learning in agriculture, but it could better articulate the significance of this integration in addressing global challenges

Suggestions:

- Enhance the technical descriptions by comparing the performance, advantages, and drawbacks of different machine learning models specifically in the context of processing hyperspectral data.

- Include a methodology section detailing the search strategy, databases used, inclusion and exclusion criteria, and the process for synthesizing the information from the selected studies.

- Incorporate one or two detailed case studies showing how hyperspectral imaging coupled with machine learning has been successfully implemented in real-world agricultural settings.

- Discuss the potential for interdisciplinary approaches combining hyperspectral imaging and machine learning with other fields such as genomics or climate science, which could open new avenues for research and application.

- Emphasize the potential impact of this technology on global agricultural practices, including how it can contribute to sustainable agriculture by allowing for more precise use of resources and early detection of crop diseases.

- Conduct a thorough proofreading session to correct grammatical errors. Consider using professional language editing services to ensure the manuscript meets the high standards of academic publishing.

- Increasing the number of figures and tables to summarize the findings or demonstrate the technologies discussed could enhance reader understanding and engagement.

- Suggested References:
Ahmed, I., & Yadav, P. K. (2022). A systematic analysis of machine learning and deep learning based approaches for identifying and diagnosing plant diseases. Sustainable Operations and Computers, 4, 96-104. https://doi.org/10.1016/j.susoc.2023.03.001

Mamba Kabala, D., Hafiane, A., Bobelin, L., & Canals, R. (2023). Image-based crop disease detection with federated learning. Scientific Reports, 13(1), 1-19. https://doi.org/10.1038/s41598-023-46218-5

Hasan, M. M., Uddin, A. F., Akhond, M. R., Uddin, M. J., Hossain, M. A., & Hossain, M. A. (2023). Machine Learning and Image Processing Techniques for Rice Disease Detection: A Critical Analysis. International Journal of Plant Biology, 14(4), 1190-1207. https://doi.org/10.3390/ijpb14040087

El-gayar, M., Soliman, H., & Meky, N. (2013). A comparative study of image low level feature extraction algorithms. Egyptian Informatics Journal, 14(2), 175-181. https://doi.org/10.1016/j.eij.2013.06.003

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

Reviewer 2 Report

Comments and Suggestions for Authors

This is an interesting study that provides a technical overview of the major applications of hyperspectral imagery in crops, prospects and challenges of combining artificial intelligence algorithms such as machine learning and deep learning in the classification and detection of crop diseases in cereals, oilseeds, fruits and vegetables. I have some comments on the presentation of the text, the logic of the lines, the structure of the article, and the citation of references in this study:

1.       Please highlight the findings of the study in the abstract section.

 

2.       Please add the reference to “Among these technologies, artificial intelligence and image analysis supported by the use of hyperspectral cameras is a growing trend”.

 

3.       The symbols used to express the range of values in Figure 1 are not uniform, so please standardize the use of symbols according to the requirements of journals.

 

4.       Some of the concluding statements in “1. Introduction” lack relevant references, such as “they have obtained new findings by processing these images with techniques based on machine learning and deep learning, thus achieving more efficient analysis methods and saving time when the crops are large. techniques based on machine learning and deep learning, thus achieving more efficient analysis methods and saving time when the crops are large. “

 

5.       The “2. Methodology” section of the text is too elaborate and complex, so please use a concise presentation of the research methodology used in this study.

 

6.       The section “3.1. Technology Based On Hyperspectral Images” is similar to the introduction to “1. Introduction” and should be streamlined. Please highlight the findings of this study on the topic of “Technology Based On Hyperspectral Images” and discuss them in comparison with related studies.

 

7.       Hyperspectral images, when using specialized sensors, can measure the radiation reflected or emitted by plants in multiple spectral bands, which allows obtaining detailed information. Hyperspectral images, when used with specialized sensors, can measure the radiation reflected or emitted by plants in multiple spectral bands, which allows obtaining detailed information about the state of the plants and their characteristics, as well as the radiation intensity of the plant. about the state of the plants and their characteristics, as well as traits of their leaves, stems, soil nutrient level, and early diagnosis to determine The expression “any type” in “any type of anomaly or disease” is not rigorous, please check.

 

8.       The structure of the study is problematic and lacks results on the content of the study, please highlight the findings that the study is innovative.

 

 

9.       In the text “4. Conclusions”, it is suggested that the results of this study be stated in a concise statement.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This manuscript represents a great effort to compile information, applicable in different areas.

However, it is necessary to take care of great details to be a scientific manuscript.

The numbering of the figures, review, give more information regarding each of the figures, the source of information or if it is self-developed. Its usefulness is pertinent, since with this information decisions are made in different areas.

It is suggested to review the scientific names

You could present a picture of Technology based on hyperspectral images, relating differences and similarities. It is interesting and with very recent literature. Surely the patents for Technology based on hyperspectral images are older; But the authors write well, the review of publications considers from 2013 to date.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

This paper proposes a technical review of the main applications of hyperspectral images in agricultural crops, perspectives and challenges that combine artificial intelligence algorithms such as machine learning and deep learning, in classification and detection of diseases of crops such as cereals, oilseeds, fruits and vegetables.
Authors explain different machine learning algorithms and their possible use cases, reviewing more than 100 papers, to analyze hyperspectral images in agriculture, for crop disease detection and identification.
The paper provides a comprehensive review of images in agricultural crops, identifying the main algorithms based on machine learning and deep learning for detection and classification, as well as the spectral ranges mainly used.

Remarks for the paper:
1. Some other similar reviews should be also included and compared with the proposed review, i.e. [1], [2], [3] (possibly, a meta-analysis subsection). Comparison with the existing reviews that cover similar topics is missing and should be included in the revised paper, explaining what is newly presented in the proposed paper.
2. UAV paper analysis might be also added in the Results section, as it is only mentioned in the conclusions. Additionally, terrestrial/aerial use cases might be added and possibly compared.
3. Conclusions are well presented, taking into account previously described papers. However, new conclusions might be added, after examining proposed references [1-3].
4. Some references are missing and should be added in the paper, such as [1-3].
5. Additional comparison might be added, regarding hyperspectral versus normal RGB images in agriculture. I.e. when should each one be used, in which cases?
6. Regarding the quality of the data presented in the paper:
a) Figure 6 should be analyzed and explained (also) before the conclusions. Currently it is only mentioned in the conclusion section, while the figure itself is in "Results and Discussion" section
b) Results and discussion section might be also divided in 2 sections.
c) Check all tables and their header sections.
d) Remove the sentence on the lines 273-275 that does not belong to the subsection 3.3.4.

Missing references:
[1] Atiya Khan, Amol D. Vibhute, Shankar Mali, C.H. Patil, " A systematic review on hyperspectral imaging technology with a machine and deep learning methodology for agricultural applications", Ecological Informatics, Volume 69, July 2022, 101678
[2] Billy G. Ram, Peter Oduor, C. Igathinathane, Kirk Howatt, Xin Sun, " A systematic review of hyperspectral imaging in precision agriculture: Analysis of its current state and future prospects", Computers and Electronics in Agriculture, Volume 222, July 2024, 109037
[3] Yu H, Kong B, Hou Y, Xu X, Chen T, Liu X. A critical review on applications of hyperspectral remote sensing in crop monitoring. Experimental Agriculture. 2022;58:e26. doi:10.1017/S0014479722000278

Comments on the Quality of English Language

Language/grammar remarks:

1.       Line 183, "The waveThe results".

2.       Line 167, probably "Fourth, all off-topic articles that did not contribute to the study were deleted."

3.       Line 186, "detectionlength".

4.       Rephrase line 92 "Second, was perform multiple search combinations…"

5.       Sentence on the lines 358-361 needs to be checked.

6.       Line 404 "harvest. in".

7.       Header "Version May 6, 2024 submitted to Sensors".

Overall, language and grammar should be checked.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have made substantial improvements to the manuscript. In light of these revisions, I believe the manuscript has been significantly enhanced and I am satisfied with the current state of the work.

Comments on the Quality of English Language

Minor editing of English language required

Author Response

Comments 1: [The authors have made substantial improvements to the manuscript. In light of these revisions, I believe the manuscript has been significantly enhanced and I am satisfied with the current state of the work.]

Response 1: [We agree with this comment. Thank you so much!]

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript has improved considerably, and it is understood

In the conclusion

 

Hyperspectral imaging technology allows crops to be monitored without any physical contact or damage; and it is a widely used tool with highly positive results. In the conclusion it mentions that it generates cost savings and higher returns; costs are not described in the manuscript.

 

Author Response

Comments 1: [The manuscript has improved considerably, and it is understood In the conclusion.

Hyperspectral imaging technology allows crops to be monitored without any physical contact or damage; and it is a widely used tool with highly positive results. In the conclusion it mentions that it generates cost savings and higher returns; costs are not described in the manuscript. ]

Response 1: [We agree with this comment. We try to put approximate costs, but given the large number of crops, there is a great variety of technology integration  between the sensors, the mounting structure and the computing servers for processing, therefore They can generate different combinations and costs can vary. At first the hyperespectral sensor is the most expensive investment, but in the long term Monitoring and prediction can save many expenses due to crop losses due to different events, improve production processes and make better decisions, allowing the initial investment to be recovered.]

Reviewer 4 Report

Comments and Suggestions for Authors

The authors have answered all of my comments.

Author Response

Comments 1: [The authors have answered all of my comments.]

 

Response 1: [We agree with this comment. Thank you so much!]