Alpha-EIOU-YOLOv8: An Improved Algorithm for Rice Leaf Disease Detection

Round 1

Reviewer 1 Report

This paper focuses on the early detection of diseases in rice leaves, which is crucial for controlling the spread of infections and improving the quality of food crops. The authors propose the use of deep learning approaches, specifically a Convolutional Neural Network (CNN) based on a modified version of YOLOv8. To assess the performance of their proposed approach, the authors conduct a comparative study with existing methods. The experimental results show that their proposed model achieves an accuracy of up to 89.9% on a dataset of 3175 images, with 2608 images for training, 326 for validation, and 241 for testing. This indicates that their approach achieves a higher accuracy rate compared to existing methods based on the results obtained in this study. Overall, the comprehensive evaluation of the work, including the methodology, the datasets used, and the model's robustness under various conditions, appears quite robust. Additionally, the practical applicability of implementing this approach in real agricultural environments would be an important aspect to consider in future research.

Regarding my comments, I notice that the work, as well as the English, is very understandable. However, I cannot help but mention some important considerations. The first is that all scientific names, whether for rice or the fungi mentioned in the paper, are not italicized, which is not acceptable. Next, I believe that Figure 1 to 3 could easily be combined into a single figure, distinguishing each image and its respective disease with letters (a, b, c). Also, Figure 6 is not in the correct format, unlike the other figures (I am referring to the legend). Additionally, the subtitles (numbers) in the sections are inconsistent, with some in italics and others not.

Author Response

Authors’ responses to the Reviewers’ comments
Manuscript number: agriengineering-2763623
Title: Alpha-EIOU-YOLOv8: An improved algorithm for rice leaf diseases detection
Authors: Dong Cong Trinh, Anh Tuan Mac, Khanh Giap Dang, Huong Thanh Nguyen, Hoc Thai Nguyen and Thanh Dang Bui

January 10, 2024

Dear M.Sc. Leanne Fan and the Reviewers,

We sincerely thank and appreciate you and the reviewers for useful comments and constructive criticisms. Enclosed please find our responses to you and the reviewers of manuscript agriengineering-2763623. “Alpha-EIOU-YOLOv8: An improved algorithm for rice leaf diseases detection”. We have comprehensively addressed all of the comments provided by you and the reviewers and incorporated all of them in the revised paper.

Please find the detailed responses below and the corresponding revisions/corrections highlighted in the re-submitted files.

We sincerely thank you for all the insightful and constructively criticisms and suggestions. We belive that they have greatly strengthened the paper.

Thank you again for taking the time to review this paper.

Yours Faithfully,

A/Prof. Thanh Bui Dang, Dr Hoc Thai Nguyen, Dr Huong Thanh Nguyen, Mr. Khanh Giap Dang, Mr. Anh Tuan Mac, and M.Sc Dong Cong Trinh

Institute for Control Engineering and Automation,

Hanoi University of Science and Technology,

 Dai Co Viet Street, Hanoi, Vietnam.

Author Response File: Author Response.pdf

Reviewer 2 Report

1.      In abstract, authors says “. In order to assess the performance of the proposed approach, a comparative 12 study between our proposed method and some existing methods is conducted.” Don’t use the word some. Give specific methods.

2.      The type of variety of paddy mentioning is missing.

3.      Figure 1. Leaf blast disease, Figure 2. Leaf folder disease and Figure 3. Brown spot disease. Can be resized and used in a common place. It looks odd.

4.      Reorientation of Figure 5. Auto detecting rice disease system with Raspberry Pi 4 model B is required. It looks simple.

5.      F1−score = 2x precisionxrecall precision+recall (5). Check the correctness of this equitation.

6.      Unable to find the future direction of this current research study.

7.      Why the constant aspect ratio is followed (The width of the images varies from 106 to 4312 and the 142 height of the images varies from 150 to 5760.)

8.      What about the identification time? Nowhere that has been projected.

9.      What are packages used in Raspberry pi SBC?

1.  Comparison of various model is missing in the results and discussion section.

1.  The following paper can be added in the references.

1. Ramesh, S., Vydeki, D.: Recognition and classification of paddy leaf diseases using Optimized Deep Neural network with Jaya algorithm. Inf. Process. Agric. 7(2), 249–260 (2020)

2. “Maheswaran, S., Sathesh, S., Rithika, P., Shafiq, I.M., Nandita, S., Gomathi, R.D. (2022). Detection and Classification of Paddy Leaf Diseases Using Deep Learning (CNN). In: Neuhold, E.J., Fernando, X., Lu, J., Piramuthu, S., Chandrabose, A. (eds) Computer, Communication, and Signal Processing. ICCCSP 2022. IFIP Advances in Information and Communication Technology, vol 651. Springer, Cham. https://doi.org/10.1007/978-3-031-11633-9_6”

Minor editing of English language required

Author Response

Authors’ responses to the Reviewers’ comments
Manuscript number: agriengineering-2763623
Title: Alpha-EIOU-YOLOv8: An improved algorithm for rice leaf diseases detection
Authors: Dong Cong Trinh, Anh Tuan Mac, Khanh Giap Dang, Huong Thanh Nguyen, Hoc Thai Nguyen and Thanh Dang Bui

January 10, 2024

Dear M.Sc. Leanne Fan and the Reviewers,

We sincerely thank and appreciate you and the reviewers for useful comments and constructive criticisms. Enclosed please find our responses to you and the reviewers of manuscript agriengineering-2763623. “Alpha-EIOU-YOLOv8: An improved algorithm for rice leaf diseases detection”. We have comprehensively addressed all of the comments provided by you and the reviewers and incorporated all of them in the revised paper.

Please find the detailed responses below and the corresponding revisions/corrections highlighted in the re-submitted files.

We sincerely thank you for all the insightful and constructively criticisms and suggestions. We belive that they have greatly strengthened the paper.

Thank you again for taking the time to review this paper.

Yours Faithfully,

A/Prof. Thanh Bui Dang, Dr Hoc Thai Nguyen, Dr Huong Thanh Nguyen, Mr. Khanh Giap Dang, Mr. Anh Tuan Mac, and M.Sc Dong Cong Trinh

Institute for Control Engineering and Automation,

Hanoi University of Science and Technology,

 Dai Co Viet Street, Hanoi, Vietnam.

Author Response File: Author Response.pdf

Reviewer 3 Report

Contributions of the paper:

• The proposed modification of the YOLOv8 framework achieved better performance than some other state-of-the-art methods.
• The paper presents a proposed approach that achieved a higher accuracy rate compared to existing approaches in detecting rice leaf diseases such as blast leaf, leaf folder, and brown spot.

Comments/Suggestions:

• Provide more details about the specific modifications made to the YOLOv8 framework to achieve better performance.
• Include a comparison of the proposed approach with other state-of-the-art methods in terms of accuracy, precision, recall, and other relevant metrics.
• Clearly define the dataset used for training, validation, and testing, including the number of images and the distribution of different rice leaf diseases within the dataset.
• Discuss the limitations of the proposed approach, such as potential challenges in detecting certain types of rice leaf diseases or variations in lighting conditions.
• Include a discussion on the computational requirements and efficiency of the proposed approach, considering the potential deployment in real-world scenarios.
• Provide insights into the practical implications of the proposed approach, such as its potential for early detection and prevention of rice leaf diseases in Vietnam.
• Consider conducting user studies or expert evaluations to validate the effectiveness and usability of the proposed approach in real-world settings.
• Discuss the potential scalability of the proposed approach, considering the possibility of applying it to larger datasets or expanding it to detect other types of plant diseases.
• Provide recommendations for future research directions, such as exploring the use of deep learning techniques for automated diagnosis and treatment of rice leaf diseases.
• The authors are invited to include some recent references, especially some references related to  Deep Convolutional Neural Networks.  For instance, the authors may include the following interesting references (and others):
  
  a. https://www.mdpi.com/2073-431X/12/8/151
  
  b. https://www.taylorfrancis.com/chapters/edit/10.1201/9781003393030-10/learning-modeling-technique-convolution-neural-networks-online-education-fahad-alahmari-arshi-naim-hamed-alqa
•  

Can be improved

Author Response

Authors’ responses to the Reviewers’ comments
Manuscript number: agriengineering-2763623
Title: Alpha-EIOU-YOLOv8: An improved algorithm for rice leaf diseases detection
Authors: Dong Cong Trinh, Anh Tuan Mac, Khanh Giap Dang, Huong Thanh Nguyen, Hoc Thai Nguyen and Thanh Dang Bui

January 10, 2024

Dear M.Sc. Leanne Fan and the Reviewers,

We sincerely thank and appreciate you and the reviewers for useful comments and constructive criticisms. Enclosed please find our responses to you and the reviewers of manuscript agriengineering-2763623. “Alpha-EIOU-YOLOv8: An improved algorithm for rice leaf diseases detection”. We have comprehensively addressed all of the comments provided by you and the reviewers and incorporated all of them in the revised paper.

Please find the detailed responses below and the corresponding revisions/corrections highlighted in the re-submitted files.

We sincerely thank you for all the insightful and constructively criticisms and suggestions. We belive that they have greatly strengthened the paper.

Thank you again for taking the time to review this paper.

Yours Faithfully,

A/Prof. Thanh Bui Dang, Dr Hoc Thai Nguyen, Dr Huong Thanh Nguyen, Mr. Khanh Giap Dang, Mr. Anh Tuan Mac, and M.Sc Dong Cong Trinh

Institute for Control Engineering and Automation,

Hanoi University of Science and Technology,

 Dai Co Viet Street, Hanoi, Vietnam.

Author Response File: Author Response.pdf

Reviewer 4 Report

The section Results and Discussion is too short. I ask the authors to reflect and add further considerations and content to support their findings.

Comments for author File: Comments.pdf

Author Response

Authors’ responses to the Reviewers’ comments
Manuscript number: agriengineering-2763623
Title: Alpha-EIOU-YOLOv8: An improved algorithm for rice leaf diseases detection
Authors: Dong Cong Trinh, Anh Tuan Mac, Khanh Giap Dang, Huong Thanh Nguyen, Hoc Thai Nguyen and Thanh Dang Bui

January 10, 2024

Dear M.Sc. Leanne Fan and the Reviewers,

We sincerely thank and appreciate you and the reviewers for useful comments and constructive criticisms. Enclosed please find our responses to you and the reviewers of manuscript agriengineering-2763623. “Alpha-EIOU-YOLOv8: An improved algorithm for rice leaf diseases detection”. We have comprehensively addressed all of the comments provided by you and the reviewers and incorporated all of them in the revised paper.

Please find the detailed responses below and the corresponding revisions/corrections highlighted in the re-submitted files.

We sincerely thank you for all the insightful and constructively criticisms and suggestions. We belive that they have greatly strengthened the paper.

Thank you again for taking the time to review this paper.

Yours Faithfully,

A/Prof. Thanh Bui Dang, Dr Hoc Thai Nguyen, Dr Huong Thanh Nguyen, Mr. Khanh Giap Dang, Mr. Anh Tuan Mac, and M.Sc Dong Cong Trinh

Institute for Control Engineering and Automation,

Hanoi University of Science and Technology,

 Dai Co Viet Street, Hanoi, Vietnam.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Authors have addressed all the points in good manner.

Author Response

 Authors’ responses to the Reviewers’ comments
Manuscript number: agriengineering-2763623
Title: Alpha-EIOU-YOLOv8: An improved algorithm for rice leaf diseases detection
Authors: Dong Cong Trinh, Anh Tuan Mac, Khanh Giap Dang, Huong Thanh Nguyen, Hoc Thai Nguyen and Thanh Dang Bui

January 19, 2024

Dear the  Reviewer,

We greatly appreciate your precious time in reviewing our manuscript and providing us with constructive criticisms and suggestions. We believe that they have greatly strengthened the paper. We have comprehensively addressed all of the comments by you and incorporated all of them in the revised paper.

Thank you again for taking time to review this paper.

Yours Faithfully,

A/Prof. Thanh Bui Dang, Dr. Hoc Thai Nguyen, Dr. Huong Thanh Nguyen, Mr. Khanh Giap Dang, Mr. Anh Tuan Mac, and M.Sc. Dong Cong Trinh

Institute for Control Engineering and Automation,

Hanoi University of Science and Technology,

 Dai Co Viet Street, Hanoi, Vietnam

Author Response File: Author Response.pdf

Reviewer 3 Report

The authors considered my comments and suggestions. Good luck.

A final proofread will be useful.

Author Response

Authors’ responses to the Reviewers’ comments
Manuscript number: agriengineering-2763623
Title: Alpha-EIOU-YOLOv8: An improved algorithm for rice leaf diseases detection
Authors: Dong Cong Trinh, Anh Tuan Mac, Khanh Giap Dang, Huong Thanh Nguyen, Hoc Thai Nguyen and Thanh Dang Bui

January 19, 2024

Dear the  Reviewer,

We greatly appreciate your precious time in reviewing our manuscript and providing us with constructive criticisms and suggestions. Enclosed please find our responses to you and the reviewers of manuscript agriengineering-2763623: “Alpha-EIOU-YOLOv8: An improved algorithm for rice leaf diseases detection”. We have comprehensively addressed all of your comments and incorporated all of them in the revised paper. Our responses to your comments are listed below and marked in blue following each specific comment.

Comments 1: A final proofread will be useful.

Response 1: We have re-read and checked our article very carefully and we have corrected some errors as follows:

Lines 284-285: Alpha IoU to alpha-IoU;

Lines 301, 321, 323: alpha-EioU to alpha-EIoU;

Line 351: F1-Score to F1-score;

Line 370: Vietnnam to Vietnam;

Line 394: Our proposed system reduces the human resources and the time needed for rice diseases detection, therefore it can easily diagnostic the diseases at the early stage;

Lines 404 -407: In further extending the number of classes, the new diseases might contain relatively small disease marks, this could be a challenge, however, in this research, our predictions accuracy on the brown spot disease and the leaf blast disease characterized by the smallest disease marks, is relatively good, which bodes well for future improvements.

We sincerely thank you for all the insightful and constructive criticisms and suggestions. We believe that they have greatly strengthened the paper.

Thank you again for taking the time to review this paper.

Yours Faithfully

A/Prof. Thanh Bui Dang, Dr. Hoc Thai Nguyen, Dr. Huong Thanh Nguyen, Mr. Khanh Giap Dang, Mr. Anh Tuan Mac, and M.Sc. Dong Cong Trinh

Institute for Control Engineering and Automation,

Hanoi University of Science and Technology,

 Dai Co Viet Street, Hanoi, Vietnam

Author Response File: Author Response.pdf

Reviewer 4 Report

Compared to the previous version in which I pointed out as the main critical issue the brevity of the Discussions section the article is significantly improved in all aspects. The results obtained are now well supported and highlighted. I congratulate the authors for the excellent work done on the changes. I have indicated a few minor corrections to be made.

Comments for author File: Comments.pdf

Author Response

Authors’ responses to the Reviewers’ comments
Manuscript number: agriengineering-2763623
Title: Alpha-EIOU-YOLOv8: An improved algorithm for rice leaf diseases detection
Authors: Dong Cong Trinh, Anh Tuan Mac, Khanh Giap Dang, Huong Thanh Nguyen, Hoc Thai Nguyen and Thanh Dang Bui

January 19, 2024

Dear the  Reviewer,

We greatly appreciate your precious time in reviewing our manuscript and providing us with constructive criticisms and suggestions. Enclosed please find our responses to you and the reviewers of manuscript agriengineering-2763623: “Alpha-EIOU-YOLOv8: An improved algorithm for rice leaf diseases detection”. We have comprehensively addressed all of your comments and incorporated all of them in the revised paper. Our responses to your comments are listed below and marked in blue following each specific comment.

Comments 1: Compared to the previous version in which I pointed out as the main critical issue the brevity of the Discussions section the article is significantly improved in all aspects. The results obtained are now well supported and highlighted. I congratulate the authors for the excellent work done on the changes. I have indicated a few minor corrections to be made.

This paper needs minor revision.

Specific comments:

Line 370: “…Vietnam.”.

Line 394: “Our proposed system reduces the human resources and to short the time needed for rice disease…”.

Lines 406-407: “…brown spot disease and the leaf blast disease characterized by are two types of diseases that have the smallest disease marks, is relatively good, which bodes well for future improvements.”

Response 1: Authors would like to thank the reviewer for insightful comments. Extended clarification and modification of the comments have now been included in the revised paper.

Line 370: Vietnnam to Vietnam;

Line 394: Our proposed system reduces the human resources and the time needed for rice diseases detection, therefore it can easily diagnostic the diseases at the early stage;

Lines 404 - 407: In further extending the number of classes, the new diseases might contain relatively small disease marks, this could be a challenge, however, in this research, our predictions accuracy on the brown spot disease and the leaf blast disease characterized by the smallest disease marks, is relatively good, which bodes well for future improvements.

We sincerely thank you for all the insightful and constructive criticisms and suggestions. We believe that they have greatly strengthened the paper.

Thank you again for taking the time to review this paper.

Yours Faithfully

A/Prof. Thanh Bui Dang, Dr. Hoc Thai Nguyen, Dr. Huong Thanh Nguyen, Mr. Khanh Giap Dang, Mr. Anh Tuan Mac, and M.Sc. Dong Cong Trinh

Institute for Control Engineering and Automation,

Hanoi University of Science and Technology,

 Dai Co Viet Street, Hanoi, Vietnam

Author Response File: Author Response.pdf