Identification Method of Corn Leaf Disease Based on Improved Mobilenetv3 Model

Round 1

Reviewer 1 Report

Clearly written and  well structured  article with  clear conclusions. 

Author Response

Dear Reviewers:

Thank you for your letter and for the reviewers’ comments concerning our manuscript entitled “Identification Method of Corn Leaf Disease Based on Improved Mobilenetv3 Model” (Manuscript ID: agronomy-2035577). Thank you for your approval and suggestions on this article. I have checked the spelling of the article. We have revised the spelling of lines 14, 21, 23, 41, 128-130, 133-138, 278, 306, 329-330, 355-356,372-373,393-395,408,428-429,434,438 in the paper. The revisions to the manuscript have been marked up using the “Track Changes” function and marked with yellow.

Reviewer 2 Report

The manuscript proposed an Improved Mobilenet model for corn leaf disease identification. The manuscript is well written however, the following are my concerns that must be addressed. 

Abstract:

What is CLDD?

I felt the authors have directly jumped into the details of the MobileNet model. The abstract should give a background, a problem statement, the objectives, and the results. The technical components are good but I feel the background, problem statement, and objectives are missing in the abstract. Considering this comment rewrite the abstract section. 

Introduction:

In the introduction, why CNN, and why pretrained / transfer learning models are used? The answer to such questions should be answered. 

Since the major contribution is towards enhancing the mobilenet model I think there should be subsection in the introduction that introduces the original MobileNet model and its functions like cross-entropy loss function, SE module, and so on. Only then the contribution can be justified. 

Materials and Method:

L.no 115 says one health leaf. I think its a typo

L.no 117 says 7757 pieces of corn leaf health. It should be "healthy corn leafs" isn't it?

In section 2.3, I feel first the author should discuss topics 2.3.1 to 2.3.4 then the proposed architecture or modifications done to the model ie., from L.no 142 to 196 should come. Consider this flow. 

What is meant by Top1 accuracy?

Which visualization module is used for heat map visualization? It should be cited. 

Conclusion is only presenting the results. Authors should adhere to the standard conclusion standard for the scientific paper. Future work is also missing. 

The following recent reference can be cited:

Andrew J, Eunice J, Popescu DE, Chowdary MK, Hemanth J. Deep Learning-Based Leaf Disease Detection in Crops Using Images for Agricultural Applications. Agronomy (2022) 12: doi: 10.3390/AGRONOMY12102395

 

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 3 Report

Authored approach in specific area driven by image recognition. Well written .

Author Response

Dear Reviewers:

Thank you for your letter and for the reviewers’ comments concerning our manuscript entitled “Identification Method of Corn Leaf Disease Based on Improved Mobilenetv3 Model” (Manuscript ID: agronomy-2035577). Thank you for your approval and suggestions on this article. I have checked the spelling of the article. We have revised the spelling of lines 14, 21, 23, 41, 128-130, 133-138, 278, 306, 329-330, 355-356,372-373,393-395,408,428-429,434,438 in the paper. The revisions to the manuscript have been marked up using the “Track Changes” function and marked with yellow.

Round 2

Reviewer 2 Report

I appreciate the authors for effectively addressing my concerns. The manuscript can now be in acceptable form. 

The following checks should be done by the authors during the final submission.

- Check lines 49-50 in the track changed version. Rewrite for clarity. 

- MobilenetV3 citation is missing at its first instance. 

- Language editing and proofreading. 

 

 

Author Response

Please see the attachment

Author Response File: Author Response.docx