GamaNNet: A Novel Plant Pathologist-Level CNN Architecture for Intelligent Diagnosis
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThis research is meaningful. It can improve the monitoring of diseases and pests during the tomato planting process, reduce manual operations, and realize automatic and modern tracking. The difficulty of monitoring lies in the similarity between some diseases and the different severity of each type of disease, which will affect the accuracy. However, this paper only considers the efficiency of the diagnosis of diseases at various times and cannot apply to the natural situation in the field. There are some small problems in this paper:
Line 210-211: The results of this article are based on the diagnostic efficiency of 10 non-simultaneous plant quarantine problems, and 10 are too few to represent all diseases.
Lines 417-449: The authors can add to the convolutional neural network to discuss the analysis of plant leaves and plant morphology in different periods of the early, middle, and late disease.
Lines 449-502: The author should consider the simultaneous occurrence of two to three pests and diseases.
Comments on the Quality of English LanguageThis research is meaningful. It can improve the monitoring of diseases and pests during the tomato planting process, reduce manual operations, and realize automatic and modern tracking. The difficulty of monitoring lies in the similarity between some diseases and the different severity of each type of disease, which will affect the accuracy. However, this paper only considers the efficiency of the diagnosis of diseases at various times and cannot apply to the natural situation in the field. There are some small problems in this paper:
Line 210-211: The results of this article are based on the diagnostic efficiency of 10 non-simultaneous plant quarantine problems, and 10 are too few to represent all diseases.
Lines 417-449: The authors can add to the convolutional neural network to discuss the analysis of plant leaves and plant morphology in different periods of the early, middle, and late disease.
Lines 449-502: The author should consider the simultaneous occurrence of two to three pests and diseases.
Author Response
Esteemed Peer,
Based on all of your considerations, we would like to express our sincere gratitude for your valuable suggestions and points of enhancement for the readers of this Special Issue. Minor edits have been conducted in the manuscript to ensure it meets the formal and academic standards appropriate for research papers.
In seeking your approval, we have respectfully provided detailed answers and explanations regarding the manuscript. Please feel free to review the attached file.
Sincerely,
Adunias Teixeira
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsThe authors propose a novel plant pathology-grade CNN architecture for intelligent diagnosis. Before the paper is published, I recommend the following revisions:
1. The proposed GamaNNet is compared with current networks. The authors should indicate the prediction speed of GamaNNet under better performance metrics.
2. If feasible, the scale information of the images and each processing layer should be introduced in Figure 3, along with a detailed description of the proposed network.
3. The authors should further explain why GamaNNet outperforms current networks and highlight the key design points of their network.
4.The size and clarity of text in the image should be improved to ensure they are clearly visible when printed. For example, the axis labels for Figures 4, 5, 6, 7, and 9 are difficult to read.
5. Improve keyword precision to ensure efficient paper search.
6. It is recommended that authors compare performance metrics with current approaches to tabular format
Author Response
Esteemed Peer,
Based on all of your considerations, we would like to express our sincere gratitude for your valuable suggestions and points of enhancement for the readers of this Special Issue. Minor edits have been conducted in the manuscript to ensure it meets the formal and academic standards appropriate for research papers.
In seeking your approval, we have respectfully provided detailed answers and explanations regarding the manuscript. Please feel free to review the attached file.
Sincerely,
Adunias Teixeira
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsAuthors claimed that it has designed the GamaNNet, a new type neural network. However, I failed to see if there is any novelty here. The architect of the network is similar or identical to the old fashioned earlier network with a few CONV layers .
The network does not seem able to give a good accuracy. In the days, even Yolo would like to give a better accuracy on that.
I also failed to see why the network can adapt to without the need of resizing. Even this is a presentation problem where authors may forget to present, I cannot see the novelty and usefulness of the claimed contribution.
Comments on the Quality of English LanguageLanguage of writing is fine.
Author Response
Esteemed Peer,
Based on all of your considerations, we would like to express our sincere gratitude for your valuable suggestions and points of enhancement for the readers of this Special Issue. Minor edits have been conducted in the manuscript to ensure it meets the formal and academic standards appropriate for research papers.
In seeking your approval, we have respectfully provided detailed answers and explanations regarding the manuscript. Please feel free to review the attached file.
Sincerely,
Adunias Teixeira
Author Response File: Author Response.pdf
Reviewer 4 Report
Comments and Suggestions for AuthorsThe article is interesting but requires minor revision.
There is no clear information on which tier and in what phase of the growing season plant leaves were selected for analysis. Varietal characteristics are also important.
It is necessary to standardize all References and bring them to the journal's requirements. There are a lot of inconsistencies.
Author Response
Esteemed Peer,
Based on all of your considerations, we would like to express our sincere gratitude for your valuable suggestions and points of enhancement for the readers of this Special Issue. Minor edits have been conducted in the manuscript to ensure it meets the formal and academic standards appropriate for research papers.
In seeking your approval, we have respectfully provided detailed answers and explanations regarding the manuscript. Please feel free to review the attached file.
Sincerely,
Adunias Teixeira
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
Comments and Suggestions for AuthorsThe author answered my question.
