Image Information Contribution Evaluation for Plant Diseases Classification via Inter-Class Similarity

Round 1

Reviewer 1 Report

General reviews,

The paper presented a new method for classification images of plant diseases with inter-class similarity. The proposed method has been compared with ref [13,14] for uncertainty, and ref. [15,16] for diversity. The main idea is to solve the problem of insufficient data for plant disease identification. 

Body of paper structure comments,

- Introduction was strongly presented to discuss the research background and point of interest. 

- Section 2, the authors provided the materials. However, if impossible the authors can give more details about the tomato leaf disease problem so that the reader can know the background of plants currently.

- Section 3, the authors presented the methods with two models such as inter-class similarity and image information contribution. Please delate full stop in Figure2, in lines 129, 134 "Where"  change to be "where". 

-Section 4, how many percentages of training data and test data when the authors train the model?

-Section 5, In Figure3 the authors should write the reference [..] of the traditional method in legends both uncertainty and diversity method, while the proposed in the red line should write as "proposed" represented "ours",  What is "ran"?.  Subsections 5.1 and 5.2 can be merged.

-Section 6 conclusion is good.

- Reference Section, please the authors should be rechecked with MDPI format, such as journal, proceeding, book, and so on.         

Author Response

Dear reviewer,

 

Thanks for your kind suggestions. Please see the attachment.

 

Thank you very much for your attention and consideration.

 

Sincerely,

Yue Yang

yue.yang@tju.edu.cn

Author Response File: Author Response.pdf

Reviewer 2 Report

The overall manuscript is well written, but the authors must address a few more points before further steps.

Figure 2 should be more apparent from the visual point of view.

Figure 4 also has the same visual issue. It should be a bit bigger and clear.

Values given in figure 5 are also not readable. It must be improved.

References would be consistent according to the style preferred by the journal.

The discussion section is very brief. It must be elaborated by communicating the previous findings in this field with the present findings of current research work.

Line 174: The word "in this paper" should be replaced by "in the present research".

 

 

 

 

Author Response

Dear reviewer,

 

Thanks for your kind suggestions. Please see the attachment.

 

Thank you very much for your attention and consideration.

 

Sincerely,

Yue Yang

yue.yang@tju.edu.cn

Author Response File: Author Response.pdf

Reviewer 3 Report

Sensing is a new field of research that enters our lives. The authors made a lot of effort but did not discuss many aspects. I believe they should refer to similar publications such as https://doi.org/10.3390/rs14051281 The article is about a digital elevation model (DEM) is an essential element of input data in the model research of watersheds. The study, gradually and with various methods, carried out a great simplification of a detailed LiDAR-derived DEM. Then, the impact of that treatment on the precision of the selected elements for modeling a watershed was assessed. The simplification comprised a reduction in resolution, with the use of statistical resampling methods, namely giving an average, modal, median, minimum, maximum, or the closest value to the pixels. 

Additionally, there is no in-depth statistical analysis. Statistical analysis was not sufficiently discussed. The entire manuscript looks like an excerpt from a report from some major work. Without statistics and research, the article does not contribute to any knowledge development. An expansion of literature, conclusions and discussions is required.

Author Response

Dear reviewer,

 

Thanks for your kind suggestions. Please see the attachment.

 

Thank you very much for your attention and consideration.

 

Sincerely,

Yue Yang

yue.yang@tju.edu.cn

Author Response File: Author Response.pdf

Reviewer 4 Report

This research article paper proposes an image information contribution evaluation method based on the analysis of inter-class similarity. This paper is well organized and the descriptions have well flowed. The introduction is well explained. To improve the quality of the paper, the authors need to consider the following items for revision.

Major comments:

1.      It is suggested that the author revises English writing.

2.      In the abstract, it is advised to add some major findings from the review and some key drawbacks/scope remain for further study.

3.      Keywords can be revised more specific and technical.

4.      Introduction part was well explained with literature. Moreover, the author mentioned “fine-grained” classification and “inter-class similarity” for the research steps. It would be clearer if the author provide a brief description of these two points in the introduction.

5.      More detailed depth in the method is needed in the method section. Need to explain more details about the feature extraction network and training network with current datasets.

6.      It is needed to explain the “High information” contribution and the “Low information” contribution for the images. How author differentiates between high and low information?

7.      In the discussion, the author mainly highlighted the similarity evaluation results between classes calculated from the initially labeled dataset and the test results of the models which are trained on the initially labeled dataset. Is it needed to discuss the model performance changes with the datasets and discuss in-depth the proposed method.

8.      It is recommended to add major findings of this article in the conclusion.

Minor comments

1.      Keywords can be improved.

 

2.      Figures quality can be improved. Please enhance the image quality. 

Author Response

Dear reviewer,

 

Thanks for your kind suggestions. Please see the attachment.

 

Thank you very much for your attention and consideration.

 

Sincerely,

Yue Yang

yue.yang@tju.edu.cn

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Give the authors to change all the references by MDPI format,

for example,

[1] D.A.; K.R. “Geostatistical resampling of lidar-derived dem in wide resolution range for modelling in swat: A case study of 290 zgłowi ˛aczka river (poland),” Remote Sensing, vol. 14, no. 5, p. 1281, 2022.

Correctly as

[1] Sliwinski, D.; Konieczna, A.; Roman, K. Geostatistical resampling of LiDAR-derived DEM in wide resolution range for modelling in SWAT: A case study of Zgłowi ˛aczka River (Poland), Remote sens. 2022, 14, 1-27.

 

Example of MDPI style,

[1]   Author 1, A.B.; Author 2, C.D. Title of the article. Abbreviated Journal Name Year, Volume, page range.

[2]  Author 1, A.B.; Author 2, C.D.; Author 3, E.F. Title of Presentation. In Proceedings of the Name of the Conference, Location of Conference, Country, Date of Conference (Day Month Year).

 

 

In the letter of response, the authors have revised, following the previous comments, that it was great. However, if the author rechecks the reference format again in MDPI style that the paper will be accepted after minor revision. 

 

Author Response

Dear reviewer:

 

Thank you for your encouragement and patient guidance. In this version, we have adjusted the reference format. Thank you very much for your helpful suggestions again.

 

Sincerely,

Yue Yang

Reviewer 2 Report

The manuscript is improved much. Few spelling checks are recommended.

Author Response

Dear reviewer:

 

Thanks for your encouragement and good suggestions again.

 

Sincerely,

Yue Yang

Reviewer 3 Report

Unfortunately, the article has not been corrected following the reviewer's comments. There are many unfinished points throughout the manuscript.

The manuscript has been significantly modernized, but the substantive part still does not convey any significant message. The manuscript does not have any advanced statistical analysis and correlation.

The discussion was not conducted in accordance with the guidelines of the journal. An expansion of literature, conclusions and discussions is required.

In this form, I advise against publishing the article in this journal.

 

English needs a lot of improvement.

Author Response

Dear reviewer:

 

Thank you for giving us a second chance to revise. In this revision, we mainly made the following modifications. Please see the attachment.

 

Sincerely,

Yue Yang

Author Response File: Author Response.pdf

Reviewer 4 Report

Some spelling checks, repeated words, and similar words/phrases like "in the present research" can be avoided or removed.

Author Response

Dear reviewer:

 

Thanks for your encouragement. In this version, we deleted some duplicate words and similar words / phrases and checked the spelling of the manuscript. Thank you very much for your good suggestions again.

 

Sincerely,

Yue Yang

Round 3

Reviewer 3 Report

Manuscript was correct according to the reviewer requirements. A lot of important suggestion was include.