A Review of Plant Phenotypic Image Recognition Technology Based on Deep Learning

Round 1

Reviewer 1 Report

The review is devoted to the use of deep learning techniques to recognize the plant phenotypic image.

The manuscript is well structured, it contains all sections for this type of publication. The abstract briefly reflects the content of the review. However, to my mind, this review is too limited and it should be extended by adding information according to the presented below remarks.

1. To my best mind, table 1 contains a very limited number of current reviews in this subject area. I think, that it is necessary to extend this table by adding both the references and descriptions of the other researchers' works.
2. Figures 1,2,... have low quality in terms of both size and text representation. To my mind, it is necessary to reformat them.
3. Table 2 also does not present all list of current traditional classifiers. If you describe SVM and Decision Trees techniques, I think that it is necessary to describe the Random Forest technique too.
4. To my mind, the review will look better in the case of adding the section with a review of general techniques of quality evaluation of the described image recognition methods.

Author Response

Dear reviewer and editor,

      We have provided a point-by-point response to the reviewer’s comments, please see the attachment. Thank you for giving us the suggestions for the manuscript. We tried our best to improve the manuscript and made some changes in the manuscript. We appreciate for reviewers’ warm work earnestly and hope that the correction will meet with approval. We sincerely hope to have the opportunity to publish in Electronics Journal. Best wishes!

 

Yours sincerely,

Lei Shu

        

Author Response File: Author Response.pdf

Reviewer 2 Report

This paper review different approaches of plant phenotypic image recognition technologies. Authors illustrate the development and application of plant phenotypic image recognition technologies (classical and deep learning methods). In the paper, they discuss about three classical and four deep learning-based methods in the context of plant phenotypic image recognition. However, I have several concerns and comments to be addressed:

1. Authors discuss more about the basic concepts rather than the application on the specific field.

2. Inconsistency in writing ( e.g., in paper ..image plant phenotypic image recognition technology in title only image recognition)

3. Inconsistency in citations (e.g., line 39, 38 missing citations)

4. Unnecessary information and repetition (e.g., lines 133-140, 165 - 168)

5. Grammar can be improved  ( e.g., line 16 “ .. point of any..” to “..point for any..” 

6. The Authors mention four types of deep learning methods. Why the specific four types of models?

7. Some phrases can be abbreviated (e.g., plant phenotypic image recognition)

8. Low-quality figures ( figure 1, 2, 3), fonts different, resolution, scaling issues. Hard to understand ( e.g., figure 10)

9. Figures captions should be corrected and improved ( e.g., is this convolutional neural network structure diagram?)

10. In lines 294-295, the relation between GRU and vanishing and exploding gradient not clear to the reader.

11. In line 313, the authors mention “SAE is a type of unsupervised learning with strong characteristic learning ability”. It’s not clear to the reader ( in the context of this field). And the figure 12 does not clearly represent the unsupervised methodology.

12. Insufficient information about the previous literature reviews (table 1). 

13. Table 3 structure can be improved ( e.g., overlapping)

Author Response

Dear reviewer and editor,

      We have provided a point-by-point response to the reviewer’s comments, please see the attachment. Thank you for giving us the suggestions for the manuscript. We tried our best to improve the manuscript and made some changes in the manuscript. We appreciate for reviewers’ warm work earnestly and hope that the correction will meet with approval. We sincerely hope to have the opportunity to publish in Electronics Journal. Best wishes!

Yours sincerely,

Lei Shu

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

I have not any questions.

Author Response

Dear reviewer and editor,

        We appreciate for reviewers’ warm work earnestly. We sincerely hope to have the opportunity to publish in Electronics Journal. Best wishes!

 

                                                                                                    Yours sincerely,

                                                                                                     Lei Shu

Reviewer 2 Report

The Authors revised the papers based on the comments. However, the overall quality is not satisfactory. I have several more concerns and comments to be addressed:

1. Table 1, authors added some more related works but, the description of the previous works still not up to the level (i.e., row 1: just keywords from the [1], row 5, last sentence: we conclude ...  ). 

2. Previous works already discuss the feature extraction. Please discuss what  are the differences between the current review and the previous review (i.e., [1] discuss leaf feature extraction which is similar for plant feature extraction)

3. [6] already mention and discuss CNN based methods.  Please discuss what are the differences between the current review and the previous reviews.

4. [7] already mention and discuss traditional methods.  Please discuss what are the differences between the current review and the previous reviews.

5. Table 2 contains repeated information (i.e., from [6])

6. There are repeated figures (i.e., figure 6 and figure 7 represent the same things) 

7. Table 3, column 4, contains repeated information, Those can be mentioned in the text for less repetition. 

Author Response

Dear reviewer and editor,

       We have provided a point-by-point response to the reviewer’s comments, please see the attachment. Thank you for giving us the suggestions for the manuscript. We tried our best to improve the manuscript and made some changes in the manuscript. We appreciate for reviewers’ warm work earnestly and hope that the correction will meet with approval. We sincerely hope to have the opportunity to publish in Electronics Journal. Best wishes!

                                                                                                    Yours sincerely,

                                                                                                               Lei Shu

Author Response File: Author Response.pdf

Round 3

Reviewer 2 Report

The authors carefully answered the questions along with new explanations. More specifically, the authors updated the figures, tables. Additional details about the previous review papers and detailed analysis have been given for better understanding. Authors also removed some repeated information to improve the paper quality.