Sweet Pepper Leaf Area Estimation Using Semantic 3D Point Clouds Based on Semantic Segmentation Neural Network
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsIn this work, the authors estimate the leaf area by combining the semantic segmentation effect image with the RGB-D image to generate a 3D point cloud map. Although this paper has
some innovative behavior, it still has some shortcomings.
1. In introduction, it is hoped that the authors will add related literature on evaluating leaf area based on deep learning.
2. Please standardize your writing. Remove redundant sentences from the manuscript, such as Line 78.
3. It is suggested that the author revise the title of 2.1, which is mostly about the dataset.
4. In the footnote of Figure 3, the means of data augmentation corresponding to each picture should be added for easy reading.
5. What are the red and green lines in the 3D point cloud images in the manuscript, and what are their meanings, such as Figures 4、7
6. Line 123:”the right camera is positioned at (0, 0, 0), the left camera is placed at (0, 0, d),” but as you can see from Figure 5, the left and right cameras should vary on the abscax.
7. Line 132: “To improve the integration of point clouds, we increase their resolution before
merging them.“ How do you do that?
8. Line 151: “ 400mm2 ” should be modified to “ 400mm2 ”
9. Which type of the three experiments do the 21 sampling points in Table 1 belong to?
10. In the experiment done in this paper, the true area of the leaf is not given. What does "error" in Table 2 mean? How is this value calculated ?
11. What are the evaluation metrics used in the experiments in the manuscript?
Author Response
Thank you very much for your comments.
The answer file is attached.
Author Response File: Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for Authors1. Authors are advised to change the sentence in abstract “In agriculture, estimating leaf area is essential for crop management. It helps farmers determine the health and growth of crops, which can inform decisions about irrigation, fertilization, and pest control. It also aids in assessing crop yield potential”. The statement is controversy.
2. The novelty projection in the manuscript is lagging
3. Results and discussion needs further improvements.
4. More literature is required to support the proposed methodology.
5. There is no mathematical analysis present in the manuscript. It is required to have more analysis.
6. It is hard to understand the process flow (Image Processing).
7. Methodology should be rewritten with clear flow.
8. The manuscript looks like very basic technology was discussed. Unable to find research content.
9. More recent references are required to support the present research.
Comments on the Quality of English LanguageModerate editing of English language required
Author Response
Thank you very much for your comments.
The answer file is attached.
Please, see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsThe paper presents a novel method for estimating the leaf area of sweet paper plants using semantic 3d point clouds based on semantic segmentation neural network. The method can be applied to automatic estimation of leaf area of other types of plants with proper training dataset.
The authors shall also clarify briefly on how the method experimented on a single isolated plant is applicable to the real field scenario where there will be other plants in the background of the image.
Author Response
Thank you very much for your comments.
The answer file is attached.
Please, see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsThank you for the improvement work you have made on all the issues.
Reviewer 2 Report
Comments and Suggestions for AuthorsNow authors have made good number of changes in the manuscript which can be acceptable.