An Efficient Processing Approach for Colored Point Cloud-Based High-Throughput Seedling Phenotyping
Round 1
Reviewer 1 Report
This is an interesting paper that uses RGB-D sensors for the monitoring of plant growth and health. The proposed method is quite straightforward. Although the proposed method is not significant in its novelty, the proposed workflow is quite sound. I think this is a good paper, please consider the following minor concerns that may help to improve this paper.
- The compared method is published in 2013 (which I think is relatively old), is there other new method on this topic to be compared? Otherwise, the term "state-of-the-art" in the abstract and the texts may not be justified.
- Can we design some experiments to directly measure the precision of heights or other metrics of the growth and health conditions of the proposed methods? Rather than only the segmentation results. Because in my experiences the precision of RGB-D sensor (the precision of the depth measurement) is at several centinmeters. Is it good enough?
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
Reviewer 2 Report
As a general comment, the experiment is described with sufficient details on the applied methodologies and algorithms. My concern is related to the lack of comparison with other existing methods mentioned in the introduction. This is a big lack for the proposed method which aims to offer an operational alternative.
Moreover, no analysis has been conducted on the applicability of different point resolutions to provide differences regarding the resolution and accuracy of the final results.
Finally, the research seems to be focused mainly on the segmentation techniques and it appears poor of data on the height and leaf area index estimation. The latter two parameters are crucial for seedling monitoring as stated by the authors in the introduction. Only for the cucumber, the estimation of the height and leaf area index has been provided. Why no estimation of these important parameters has been conducted for the other seedlings in the experiment?
For a method that aims to be alternative to the existing ones, the research should have provided more examples of estimation of these two parameters for the other types of seedlings involved in the work.
Also for cucumber, the discussion of the result appears not enough in deep.
For this reason, I regret but I think that the research needs supplementary work and it should be rejected.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 3 Report
The paper is interesting, describing an innovative, low cost alternative for plant phenotyping.
The research is adequately designed, however some questions arise:
- what is the performance of other systems in the same setup
- Up to what age can seedlings be monitored using the described techniques?
- Is the system adaptive to various types of soil, in terms of texture and color?
- Can accurate measurements be performed on anthocyanin rich species, such as red basil?
Please check grammar, as in line 222, "raise the robust".
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 2 Report
Dear Authors, thank you so much for your clarifications and improvements.
The paper has been significantly improved, providing much more details for the comparison with other research. Also, details about applied methods and result accuracy have been provided.
Lack of information in the previous version had led me to reject the paper.
Now, authors have demonstrated that the work has been done with a scientific approach, collecting many valuable data which has been organised correctly.
In my opinion, thanks to the effective improvement achieved, the paper deserves to be published.
Author Response
We apologized for the inadequacy of our previous manuscript. It is our honor for our revised manuscript to receive your approval. Thanks again for your rigorous review.
