Deep Learning Sensor Fusion in Plant Water Stress Assessment: A Comprehensive Review
Round 1
Reviewer 1 Report
The authors have presented a paper in which they review the current solutions that use deep learning for plant stress monitoring. The authors include in their survey a considerable number of papers and discuss the current challenges and future trends, as expected in a good survey. In general terms, the paper is well-written and well organized. There are only a few issues that must be solved and are mentioned below:
- The abstract must be extensively improved. The authors have to describe the problem in just two sentences and the focus on the aim of the paper and on their results. In the current version of the abstract, authors use do not highlight their results. The authors should include the major findings as the future prospects together with the challenges of the application domain in the abstract itself.
- For the keywords, authors must select words that are not included in the title.
- A section describing in detail the methodology followed by the authors to collect the papers included in their review must be added. In this section, the authors can detail the platform used to search the papers, the keywords used, the number of papers collected, and the number of papers used. In addition, any constrain, for example, in terms of year of publication that have been included in the search, must be mentioned.
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
Reviewer 2 Report
The review paper titled “Deep learning sensor fusion in plant water stress assessment: A comprehensive review” aims to catalogue studies related to deep learning applications in plant water stress domain. I would add few constructive comments to further improvements in this important subjected paper.
Introduction: This part has been written nicely as authors briefly covered all aspects of water stress and its quantification.
Section 2 heading is misleading. In this section the authors describe the major differences between machine learning and deep learning as no sensor fusion deep learning approach has been discussed. It should be modified as per discussion in this section.
Section 3 seems to be disconnected from the theme of the paper. This section may change into more generic and categorized into soil moisture estimation DL models, soil moisture modeling DL models and vision based DL models to aid water stress identification. In current form, this section does not fit here.
Section 4 is very short and require more information as this seems to be incomplete review for all the sub sections. The authors should focus more on comprehensive review on the satellite based soil moisture estimation as very brief details have been provided. Similarly, the evapotranspiration review is incomplete as many recent articles are missing from this section.
This review survey may have potential for publication after major improvements. In current form, this is not suitable for publication.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 2 Report
The authors cleared all my concerns. The paper is in good form now.
