Rapidly Quantifying Interior Greenery Using 360° Panoramic Images

Round 1

Reviewer 1 Report

The manuscript reports an interesting methodological approach of acquiring greenery in interior spaces. The approach is based on 360 ° panoramas. While I generally think that this is method has potential and consists of a lot of creativity, there are a couple of points I would like to mention for a revision round:

• In section 3.2, you report a strong correlation between manually and automatically estimated amounts of greenery. Could you be more transparent about the manual extraction method? How did you map the data? Who did it? And do you see any limitations of this approach which might have an effect on the results oft he correlation?

 

• You refer your results to an improvement of human wellbeing (see section 5). Could you be a bit more precise in how far the amount of green could have a positive effect for human life? And how does your approach deal with the problem that different plant species (reflecting similar wavelengths in yellow and green) might have different effects on wellbeing due to different characteristics? Wouldn’t it make sense to consider additional sensors leaving the field of optical wavelengths, such as near infrared? In this way, vitality of plants could better be tracked and analyzed.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

The manuscript "Rapidly quantifying interior greenery using 360° panoramic images" is submitted to Remote Sensing as a full manuscript to Forests (MDPI). The subject of the research presented in the manuscript contains 11 pages and describes an approach to detect green surfaces in different indoor environments. Extracting interior greenery allows estimating human wellbeing in working places and also at residences that account for the majority of the daily usual lifetime.

The practical applications are useful information for ranking the environment quality of different places that indirectly could be used to relate to health issues of residents, employers, and collaborators. I found it particularly useful to have such devices and tools, but I feel that the proposed application is still fragile and the manuscript requires therefore improvements before publication.

My major concern is the need to better describe the procedures of converting epipolar geometry from spherical images for better computing the perimeter and also area. Cutting the edges is still not enough/sufficient for making a better surface estimation. The information could be accounted for in different environments by measuring the edges of a given wall inside the room. Supposing that a given wall of a living room has 3 meters in height, and 5 meters long, then the estimated surface should be 15 square meters. This could be a calibration/validation parameter, besides having some artificial targets (control in geometry and radiometry).

Another major point is the number of samples used in the investigation. There are too many values up to 5 shown in Figure 4 whereas few values from 5 to 35. The X- and Y-axis are also different which makes 1:1 comparison difficult. Finally, the discussion section is too short and poor without bringing existing methods available as Apps for common electronic devices such as tablets, smartphones, etc. that are also able to capture panoramic images and extract targets. In summary, the scripts used as well the software, processing time, and other procedures shall be clearly described and make available (e.g. sections 2.2 and 2.3).

Other comments

Abstract: please show some more quantitative results for the different scenarios and eventual issues with artificial green surfaces;

Introduction: it is important to mention working places and relate your research indirectly with those mapping trees and green areas in public places (check for literature);

L165-169: add some sentences preparing the reader for the next subtopics;

L171: please present some numbers about targets and linear features as mentioned above to show the accuracy of the proposed method to retrieve areas and perimeters;

L188: figure 4 shall improve font size and show 1 by 1 line. The scale and amplitude of both X and Y axis must be the same for better comparison;

L193: explain in the text whether the differences are found in small, intermediate, or large environments;

Table 3: please add a note saying that the positive value means;

Figure6: describe better the axis labels and figure caption;

L233: comment on this and also perspectives for this detection (other sensors?);

L250: please enhance discussion based on future perspectives. In general, for several countries, the use of plants for decoration and even small gardens at home is increasing. How does quantifying such elements benefits society/human being? I believe authors could add some very nice sentences about;

L257: bring some more quantitative analysis;

Thank you for allowing me to evaluate this research application.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The authors submitted a revised version of the manuscript. In their response letter, they refer to the points raised in review round no. 1. The argumentation in the letter (justifying the changes) is sound. Accordingly, I would recommend to accept this manuscript for publication.

Reviewer 2 Report

The majority of the comments were properly addressed in the revised version of the manuscript. I would ask authors to increase their figures when possible. In summary, the paper extended now the number of pages now by detailing the results and discussion sections. Additionally, new figures and references higlighted the content of the research. Therefore, I am recommending the revised version for publication now.