Reviewer Comments:
There are a couple of minor grammatical issues:
“… which utilizes seven colors of light emitting diodes (LED; 450, 521, 593, 625, 660, 730, and 870 nm) for acquiring multispectral images and a longpass filter (> 665 nm) to create mask image based on chlorophyll fluorescence.”
… to create a mask image …
“Differences chlorophyll levels between low and sufficient fertilizer treatments were evident from the average NDVI value of the plants and easily visualized in false color NDVI images (Figure 1).”
Differences in chlorophyll levels …
In general, I think this short paper is easy to read and understand. The authors present a low-cost system for multispectral imaging using monochrome cameras and LEDs. They show significant correlations between their method and other approaches/measurements.
The methods seem to be appropriate for the objective and the imaging system is reasonably well documented.
There is no data or code availability statement, so I am unable to reproduce any results or examine code.
We are planning to write a manuscript with the data used in this proceeding. Therefore, we do not want to jeopardize our later manuscript by the proceeding.
I think that that introduction could be expanded. I was surprised that the abstract is nearly 100 words longer than the introduction. I think that the intro could benefit from more of a ”…what do we know? What do we want to know? here we show…” format. I do not have a sense of the state-of-the-art in stationary multi-spectral phenotyping.
We edited the abstract and introduction to make clear what we want to present.
There is no description of the experiment design for the fertilizer experiment in the methods. In the results, there is a figure (Fig 1) but very little description of the finds of this experiment.
Again, we do not want to share the detail at this moment for later publication. On the later publication, we will share all the details including the codes
I might try to reframe the results and discussion as (1) the system, (2) the ”calibration” experiments, (3) the fertilizer case study.
We changed structure of it as you suggested.
The conclusions did not leave me with a sense of the contribution to the state-of-the-art. It is well-known that images are sources of data and that R-Pi computers are useful tools. What does this study demonstrate and why should the readers be interested in this work?
We rewrote the conclusion as you suggested.
The proceeding is well written and presents a cost-effective imaging platform that will be of interest to the community. Given the focus of the proceeding it would be good to include links to the code that was used. The proceeding highlights cost-effective technologies that can be used to create accurate imaging and image analysis pipelines.
In general, we shorten some unnecessary sentences and rephrased points we would like to present, especially in introduction and conclusion.