Using kites for 3-D mapping of gullies at decimetre-resolution over several square kilometres: a case study on the Kamech catchment, Tunisia

Feurer, Denis; Planchon, Olivier; El Maaoui, Mohamed Amine; Ben Slimane, Abir; Boussema, Mohamed Rached; Pierrot-Deseilligny, Marc; Raclot, Damien

doi:https://doi.org/10.5194/nhess-18-1567-2018

Articles | Volume 18, issue 6

https://doi.org/10.5194/nhess-18-1567-2018

© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 3.0 License.

Special issue:

The use of remotely piloted aircraft systems (RPAS) in monitoring...

https://doi.org/10.5194/nhess-18-1567-2018

© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 3.0 License.

Articles | Volume 18, issue 6

Research article

|

07 Jun 2018

Research article |

| 07 Jun 2018

Using kites for 3-D mapping of gullies at decimetre-resolution over several square kilometres: a case study on the Kamech catchment, Tunisia

Denis Feurer, Olivier Planchon, Mohamed Amine El Maaoui, Abir Ben Slimane, Mohamed Rached Boussema, Marc Pierrot-Deseilligny, and Damien Raclot

Download

Final revised paper (published on 07 Jun 2018)
Preprint (discussion started on 08 Feb 2017)

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

RC1: 'Review of "Potential of kite-borne photogrammetry for decimetric and kilometre square 3D mapping: an application for automatic gully detection"', Mitchell Bryson, 02 Mar 2017
- AC1: 'Reply to RC1', Denis Feurer, 14 Jun 2017
RC2: 'This paper deals with the acquisition and application of very high resolution topographic data through a kite using photogrammetric methods on a very local area where it is difficult to get authorizations to fly for UAV. The methodology is well described.', Anonymous Referee #2, 11 May 2017
- AC2: 'Reply to RC2', Denis Feurer, 14 Jun 2017

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

ED: Reconsider after major revisions (further review by Editor and Referees) (23 Jun 2017) by Daniele Giordan

AR by Denis Feurer on behalf of the Authors (06 Jul 2017) Author's response Manuscript

ED: Referee Nomination & Report Request started (09 Jul 2017) by Daniele Giordan

RR by Anonymous Referee #3 (25 Sep 2017)

Suggestions for revision or reasons for rejection

Although this is a potentially interesting contribution to SfM methodologies and gully delineation using kite platforms, there are many things to improve in the aims, structure and writing. I attached a more detailed review on the pdf file.
- Aims: to me the objective of this manuscript has not been well defined. In too many occasions, the authors assume that they are making a more novel contribution than they might be making. Kite photogrammetry is quite an old approach, and its potencial (see title) has been well established, similarly also to SfM and gully delineation. The authors should try to rethink and justify which part of this work provides new insights and provides a meaningful contribution. I found the references quite incomplete sometimes, especially regarding gully detection algorithms. Why are you offering a new method? Please review more critically previous works and the need for a new method. In addition, please specify clearly what are the objectives and develop the methods accordingly.
- Structure and writing: This is my main concern. The objectives are not clearly defined, the actual perfomed experiments ill-explained and frequently methods, introduction and results are mixed up. To me this careless writing has been irritating. The tone of the language is also very informal many times, it seems to be written for a report rather than for a scientific journal. I also believe (not native english speaker) that the writing needs a major revision and I doubt this work has been edited by a proffesional english reviewer.
I recommend major revision to give the authors an opportunity to build a better focused manuscript after undegoing major improvements to achieve the standards of a journal like HESS.

Referee Report: PDF

Hide

RR by Anonymous Referee #4 (06 Oct 2017)

Suggestions for revision or reasons for rejection

General Comments:

This manuscript explores the potential for using kites as aerial platforms to generate digital elevation models (DEM) using structure from motion. The authors subsequently attempt to detect gully heads and channels based on a DEM analysis. The main focus of this manuscript appears to be on the technique of image acquisition, with a lesser focus on the structure from motion DEM generation and the gully analysis.

Generally, I think this manuscript details an interesting study. The process of using a kite to collect photos is interesting, and the authors have clearly thought through some of the important details to make this type of research successful.

The application of gully head/channel analysis comes across as an after-thought though, and it is not clear how well their approach achieves the mapping. If extracting gullies were the main goal of the paper, I would expect a stronger analysis of their results shown in Figure 8. Currently, they spend very little time analyzing the gully extraction results. How much of the area identified as a gully channel was a false-positive? To my eye, in Figure 8 everything that is a road or a building is mapped as a gully channel. I would encourage the authors to quantitatively show how much of the area that is mapped as a gully is actually a gully. Now how much of figure 8 that is mapped as a gully is something else? Moreover, they state in the abstract that they are going to map gully heads, but I did not see any data indicating this was done.

As it is currently presented, this is really a “methods” paper, and I think if the authors wanted to stop there, they would be making a fine contribution to a journal that is focused on methods. But they have come very close to making a useful contribution to researchers interested in gullies, so I would encourage the authors to dig more deeply into this. Try to go beyond just highlighting the areas that are incised. Try to identify the gully heads. See if you can add some criteria like slope or drainage area to filter out roads and buildings where there are no gullies. The DEM differencing using the Gaussian filter approach is interesting, but it isn’t really a breakthrough that merits publication. If the authors added a few more steps to really show that they can identify gullies, I think it would make this paper much more useful to the many researchers who study gullies.

I am including many line-by-line comments in the attached pdf.

Referee Report: PDF

Hide

ED: Reconsider after major revisions (further review by editor and referees) (04 Dec 2017) by Daniele Giordan

AR by Denis Feurer on behalf of the Authors (23 Feb 2018) Author's response Manuscript

ED: Referee Nomination & Report Request started (12 Mar 2018) by Daniele Giordan

RR by Anonymous Referee #4 (29 Mar 2018)

RR by Anonymous Referee #5 (13 Apr 2018)

Suggestions for revision or reasons for rejection

This is a nicely written paper, with interesting modeling and mapping results. However, the text suffers from some smaller technical inaccuracies while it could also use extra references. Below, I give my list of comments. Once these are taken into account, I can accept this paper.

• Abstract and text: What is a decimetre DEM? Do you mean positional accuracy of 1 decimetre, spatial resolution of 1 decimetre? Please use terminology and explanations that are clear;
• Page 1 line 10: you do not explain your acronym RPAS (in your abstract, yes, but the abstract and main text have to be considered two spate texts). Also, RPAs is never explained. For me, and many others in the community, drone, UAS or RPAS refers to a motorised unmanned platform (this definition is given in multiple books). As such, a kite is not an RPAS.
• Page 1 line 25 “Structure from motion (SfM) and multi-view stereo algorithms, recent developments in photogrammetry, represent new sources of very” No, these algorithms are not the sources of the DEMs, they enable the computation of these DEMs. So write “Structure from motion (SfM) and multi-view stereo algorithms, recent developments in photogrammetry, represent new means of computing very high…”;
• Page 1 line 27 and following: SfM DEMs is a misnomer. If you do an SfM DEM, you just end up with a sparse point cloud. At least put SfM+MVS DEM; This is a mistake that is present all over the paper. SfM does not generate dens e3D geometry. It is an algorithm to orient (interior and exterior) camera sand gives – as a by-product – a sparse point cloud. One need a dense matcher to create a dense 3D surface geometry;
• Page 6 lien 10 and following as well as table 7. The term “weight” is mentioned, but the authors mean “mass”. Mass and weight are not identical and the units given here indicate mass, not weight.
• Page 8 line 13. Before, you wrote that MicMac is a dense matcher, so it implements an MVS approach. On line 13, you write “..is typical of SfM algorithms”. This is thus wrong, as – again – SfM is not identical to MVS and one can easily have one without the other. So, please make sure you use the correct terminology.
• Page 13 point 3.2. For exactly this reasons (dealing with insufficient or too much wind for a given kite), people have been using Helikite’s for aerial mapping (see reference in the point below), because you combine the best of the balloon and kite platform into one platform;
• Page 18 line 22. That is not entirely true. In “Verhoeven, G., Loenders, J., Vermeulen, F., Docter, R., 2009. Helikite aerial photography - a versatile means of unmanned, radio controlled, low-altitude aerial archaeology. Archaeological Prospection 16 (2), 125–138.”, the authors discuss for halve a page the importance of line sag, stretch, dimeter and mass of the tether, while also advising a Dyneema line for exactly those reasons. Also check the journal “The Aerial Eye” for many articles specifically on these technical kite issues;
• Page 19 line 23. Another paper that evaluated the error of a DEM obtained from aerial images over a large area using a tethered platform is “Verhoeven, G., Taelman, D., Vermeulen, F., 2012. Computer vision-based orthophoto mapping of complex archaeological sites: the ancient quarry of Pitaranha (Portugal-Spain). Archaeometry 54 (6), 1114–1129.” This is also one of the first papers that made the DEM using a tethered platform and an SfM+MVS pipeline;
• Page 21 line 29: With only SfM, you will not get far.
• Appendix: Security issues and the importance of wearing Kevlar handgloves is also discussed in the Helikite paper mentioned before.

Hide

ED: Publish subject to minor revisions (review by editor) (20 Apr 2018) by Daniele Giordan

AR by Denis Feurer on behalf of the Authors (27 Apr 2018) Author's response Manuscript

ED: Publish subject to technical corrections (08 May 2018) by Daniele Giordan

AR by Denis Feurer on behalf of the Authors (09 May 2018) Manuscript

Short summary

We present a method for acquiring very-high-resolution images for 3-D mapping of gullies over kilometre-square areas using kites. Kites used in appropriate conditions can be an advantageous alternative to light unmanned aircraft when local regulations or weather conditions hamper their use. We proved that kites can acquire images, allowing for high-quality 3-D coverage of large areas. We automatically detected and mapped gullies from a decimetre kite DEM with 74 % accuracy of the length.