Measurement of riverbed material grainsizes is now a routine part of fieldwork in fluvial geomorphology and lotic ecology. In the last decade, several authors have proposed remote sensing approaches of grain size measurements based on terrestrial and aerial imagery. Given the current rise of small Unmanned Aerial System (sUAS) applications in geomorphology, there is now increasing interest in the application of these remotely sensed grain size mapping methods to sUAS imagery. However, success in this area has been limited due to two fundamental problems: lack of constraint of image scale for sUAS imagery and blurring effects in sUAS images and resulting orthomosaics. In this work, we solve the former by showing that SfM-photogrammetry can be used in a direct georeferencing (DG) workflow (i.e. with no ground validation) in order to predict image scale within margins of 3%. We then propose a novel approach of robotic photosieving of dry exposed riverbed grains that relies on near-ground images acquired from a low-cost sUAS and which does not require the presence of ground control points or visible scale objects. We demonstrate that this absence of scale objects does not affect photosieving outputs thus resulting in a low-cost and efficient sampling method for surficial grains.

No abstract

An estimated 76% of global stream area is occupied by channels with widths above 30m. Sentinel-2 imagery with resolutions of 10m could supply information about the composition of river corridors at national and global scales. Fuzzy classification models that infer sub-pixel composition could further be used to compensate for small channel widths imaged at 10m of spatial resolution. A major challenge to this approach is the acquisition of suitable training data useable in machine learning models that can predict land-cover type information from image radiance values. In this contribution, we present a method which combines unmanned aerial vehicles (UAVs) and Sentinel-2 imagery in order to develop a fuzzy classification approach capable of large-scale investigations. Our approach uses hyperspatial UAV imagery in order to derive high-resolution class information that can be used to train fuzzy classification models for Sentinel-2 data where all bands are super-resolved to a spatial resolution of 10m. We use a multi-temporal UAV dataset covering an area of 5.25km 2. Using a novel convolutional neural network (CNN) classifier, we predict sub-pixel membership for Sentinel-2 pixels in the fluvial corridor as divided into classes of water, vegetation and dry sediment. Our CNN model can predict fuzzy class memberships with median errors from À5% to +3% and mean absolute errors from 10% to 20%. We also show that our CNN fuzzy predictor can be used to predict crisp classes with accuracies from 95.5% to 99.9%. Finally, we use an example to show how a fuzzy CNN model trained with localized UAV data can be applied to longer channel reaches and detect new vegetation growth. We therefore argue that the novel use of UAVs as field validation tools for freely available satellite data can bridge the scale gap between local and regional fluvial studies.

Speed of spoligotyping could be a benefit in the clinical setting.