Source and magnitude of error in an inexpensive image-based water level measurement system

Abstract

Recent technological advances have opened the possibility to use webcams and images as part of the environmental monitoring arsenal. The potential sources and magnitude of uncertainties inherent to an image-based water level measurement system are evaluated in an experimental design in the laboratory. Sources of error investigated include image resolution, lighting effects, perspective, lens distortion and water meniscus. Image resolution and meniscus were found to weigh the most in the overall uncertainty of this system. Image distortion, although largely taken into account by the software developed, may also significantly add to uncertainty. Results suggest that ''flat'' images with little distortion are preferable. After correction for the water meniscus, images captured with a camera (12 mm or 16 mm focal lengths) positioned 4-7 m from the water level edge have the potential to yield water level measurements within ±3 mm when using this technique.

Water Level Measurements from Drones: A Pilot Case Study at a Dam Site

“…Moreover, unmanned aerial vehicles can improve crowdsourced data collection, supporting citizen observatories, lowering data uncertainty, and filling in the gap due to the irregular distribution of engaged volunteers. In agreement with the analysis performed by Gilmore et al [52], the major sources of uncertainty are found to be the perspective and the pixelization, while the scale of the experiment allowed us to get rid of the effect of meniscuses formed by water at the contact with the background. It is worth underlining the fact that the water surface waviness can substantially affect the water level estimates, as waves ripple the water, causing a blurry water level edge.…”

“…The combination of the waves and of both angle and intensity of the incoming light source (e.g., sun, clouds) creates the sharp change in pixel gray scale in an image. Gilmore [52] investigated different sources of uncertainty affecting water level estimates from images. In addition to the uncertainties associated to the local environment as the change in lighting, other sources of uncertainty are associated with the image quality, such as image focus, image resolution, perspective, and lens distortion.…”

Watershed services in the humid tropics: Opportunities from recent advances in ecohydrology

“…For example, low-cost water quality sensors have been developed and tested for parameters such as pH and conductivity, temperature, toxicity, and turbidity (Banna et al, 2014;Chapin, Todd, & Zeigler, 2014;Murphy et al, 2015;Tuna, Arkoc, & Gulez, 2013;Yagur-Kroll et al, 2015), although few sensors have actually been deployed in the field. Offthe-shelf cameras have also been applied successfully to record water level (Gilmore, Birgand, & Chapman, 2013) and discharge (Bradley, Kruger, Meselhe, & Muste, 2002;Tsubaki, Fujita, & Tsutsumi, 2011), plant phenology (Crimmins & Crimmins, 2008;Nijland et al, 2014), and cloud cover (Scholl, 2015). To compile data, wireless sensor networks can also be used to provide connected and sometimes real-time data on a range of environmental parameters within an area (Kido et al, 2008;Zia, Harris, Merrett, Rivers, & Coles, 2013).…”

Monitoring ephemeral, intermittent and perennial streamflow: a dataset from 182 sites in the Attert catchment, Luxembourg

“…Bradley et al, 2002;Muste et al, 2008;Tsubaki et al, 2015), stage (e.g. Shin et al, 2007;Royem et al, 2012;Gilmore et al, 2013;Schoener, 2017) or discharge (Lüthi et al, 2014) found its way into hydrological applications. Streamflow measurements using particle image velocimetry (PIV) require video imagery and have been used in combination with artificial and natural tracers (Bradley et al, 2002;Creutin et al, 2003;Muste et al, 2008).…”

“…Evaluation of measured heights from the camera system with stage heights from the United States Geological Survey (USGS) showed a good agreement with a relative difference of 16 %. Gilmore et al (2013) identified image resolution, lighting effects, perspective, lens distortion and the water meniscus as sources of error. They found image resolution and the meniscus contributing most to errors in detected water level, while the influence of lens distortion largely depends on the consideration of the distortion in the software.…”