Elsevier

Computers & Geosciences

Volume 155, October 2021, 104866
Computers & Geosciences

Development of a three-axis accelerometer and large-scale particle image velocimetry (LSPIV) to enhance surface velocity measurements in rivers

https://doi.org/10.1016/j.cageo.2021.104866Get rights and content

Highlights

  • A three-axis accelerometer combined with LSPIV was developed.

  • Dual cameras were built to capture images for analyzing surface velocity.

  • LSPIV with a three-axis accelerometer improved the surface velocity measurements.

  • The effects of different factors on surface velocity measurements were investigated.

Abstract

A three-axis accelerometer was combined with large-scale particle image velocimetry (LSPIV) to obtain nonintrusive and safe surface velocity measurements. Dual cameras were established at the Yu-Feng gauging station of Shimen Reservoir to capture near-field and far-field images and analyze the surface velocity of rivers. Surface velocity measurements were obtained with a flow meter, LPSIV with ground control points, and LSPIV with a three-axis accelerometer to compare the measurement accuracy. The results show that the relative root mean square error (RMSE) values for LSPIV with ground control points and with the three-axis accelerometer are 21% and 18%, respectively. The LSPIV technique with the three-axis accelerometer slightly improved the measurement accuracy of the surface velocity. However a three-axis accelerometer can be utilized to replace the traditional ground control points for yielding the camera pose parameters. Furthermore, the effects of the camera pose, three-axis acceleration, interrogation area (IA), and image resolution on surface velocity measurements were explored. The camera pose parameters, namely, roll (θ) and tilt (τ), and three-axis acceleration parameters, including Xa and Ya, influence surface velocity measurements. If the measurement error of the surface velocity is controlled within ±10%, the acceptable variational ranges of θ, τ, Xa, and Ya are 6.2°, 1.3°, 0.11 g, and 0.02 g, respectively. The IA size and image resolution also significantly affect the accuracy of surface velocity measurements. Therefore, the selection of a suitable IA size and image resolution is crucial for accurately measuring surface velocity.

Introduction

The precise measurement of river flow velocity is crucial for hydroenvironmental flow, water resource management, sediment transport, flash flooding disaster mitigation, bank erosion control, and hydrologic and hydraulic engineering research (Fujita et al., 1998; Kantoush et al., 2011; Tauro, 2016; Yeh et al., 2017; Guillen et al., 2017; Tauro et al., 2018). However, the availability of and capability to obtain precise observations in river regions are limited due to the difficulty of accessing the measuring environment, which can be a high-risk area, especially during high-flow periods (Coata et al., 2006; Jodeau et al., 2008; Hrachowitz et al., 2013). Different kinds of intrusive measurement instruments, such as impellers, rotator velocimeters, acoustic Doppler velocimeters, and acoustic Doppler current profilers, provide good spatial and temporal resolutions but require expensive instrument maintenance (Hannah et al., 2011; Muste et al., 2011; Gharahjeh and Aydin, 2016). The success of non-intrusive approach is able to provide measurements that are not possible with the intrusive ones during floods. Therefore, the development of remote methods to overcome inconveniences in measuring surface velocities is an urgent requirement (Legleiter et al., 2017; LeGrand et al., 2020).

To enhance the available measurement techniques, state-of-the-art approaches to measure the water surface velocity and river discharge are needed. Among these techniques, large-scale particle image velocimetry (LSPIV) using fixed stations is a remote surface velocity measurement system that displays promising potential for measuring surface velocity in river segments via the nonintrusive and continuous capture of images of floating objects (Muste et al., 2014; Tauro et al., 2015; Ran et al., 2016; Li et al., 2019). LSPIV provides a safe and easy option for the measurement of surface velocities and flows at large scales during flash flood events. Advanced techniques are not only utilized to measure surface velocity in rivers but also applied to assess debris flow velocities, manage stormwater detention basins, and measure street-scale urban drainage and urban overland runoff velocities (Leitao et al., 2018; Theule et al., 2018; Naves et al., 2019; Zhu and Kouyi, 2019). However, LSPIV measurements are highly sensitive to the ground control points used for image calibration and orthorectification and the orthogonal images applied for further processing (Kim et al., 2008; Meselhe et al., 2004; Hauet et al., 2008).

In recent years, the nonintrusive LSPIV methods have been widely applied to measure surface velocities at field sites under low/normal flow conditions (Sun et al., 2010; Bechle et al., 2012; Gunawan et al., 2012; Tauro et al., 2014, 2017, 2019) and flash flood conditions (Jodeau et al., 2008; Le Coz et al., 2010; Dramais et al., 2011; Fujita and Kunita, 2011; Ran et al., 2016; Le Boursicaud et al., 2016; Tauro et al., 2016; Guillen et al., 2017; Huang et al., 2018). However, establishing ground control points is an indispensable step before using LSPIV for measuring surface velocity in both low/normal and flash flood flow conditions. Ground control points are typically established based on the stations set up along the banks of rivers for LSPIV measurements. The current experimental approaches used to access the banks of rivers expose researchers to unpredictable risk. Additionally, establishing ground control points is laborious and requires expensive experimental instruments. However, the control point issue was already introduced in Tauro et al. (2014) where lasers were suggested in order to estimate the pixel dimensional step.

To overcome these disadvantages of using ground control points to obtain the poses of cameras, the objective of this study is to develop a three-axis accelerometer with LSPIV for precisely measuring the surface velocity of rivers and for potentially replacing the traditional ground control points to yield camera pose parameters. To determine the accuracy of the measured surface velocity, field measurements of the cross-sectional surface velocity obtained using a flow meter in a river. The measured surface velocity using LSPIV with ground control points was compared with that using LSPIV with a three-axis accelerometer. Furthermore, the effects of the camera pose, interrogation area (IA), and image resolution on the surface velocity measurements were explored and analyzed using LSPIV techniques with a three-axis accelerometer.

Section snippets

Study site

The study site, the Yufeng gauging station, is situated in the upstream catchment area of the Shimen Reservoir, which is located in the northern part of Taiwan (Fig. 1a). Shimen Reservoir is one of the most important reservoirs in Taiwan; it is located on the Dahan River, a tributary of the upper reaches of the Tamsui River. The reservoir was completed in 1964. The catchment area is 763.4 km2, and the total storage capacity is 309 million m3. The Yufeng gauging station is located at Yufeng

Results

In the comparison of cross-sectional surface velocities, three time frames were selected on October 22, 2019, November 29, 2019, and March 15, 2020, and comparisons of the observed surface velocities and LSPIV results were made. Fig. 9 illustrates the comparison of the measured surface velocities using the flow meter measurements, the LSPIV values with ground control points, and the LSPIV values with the three-axis accelerometer along the river cross-section. In the figure, the X-axis equals at

Discussion

To explore the effects of the camera pose using three-axis acceleration, IA, and image resolution on the measured results of the surface velocity, the data obtained via LSPIV with the three-axis accelerometer on November 29, 2019 were used as a baseline. In each run, one parameter/variable was changed, and the surface velocity results were analyzed.

Conclusions

A three-axis accelerometer with LSPIV was developed to measure the surface velocity in rivers and applied at the Yufeng gauging station, which is located in the upstream section of Shihmen Reservoir in northern Taiwan. Dual cameras were employed to capture far-field and near-field images and measure the surface velocity in the river. The surface velocity was measured using a flow meter and two approaches, including LSPIV with ground control points and LSPIV with a three-axis accelerometer. The

Computer code availability

The code was developed by Wen-Cheng Liu and Wei-Che Huang. The contact address is the National United University, Miaoli Taiwan. The e-mail is [email protected] and [email protected]. The code is written in Matlab environment. Running the code requires the Matlab version 9.7. The users can download the code freely at https://github.com/e11856824/LSPIV.

Author statement

W.C. Liu: Conceptualization, Investigation, Resources, Writing – original draft, reviewing, editing, Supervision, Project administration, and Funding acquisition; W.C. Huang: Methodology, Software, Validation, Formal analysis, correction of experimental data, and Visualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This study was supported by the Ministry of Science and Technology (MOST), Taiwan, under grant nos. 107-2119-M-239-002 and 108-2119-M-239-001. The financial support is greatly appreciated. The authors would like to express their appreciation to associate editor and two anonymous reviewers who provided useful comments to improve this article.

References (42)

  • W. Li et al.

    Stereo-imaging LSPIV (SI-LSPIV) for 3D water surface reconstruction and discharge measurement in mountain river flows

    J. Hydrol.

    (2019)
  • M. Muste et al.

    Capabilities of large-scale particle image velocimetry to characterize shallow free-surface flows

    Adv. Water Resour.

    (2014)
  • M. Muste et al.

    Consideration on direct stream flow measurements using video imagery: outlook and research needs

    J. Hydro-environ. Res.

    (2011)
  • J. Naves et al.

    Using a 2D shallow water model to assess Large-Scale Particle Image Velocimetry (LSPIV) and Structure from Motion (SfM) techniques in a street-scale urban drainage physical model

    J. Hydrol. (Wellingt. North)

    (2019)
  • F. Tauro et al.

    PTV-Stream: a simplified particle tracking velocimetry framework for stream surface velocity monitoring

    Catena

    (2019)
  • A. Bechle et al.

    Development and application of an automated river-estuary discharge imaging system

    J. Hydraul. Eng.

    (2012)
  • J. Coata et al.

    Use of radars to monitor stream discharge by noncontact method

    Water Resour. Res.

    (2006)
  • I. Fujita et al.

    Large-scale particle image velocimetry for flow analysis in hydraulic engineering applications

    J. Hydraul. Res.

    (1998)
  • N.F. Guillen et al.

    Use of LSPIV in assessing urban flash flood vulnerability

    Nat. Hazards

    (2017)
  • D.M. Hannah et al.

    Large-scale river flow archives: important, current status and future needs

    Hydrol. Process.

    (2011)
  • A. Hauet et al.

    Experiment system for real-time discharge estimation using an imagine-based method

    J. Hydrol. Eng.

    (2008)
  • Cited by (8)

    View all citing articles on Scopus
    View full text