Remote Sensing of River Velocity Using Drone Video and Optical Flow Algorithm

Jyoti, Jamir Shariar; Medeiros, Henry; Sebo, Spencer; McDonald, Walter

doi:10.1061/9780784483060.018

Cited by 1 publication

(2 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Traditional RSV measurements usually rely on propeller velocimeters and acoustic doppler current profilers [3]- [6], which are based on contact measurements and can only retrieve RSV of a relatively small area. As these methods are both of limited efficiency and dangerous to operate under extreme conditions, image-based velocimetry methods, such as large-scale particle image velocimetry (LSPIV) [7]- [9], space-time image velocimetry (STIV) [10]- [12], and optical flow velocimetry (OFV) [13]- [15] have been proposed as substitution for efficient and convenient RSV estimation. But when it comes to practical application, LSPIV-based methods depend on tracers for good performance, and STIVbased methods are sensitive to normal optical noises and inapplicable to river with vortices and turbulence, whereas OFV-based methods show low requirement of tracers [16] and produced pixel-wise flow field.…”

Section: Introductionmentioning

confidence: 99%

“…OFV-based methods utilize optical flow estimation for pixel-level displacement estimation and their accuracy mainly depends on the accuracy of the optical flow estimation model. As the optical flow estimation methods based on deep learning perform better than the traditional ones and are free of burdensome hyper-parameter settings, many research groups have employed them and have reported satisfactory results [14], [15], [17].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

River Surface Velocimetry Based on Virtual River Dataset and Modulated-RAFT

Zhang

Cao

et al. 2023

IEEE Access

View full text Add to dashboard Cite

Many image-based velocimetry methods, such as large-scale particle image velocimetry (LSPIV), space-time image velocimetry (STIV) and optical flow velocimetry (OFV), have been proposed to estimate river surface velocity (RSV) efficiently and precisely. Among these methods, OFV-based methods have drawn great attention due to its high field resolution and low requirement of tracers. As a powerful approach for estimate optical flow accurately and efficiently, deep optical flow estimation is utilized in OFV-based methods as well. However, these methods often use irrelevant datasets for training due to the immeasurability of optical flow. Models obtained by such approaches suffer from limited generalization because of domain drift. Besides, high similarity of river surfaces often leads to ambiguous correlation volume extracted by deep optical flow estimation model, which can cause mismatch. This work proposed a method for accurately and robustly RSV estimation under velocity range of 0-6.0 m/s. Specifically, we introduced a method for generating relative optical flow datasets and proposed MRAFT, a deep optical flow estimation model combining with correlation volume modulation. Experiment results demonstrate that the generated virtual river datasets effectively improve the generalization of the model and MRAFT remarkably alleviates the mismatch. Our work facilitates the application of deep optical flow on RSV estimation and provides other related researches with optical flow datasets for fine-tuning.

show abstract