Remote River Velocity Measurement via Optical Flow
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for measuring stream flow velocity, such as current meters and acoustic Doppler meters, are expensive, time-consuming, and require on-site presence, making them impractical for extreme flood conditions or areas with heavy debris.
Innovation Solution
A system and method using image data from a camera, potentially coupled to a drone, which transmits data to a remote computer system for processing with a trained machine learning algorithm implementing an optical flow algorithm to estimate stream flow velocity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional current meters or acoustic Doppler meters are used to measure stream flow velocity, then measurement accuracy is improved, but on-site presence is required which becomes impossible or impractical during heavy floods
Solution Approach 1:
The patent uses video images as a copy or representation of the actual water flow, allowing remote measurement without physical presence in the stream. The optical flow algorithm processes these visual copies to extract velocity information, eliminating the need for operators to be on-site during hazardous flood conditions while maintaining measurement accuracy
Solution Approach 2:
The patent replaces mechanical measurement devices (current meters, acoustic Doppler meters) with an optical-based system using video cameras and image processing algorithms. This substitution allows velocity measurement through visual analysis rather than mechanical or acoustic methods, enabling remote operation during extreme flood events
2Measurement precision
If conventional measurement techniques are used during extreme floods, then velocity measurement is obtained, but the process becomes time-consuming and expensive
Solution Approach 1:
The system enables continuous velocity measurement by processing video frames in real-time or near-real-time. Multiple consecutive frames are analyzed to track particle displacement over time, providing continuous velocity data throughout the flood event rather than discrete spot measurements, thereby reducing total measurement time while maintaining accuracy
Solution Approach 2:
The optical flow algorithm automatically tracks water surface particles and computes velocity fields without requiring manual intervention for each measurement point. The system self-adjusts by identifying natural tracer particles in the water and computing their displacement, eliminating the need for external seeding or manual measurement at multiple locations
3Ease of operation
If particle tracking methods are used to measure stream velocity remotely, then on-site presence is eliminated, but particles can be lost and real-time analysis is not achieved
Solution Approach 1:
Instead of tracking a single particle that may be lost, the optical flow algorithm tracks numerous water surface particles simultaneously across the entire video field. This excessive tracking of multiple particles ensures that even if some particles are lost or become untrackable, sufficient data remains to compute reliable velocity fields, thereby improving tracking reliability
Solution Approach 2:
The system uses feedback from consecutive video frames to continuously update particle positions and velocities. By comparing each frame with the previous frame and integrating displacement information over time, the algorithm maintains reliable tracking even when individual particles temporarily disappear, using feedback from surrounding particles and temporal continuity
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach provides a fast, non-invasive, and remote method to measure stream flow velocity, overcoming the limitations of conventional techniques by enabling accurate measurements during extreme floods without the need for external seeding or on-site presence.
Implementation Method 1
the image data are input to a trained machine learning algorithm stored on the computer system, generating output as water flow velocity data, wherein the trained machine learning algorithm implements an optical flow algorithm
Data Source
AI summary
Described here are systems and methods that utilize visual imagery and an optical flow-based computer vision algorithm to measure river velocity in streams or other flowing bodies of water. The systems and methods described in the present disclosure overcome the barriers of conventional flow measurement techniques by providing a fast, non-intrusive, remote method to measure peak flows.