River Water Level Detection Using Graph-Cut Image Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing water level measurement methods, particularly those using water gauges and image processing, face challenges such as high installation and maintenance costs, difficulty in observing fine water level variations, and reduced accuracy due to environmental factors and reflection issues, which can lead to delayed evacuation instructions.
Innovation Solution
A water level measurement system utilizing a flow processing unit to acquire and calculate motion information, a graph-cut processing unit to specify the water surface region, and a water level calculating unit to determine the water level based on the boundary between the water surface and other regions, along with an edge extraction processing unit to enhance measurement accuracy and a setting unit to adjust the camera angle, all integrated into an image processing technique.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If water gauges are installed to measure water level with high accuracy, then measurement precision is improved, but installation cost and maintenance cost increase
Solution Approach 1:
The patent replaces the mechanical water gauge system with an image processing system using a monitoring camera. The camera captures images of the river, and image processing algorithms automatically detect the water level by identifying the boundary between water and non-water regions. This substitution eliminates the need for physical water gauges, repeaters, and associated mechanical components, thereby reducing installation and maintenance costs while maintaining measurement capability.
Solution Approach 2:
The patent uses optical copying by capturing visual information of the water level through a camera. Instead of directly measuring physical pressure or depth with gauges, the system creates an optical copy (image) of the water level scene and processes it to extract measurement data. This approach allows non-contact measurement and reduces the need for physical infrastructure.
2Device complexity
If a small number of water gauges is installed, then installation cost is reduced, but the ability to observe fine water level variation between upstream and downstream deteriorates
Solution Approach 1:
The monitoring camera serves multiple functions: it monitors the overall river scene for safety purposes and simultaneously enables precise water level measurement through image processing. This multi-functionality allows the system to observe fine water level variations along the river without requiring multiple dedicated measurement devices, as a single camera can capture the entire river section in its field of view.
3Device complexity
If image processing is used to measure water level with existing monitoring camera, then installation cost is reduced, but measurement accuracy deteriorates due to reflection and environmental factors
Solution Approach 1:
The patent applies local quality by focusing the image processing on specific regions of interest within the camera field of view. The system identifies and processes the boundary region between water and non-water areas with higher detail and attention, while other regions are processed with standard algorithms. This localized focus improves measurement accuracy by concentrating computational resources on the critical measurement zone.
Solution Approach 2:
The patent employs parameter changes by adjusting image processing parameters such as threshold values, color space conversions, and boundary detection sensitivity based on environmental conditions. The system dynamically modifies processing parameters to account for variations in lighting, reflection, and weather conditions, thereby maintaining measurement accuracy across different environments despite using an existing monitoring camera.
4Measurement precision
If expert setting of angle of view is required for high accuracy image processing, then measurement precision is improved, but ease of operation deteriorates due to increased working processes
Solution Approach 1:
The patent implements self-service by enabling the system to automatically determine and adjust the optimal angle of view and processing parameters without requiring expert intervention. The image processing algorithm autonomously analyzes the captured images, identifies the river geometry and water level boundaries, and configures itself for accurate measurement. This automation eliminates the need for manual expert setup while maintaining high measurement precision.
5Device complexity
If water level is displayed only as measurement value, then device complexity is reduced, but ease of operation deteriorates due to difficulty in visual judgment
Solution Approach 1:
The patent merges the water level measurement information with the visual image of the river. Instead of displaying separate numerical values, the system overlays the measured water level data directly onto the corresponding location in the camera image. This integration allows observers to simultaneously view the visual state of the river and the quantitative measurement, enabling rapid and accurate visual judgment of dangerous situations without requiring complex display systems.
Data Source
AI summary
In a technology for detecting the water level of a river by image processing, after an angle of view setting process, an area setting unit sets an arbitrarily defined certain range from the center of the set angle of view as a processing area. Then, a flow processing unit calculates, from the processing area, motion information and a flow direction, computes a flow density, determines a region having a high density and similar flow directions to be a flow of water currents, and deletes flows in the other directions. Thereafter, a graph-cut processing unit creates an object seed and a background seed for graph-cutting, and detects a water surface by automatic graph-cutting. After an edge extraction processing unit has performed edge extraction, a water level calculation processing unit determines an edge that satisfies a predetermined condition to be a water level line, and outputs a water level measurement result.