Rotating Object Abnormality Detection Using Virtual Stereo Displacement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in accurately detecting abnormalities in rotating objects like wind power propellers and turbine blades without increasing apparatus size, and they struggle with displacement measurement accuracy due to vibration noise and motion blur.
Innovation Solution
A state determination apparatus that includes a fixed image acquisition unit, a difference calculation unit, an out-of-plane displacement calculation unit, an in-plane displacement calculation unit, and an abnormality determination unit, which acquires fixed images of a rotating object, calculates displacement amounts, and determines abnormalities based on displacement distributions and temporal changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a stereo camera is used to capture moving images of rotating objects, then detection accuracy can be improved, but apparatus size increases
Solution Approach 1:
The patent uses a single camera to capture images and creates a virtual stereo effect through image processing. The image capturing apparatus captures images of the rotating object, and the processing unit calculates displacement amounts by comparing images taken at different times, effectively creating a virtual three-dimensional measurement system without requiring physical stereo cameras.
Solution Approach 2:
The patent replaces the mechanical stereo camera system with a single camera and computational algorithms. Instead of using two physical cameras to capture stereoscopic images, the system uses one camera and processes images over time to calculate displacements, substituting mechanical hardware with computational methods.
2Measurement precision
If displacement measurement is performed on rotating objects, then abnormality detection is enabled, but vibration noise degrades measurement accuracy
Solution Approach 1:
The patent performs preliminary actions by capturing multiple images at different times before final analysis. The image capturing apparatus continuously captures images of the rotating object, and the processing unit stores these images for later comparison. This preliminary data collection allows the system to separate true displacement signals from vibration noise through temporal analysis.
Solution Approach 2:
The patent implements feedback mechanisms where the processing unit continuously compares displacement calculations with the captured images. By feedback-looping the image capture and displacement calculation processes, the system can identify and filter out vibration noise patterns, improving measurement accuracy through iterative refinement.
3Productivity
If moving images are captured of rotating objects, then dynamic monitoring is enabled, but motion blur degrades measurement accuracy
Solution Approach 1:
The patent uses periodic action by capturing images at specific time intervals rather than continuously. The image capturing apparatus takes discrete images at predetermined times, allowing the processing unit to calculate displacements between these periodic frames. This periodic sampling enables dynamic monitoring while minimizing motion blur effects compared to continuous capture.
Data Source
AI summary
A state determination apparatus acquires each of time-series fixed images obtained by shooting a rotating object via a first optical device that rotates and emits incident light; calculates a displacement amount of a plurality of specific points on the object, and calculates a difference between the displacement amount of a reference specific point and the displacement amount of each specific point other than the reference specific point; calculates, as an out-of-plane displacement, a displacement of a surface of the object on an image plane following movement of the object in a normal direction of the image plane, using the difference; calculates, as an in-plane displacement, a displacement of the surface of the object on the image plane following movement of the surface of the object; and specifies an abnormality of the object, based on a distribution of the in-plane displacement or a temporal change of the out-of-plane displacement.