Wireless Charging ROI Imaging for False-Alarm Object Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wireless charging systems face challenges in accurately detecting foreign objects within the multi-dimensional region of interest due to false-alarm regions and excluded areas, leading to inefficiencies and safety concerns.
Innovation Solution
A method is introduced to set a multi-dimensional region of interest (ROI) based on the multi-dimensionality of the charging space, using a computing apparatus to collect and process plane images from a camera, adjusting the exclusion ratio, and determining the presence and location of surveillance targets by calculating rectangular coordinates based on distance and angle, thereby distinguishing essential and false-alarm regions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If a traditional three-dimensional ROI is set in the charging space, then the monitoring coverage is improved, but false-alarm regions and excluded areas cause detection accuracy to deteriorate
Solution Approach 1:
The patent divides the three-dimensional ROI into multiple two-dimensional plane images captured from different angles and positions. Each plane image is processed independently to detect surveillance targets, and results are integrated to achieve comprehensive monitoring while eliminating false-alarm regions that occur in traditional 3D ROI approaches.
Solution Approach 2:
The patent transforms the three-dimensional ROI monitoring problem into multiple two-dimensional plane image processing tasks. By capturing and analyzing plane images from different dimensions (multiple angles and positions), the system achieves accurate detection of surveillance targets while avoiding the false-alarm and excluded area problems inherent in traditional 3D ROI methods.
2Measurement precision
If the ROI is reset every time the power transmitting coil and power receiving coil are changed, then the monitoring accuracy is improved, but the system response time and productivity deteriorate
Solution Approach 1:
The patent pre-establishes multiple plane images corresponding to different coil configurations and positions before actual charging operations begin. When coil changes occur, the system can immediately switch to the pre-prepared corresponding plane images without requiring time-consuming ROI resetting, thus maintaining monitoring accuracy while improving system response time.
Solution Approach 2:
The patent creates a universal set of plane images that can be applied across different coil configurations. The same plane image processing methodology and surveillance target detection algorithms work universally for various power transmitting and receiving coil combinations, eliminating the need to reset ROI parameters for each specific coil pair.
3Reliability
If the ROI includes excluded areas to avoid false alarms, then false-alarm reduction is improved, but detection capability in those excluded areas deteriorates
Solution Approach 1:
The patent segments the monitoring space into multiple plane images from different angles, allowing each plane to have its own optimized ROI settings. Surveillance targets detected in one plane that would fall into excluded areas of another plane are still captured through alternative viewing angles, maintaining detection capability while reducing false alarms through multi-plane verification.
Solution Approach 2:
By transitioning from a single three-dimensional ROI with excluded areas to multiple two-dimensional plane images, the patent enables surveillance targets to be detected from alternative dimensional perspectives. A target located in an excluded area of one plane may be clearly visible in another plane, thereby maintaining detection capability while the multi-plane approach inherently reduces false alarms through cross-validation.
Data Source
AI summary
The present disclosure relates to an image processing method for a multi-dimensional region of interest (ROI) in a wireless charging system. More specifically, the method includes determining whether a surveillance target exists in the multi-dimensional ROI by performing, regarding the surveillance target, the process of removing a false-alarm region and including an excluded ROI.