Camera-Based Pointer Recognition Motion Blur Compensation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing pointer recognition systems for large format touch screens are costly and complex, relying on sophisticated image sensors and additional hardware components, which is not feasible for inexpensive but reliable alternatives.
Innovation Solution
An interactive system with cameras and a processing structure that compensates for motion blur by generating a mean intensity profile and determining pointer type using models like erf and Butterworth, disentangling point-spread function and real pointer widths, and associating digital ink attributes, with infrared pass filters and LEDs for improved accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If sophisticated image sensors and additional hardware components are used, then pointer recognition reliability is improved, but system cost and complexity increase
Solution Approach 1:
The system performs preliminary actions by capturing multiple video frames before final pointer type determination. The processing structure analyzes a sequence of frames to compensate for motion blur and accurately identify pointer characteristics, ensuring reliable recognition without requiring complex hardware components
Solution Approach 2:
The patent uses optical copying through camera-based image capture to represent the physical pointer on the touch surface. By capturing visual information and processing it algorithmically, the system achieves reliable pointer identification without needing complex sensing hardware or additional hardware components
2Measurement precision
If motion blur compensation is applied, then pointer type determination accuracy is improved, but processing time increases
Solution Approach 1:
The system performs preliminary frame capture and background image generation before actual pointer recognition. By pre-processing the touch surface environment and having multiple reference frames ready, the system can quickly compensate for motion blur during actual pointer type determination without significant processing delays
Solution Approach 2:
The processing structure uses feedback from multiple video frames to refine pointer type determination. By continuously analyzing frame sequences and comparing against background images, the system accurately compensates for motion blur while maintaining efficient processing through iterative refinement rather than exhaustive analysis
3Measurement precision
If point-spread function width is disentangled from real pointer width, then measurement precision is improved, but computational complexity increases
Solution Approach 1:
The patent creates a computational model that copies the point-spread function characteristics from the camera system and separates them from the actual pointer dimensions. By using mathematical modeling to represent the PSF as a distinct element, the system can disentangle it from the real pointer width measurement without requiring complex hardware or excessive computational resources
Solution Approach 2:
The system changes parameters by introducing separate variables for PSF width and real pointer width in the mathematical model. By parameterizing the image formation process with distinct terms for optical blur and object dimensions, the system achieves precise measurements through algorithmic separation rather than complex hardware differentiation
Data Source
AI summary
The present invention relates to a method and system of improved pointer recognition. A camera-based interactive system has two cameras with fields of view observing an interactive surface. A processor receives a plurality of video frames from the cameras and recognizes a pointer within the video frames. Motion blur of the pointer is compensated to determine the pointer type from the compensated frames using a model-based approach.