Multi-Pointer Detection Using Vertical Intensity Profile Curve Fitting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Interactive input systems employing machine vision technology face challenges in accurately determining the locations of multiple pointers due to ambiguity and occlusion issues, where one pointer may obscure another, complicating the computation of (x, y) coordinates.
Innovation Solution
The method involves generating a vertical intensity profile (VIP) from captured image frames, analyzing peak locations, and fitting curves using nonlinear least squares algorithms to resolve peak locations, with the Akaike Information Criterion (AIC) used to determine the best model fit, thereby accurately registering pointer positions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple pointers are detected in captured image frames, then the system can support multi-touch interactions, but pointer ambiguity and occlusion make coordinate computation more complicated
Solution Approach 1:
The patent segments the complex problem of multiple pointer detection by processing each camera's image data independently to generate individual pointer position estimates, then combining these estimates through triangulation. This segmentation approach breaks down the complex multi-pointer problem into manageable single-pointer detection units that can be processed separately and then integrated.
Solution Approach 2:
The patent introduces an intermediary computational process that generates multiple possible pointer positions and uses probability calculations to resolve ambiguity. The system creates a set of candidate positions and uses statistical methods to identify the most likely actual pointer locations, serving as an intermediary between raw image data and final coordinate determination.
2Device complexity
If traditional triangulation methods are used for multiple pointers, then the system maintains simplicity, but measurement precision deteriorates due to pointer occlusion and ambiguity
Solution Approach 1:
The patent implements feedback by using detected pointer positions to inform subsequent detection iterations. The system refines pointer location estimates by comparing expected pointer positions with actual detections across multiple frames, adjusting calculations to account for occlusion and improve overall measurement precision.
Solution Approach 2:
The patent changes detection parameters dynamically based on detected conditions. When multiple pointers are detected in close proximity, the system adjusts its analysis parameters to focus on resolving the specific geometric relationships between pointers, changing the computational approach from standard triangulation to more sophisticated multi-point analysis.
3Measurement precision
If visual indicators are displayed to resolve pointer ambiguity, then measurement precision improves, but device complexity increases due to additional display and processing requirements
Solution Approach 1:
The patent creates visual copies or representations of the detected pointer positions and their associated uncertainty. By displaying multiple possible pointer locations as visual indicators, the system provides a graphical copy of the computational uncertainty, allowing users to see the range of possible interpretations and improving measurement precision through visual verification.
Data Source
AI summary
A method of determining locations of at least two pointers in a captured image frame comprises generating a vertical intensity profile (VIP) from the captured image frame, the VIP comprising peaks generally corresponding to the at least two pointers; determining if the peaks are closely spaced and, if the peaks are closely spaced, fitting a curve to the VIP; analyzing the fitted curve to determine peak locations of the fitted curve; and registering the peak locations as the pointer locations.