Touch Screen Coordinate Calculation Using Precomputed Reference Templates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing touch screen devices face inaccuracies and non-linearity in calculating finger position coordinates, particularly when the finger is near the edge or corner of the screen, using standard methods like virtual sensors and centroid algorithms.
Innovation Solution
The system employs reference templates created during a training phase, using actuation signals from objects of known sizes to solve inverse finger response functions, enabling precise finger position and size estimation by comparing actual signals against threshold values and solving equations based on these templates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If standard methods (virtual sensors, centroid, linear interpolation) are used to calculate finger position coordinates, then the calculation process is simple, but the accuracy and linearity deteriorate especially near edges or corners
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing correction values in lookup tables during a training phase. The system预先 determines the relationship between sensor readings and actual finger positions across the entire touch screen, including edge and corner regions, and stores these calibration data for later use. This eliminates the need for complex real-time calculations while maintaining high accuracy throughout the display area.
Solution Approach 2:
The patent transforms the coordinate calculation from using simple geometric methods to using empirically derived correction parameters. By introducing lookup tables that contain pre-computed offset values and scaling factors based on actual touch measurements, the system adjusts the coordinate parameters dynamically based on touch location, thereby improving accuracy without requiring complex algorithms during operation.
2Measurement precision
If reference templates and inverse finger response functions are used to improve accuracy, then the measurement precision improves, but the device complexity increases
Solution Approach 1:
The system performs all complex template matching and inverse function calculations during an initial training phase, storing the results in lookup tables. During actual touch operations, the system only needs to perform simple table lookups and basic arithmetic operations, thereby maintaining high measurement precision while keeping runtime complexity low.
Solution Approach 2:
The patent creates simplified copies of the complex reference template matching process by pre-computing and storing the essential correction information in lookup tables. Instead of performing full template matching during each touch event, the system uses these tabulated copies of the calibration data to quickly determine accurate coordinates through simple interpolation and offset application.
3Measurement precision
If standard centroid algorithms are used, then the ease of operation is maintained, but the measurement precision deteriorates near edges or corners
Solution Approach 1:
The patent applies local quality by using different correction strategies for different regions of the touch screen. The lookup tables contain region-specific calibration data that accounts for local variations in sensor response, particularly at edges and corners where the standard centroid algorithm fails. This allows the system to maintain simple operation throughout while achieving high precision locally in problematic areas.
Data Source
AI summary
Systems and methods and techniques are disclosed for determining the position and size of an object touching a touch-sensitive display. One embodiment may comprise a set of reference templates—where each reference template is a model and/or data associated with a reference object of a given size. The reference templates may be determined on a priori basis—possibly created during a training phase for the touch display system. The training phase may be performed by stepping an object across a touch display and noting its actuation signal—while the modeling step may be performed by calculating actuation signals based on a model of objects of a given size. With an actual object, its position and size may be determined, calculated and/or estimated by solving a set of inverse finger response functions—e.g., as a set of over-determined linear equations.