Touch Gesture Recognition via Geometric Shape Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing touch screen technologies face limitations in efficiently recognizing advanced gestures and require pre-learning or training, occupying device storage, and are not capable of identifying touch inputs when the screen is in a switched-off state.
Innovation Solution
A method and system that uses predefined parameters such as right-left slant length, left-right slant length, number of touch points, hypotenuse, rectangularity, and elongatedness to identify hand shapes or touch patterns, allowing device-specific actions even when the screen is off, by processing mutual capacitance data and determining the orientation of touch regions using an average angle method.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If pre-learning or training of a model is implemented to enable real-world gesture recognition, then gesture recognition accuracy is improved, but device storage is occupied
Solution Approach 1:
The patent extracts only the essential parameters needed for gesture recognition (touch coordinates, contact area, duration) from the complete touch input data, eliminating the need to store and process large models while maintaining recognition accuracy
Solution Approach 2:
Instead of using pre-trained models to interpret gestures, the patent inverts the approach by using simple geometric parameter calculations (contact area, aspect ratio, orientation angles) to directly identify gestures, eliminating the need for storage-heavy training data
2Adaptability or versatility
If advanced gestures are implemented with multiple fingers and complex patterns, then gesture functionality is improved, but gesture recognition complexity increases
Solution Approach 1:
The patent segments multi-finger touch inputs into individual contact areas, calculating parameters (area, aspect ratio, orientation) for each finger separately and then combining them to identify composite gestures like pinching or grabbing, making complex gesture recognition manageable through systematic breakdown
Solution Approach 2:
The patent uses changes in geometric parameters (contact area size, aspect ratio, orientation angles) to differentiate between various gestures, allowing the system to recognize complex gestures through parameter analysis rather than complex pattern matching algorithms
3Reliability
If the screen is activated to process touch inputs, then touch input detection is improved, but energy consumption increases
Solution Approach 1:
The patent performs preliminary gesture recognition by analyzing touch parameters (contact area, orientation, duration) directly from the touch sensor data before screen activation is required, enabling the system to recognize wake-up gestures and activate the screen only when necessary, thus reducing overall energy consumption
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables efficient and convenient user interactions by allowing device-specific actions without the need for screen activation, reducing storage requirements and improving gesture recognition accuracy.
Implementation Method 1
processing mutual capacitance data
Data Source
Figure 1~3
Figure 4a
Figure 4b
AI summary
A method of performing a touch action in a touch-sensitive device is provided. The method includes detecting a shape of a contact area associated with a touch input provided on a touch screen, determining whether the detected shape is valid based on a predetermined criteria, detecting a gesture based on a movement of a validated shape, determining whether the detected gesture is valid by matching the detected gesture with one or more predefined gestures, and performing the touch action based on a determination that the detected gesture is valid and a determination that the detected shape is valid.