Swipe Input Character Recognition Normalization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The variability in swipe input trajectories on touch-based software keyboards often results in unintended characters being input, due to non-uniform trajectories, leading to inaccuracies in character recognition.
Innovation Solution
A system and method that utilize a trained model to determine input characters by normalizing the keyboard and trajectory information, extracting shape and relative position data from swipe inputs, and applying this information to a neural network-based model to accurately identify intended characters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If swipe input is used for character input, then input speed and convenience are improved, but character recognition accuracy deteriorates due to non-uniform trajectories
Solution Approach 1:
The patent transforms the raw swipe trajectory into normalized parameters including curvature, length, and key press sequence. By changing the representation parameters from raw coordinate data to extracted geometric features, the system maintains high input speed while improving character recognition accuracy through more discriminative feature parameters.
Solution Approach 2:
The patent introduces an intermediary processing layer between swipe input and character recognition. This intermediary extracts and normalizes trajectory features (curvature, length, key sequence) that mediate between the variable swipe input and the character recognition model, enabling accurate recognition despite trajectory variations.
2Measurement precision
If trajectory normalization is applied, then character recognition accuracy is improved, but processing complexity increases
Solution Approach 1:
The patent extracts only the essential and most discriminative features from the swipe trajectory (curvature, length, key press sequence) rather than processing the entire raw trajectory data. This extraction approach improves recognition accuracy by focusing on key characteristics while limiting processing complexity by excluding redundant information.
Solution Approach 2:
The patent segments the trajectory analysis into distinct processing stages: key press detection, curvature calculation, length measurement, and sequence extraction. This segmentation allows each component to be processed independently with optimized algorithms, improving overall accuracy while managing complexity through modular processing.
Data Source
AI summary
Provided are an artificial intelligence (AI) system and an application thereof, which simulate functions of a human brain, such as recognition and determination, by using a machine learning algorithm, such as deep-learning. A method of processing, by a device, a keyboard input, based on training, may include: displaying a keyboard on a screen of the device; receiving a swipe input of a user, the swipe input connecting a plurality of keys on a displayed keyboard; extracting a trajectory connecting the plurality of keys; applying, to a trained model for a keyboard input, based on the trajectory, trajectory information indicating a shape of the trajectory and a relative position of the trajectory with respect to the keyboard; and determining at least one character corresponding to the trajectory, based on a result of the applying the trajectory information.