Predictive Handwriting Input System for Touchscreen Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The inefficiency of entering handwritten characters on touchscreen devices, particularly when large numbers of characters need to be input quickly, due to the size of the screen and the cumbersome nature of traditional input methods.
Innovation Solution
Implementing a predictive input system that uses image recognition to identify handwritten characters and geometric shapes, ranking candidate words based on contextual data, and allowing users to select and complete words with intuitive gestures like strikethrough or leaping inputs, facilitating quick and accurate text entry.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If traditional touchscreen input methods are used, then the device can accept handwritten input, but the input process becomes cumbersome and inefficient
Solution Approach 1:
The system performs preliminary actions by predicting candidate words before the user completes handwriting input. As the user writes characters, the system proactively generates and displays predicted words in real-time, allowing the user to select from pre-computed options rather than writing complete words manually. This resolves the contradiction by maintaining simple touchscreen interaction while dramatically increasing input speed through advance word prediction.
Solution Approach 2:
The patent introduces an intermediary prediction system that mediates between the user's partial handwritten input and the final complete word. Instead of directly converting incomplete handwriting to text, the system inserts a layer of candidate word generation and display, where predicted words serve as intermediaries that the user can quickly select. This intermediary layer bridges the gap between simple input and high productivity.
2Area of stationary object
If the touchscreen size is small, then the device is portable, but entering large numbers of handwritten characters in short time periods becomes difficult
Solution Approach 1:
The prediction system performs preliminary word generation based on partial character input, allowing users to select from pre-computed candidate words rather than writing complete words on a small screen. This enables high input rates despite limited touchscreen area, as users only need to write enough characters to trigger useful predictions rather than completing entire words manually.
Solution Approach 2:
The system adds a temporal dimension to the input process by displaying evolving predictions over time as the user writes. Instead of requiring complete spatial information (full word) before recognition, the system uses temporal progression of partial inputs to generate and refine predictions, effectively compensating for the limited spatial area of the touchscreen through time-based prediction refinement.
3Productivity
If predictive input is implemented, then text input speed increases, but the system complexity increases
Solution Approach 1:
The system implements preliminary word prediction using contextual analysis of previously input text. By analyzing the context beforehand and generating candidate words in advance, the system enables fast input without requiring complex real-time processing during the selection phase. The complexity is front-loaded in the prediction stage rather than the interaction stage.
Solution Approach 2:
The prediction system uses feedback from user selections and continuation inputs to refine and update candidate word lists dynamically. As users select words or continue writing, the system learns from this feedback to improve prediction accuracy, managing complexity through adaptive learning rather than static complex rule sets.
Data Source
AI summary
A computer-implemented method for use in an electronic device includes receiving a first handwritten input comprising at least one handwritten character, and obtaining a plurality of candidate words that correspond to at the least one handwritten character. A first electronic instruction is generated to display information identifying one or more of the candidate words on a display of the electronic device. A second handwritten input indicative of a selection of one or the displayed candidate words is received, and a second electronic instruction is generated to display the selected candidate word on the display of the electronic device.