Handwriting Ink Stroke Grouping by Spacing and Attributes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current handwriting recognition engines face challenges in accurately grouping ink strokes for context determination, often misinterpreting user intentions due to limitations in analyzing context and handling annotations, leading to incorrect machine input conversions.
Innovation Solution
A method that groups handwriting ink strokes based on both spacing and attributes such as color, width, size, and recognition confidence scores, allowing for intelligent selection of which ink strokes to send to the recognition engine, ensuring accurate conversion of user input.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If handwriting ink strokes are sent to recognition software for conversion, then machine input can be obtained, but the recognition accuracy deteriorates due to inability to properly group strokes and determine context
Solution Approach 1:
The patent segments the plurality of handwriting ink strokes into multiple groups based on spacing attributes and stroke attributes. Each group represents a distinct textual element or word, allowing the recognition engine to process contextually meaningful units rather than individual strokes, thereby improving recognition accuracy while preserving context information.
Solution Approach 2:
The patent performs preliminary grouping and context determination before sending strokes to the recognition engine. By pre-organizing strokes into meaningful groups using spacing and attribute analysis, the system prepares the data in a context-rich format that enhances the recognition engine's ability to accurately convert handwriting to machine input.
2Ease of operation
If all ink strokes are grouped together for recognition, then context determination is simplified, but user annotations and intentions are misinterpreted
Solution Approach 1:
The patent applies local quality by differentiating grouping criteria for different regions of the handwriting input. Annotations are detected as distinct groups separate from the main text body, allowing the system to preserve user intentions while maintaining ease of context determination for the primary text content.
Solution Approach 2:
The patent implements dynamic grouping where the grouping structure adapts based on the detected content type. Different grouping strategies are applied to annotations versus main text, enabling the system to handle diverse user intentions appropriately while maintaining operational simplicity through automated detection rules.
3Productivity
If ink strokes are grouped based only on spacing, then grouping is computationally simple, but grouping accuracy deteriorates due to inability to distinguish different stroke types
Solution Approach 1:
The patent extends the grouping criteria from单一的 spacing parameter to multiple parameters including spacing, stroke width, stroke color, and other stroke attributes. This multi-parameter approach significantly improves grouping accuracy by distinguishing different stroke types and intentions, while the system maintains computational efficiency through optimized attribute analysis.
4Device complexity
If recognition engine receives individual ink strokes, then processing is straightforward, but conversion accuracy to machine input deteriorates due to lack of context
Solution Approach 1:
The patent performs preliminary grouping of ink strokes into contextually meaningful units before sending to the recognition engine. This pre-processing step organizes strokes into groups that represent complete words or textual elements, providing the recognition engine with context-rich input that significantly improves conversion accuracy while maintaining manageable processing complexity.
Data Source
AI summary
One embodiment provides a method including: receiving a plurality of handwriting ink strokes; determining, using a processor, at least one grouping of the plurality of handwriting ink strokes, wherein the grouping is determined using spacing associated with the plurality of handwriting ink strokes and at least one attribute associated with at least one portion of the plurality of handwriting ink strokes; sending the at least one grouping to a recognition engine; receiving machine input from the recognition engine; and displaying, on a display device, the machine input. Other aspects are described and claimed.