Handwriting Recognition Score Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Typical handwriting recognition systems face challenges in decoding characters due to limited screen space on user devices, particularly when users input handwritten text using delayed strokes or overlapping methods, making it difficult to determine the best recognition without knowing the input manner.
Innovation Solution
A combined recognition system utilizing both delayed stroke and overlapping recognizers, which generate recognition scores for multi-character handwritten inputs, allowing for simultaneous processing and selection of the optimal recognition based on positional information and shared computation resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a complete handwriting recognition system is used to handle formatting and segmentation, then recognition accuracy is improved, but device complexity increases and performance deteriorates on devices with limited screen space
Solution Approach 1:
The patent segments the handwriting recognition task into distinct components: a delayed stroke recognizer that handles characters written after the input area is filled, and an overlapping recognizer that handles characters that overlap with previously written characters. Each recognizer is specialized for specific input patterns, allowing the system to maintain high accuracy without requiring a single complex complete recognition system.
2Adaptability or versatility
If multiple recognizers are applied to handle different input patterns, then adaptability to various handwriting methods is improved, but device complexity increases
Solution Approach 1:
The patent creates a unified recognition system that incorporates both delayed stroke and overlapping recognition capabilities within a single integrated framework. The system universally handles multiple input patterns (delayed strokes, overlapping characters, and normal sequential input) through a cohesive architecture that shares common resources while providing specialized processing for different handwriting methods.
3Ease of operation
If recognition is performed without knowing the input manner, then ease of operation is improved, but measurement precision deteriorates because it is difficult to determine the best recognition
Solution Approach 1:
The patent implements a feedback mechanism where the system evaluates recognition scores from both the delayed stroke recognizer and the overlapping recognizer, then selects the recognition result with the higher score. This feedback-based selection process automatically determines the best recognition without requiring users to specify their input manner, maintaining ease of operation while improving accuracy through intelligent score comparison and selection.
Data Source
AI summary
An optimal recognition for handwritten input based on receiving a touch input from a user may be selected by applying both a delayed stroke recognizer as well as an overlapping recognizer to the handwritten input. A score may be generated for both the delayed stroke recognition as well as the overlapping recognition and the recognition corresponding to the highest score may be presented as the overall recognition.