Hybrid Character Completion via Handwritten Stroke Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing digital handwriting systems struggle with efficient and reliable recognition of text input, particularly when editing typeset content that requires the addition of diacritical marks or other symbols, and they limit usability by requiring users to learn specific gestures for editing.
Innovation Solution
A computing device and method for processing text in digital documents, which includes a display interface, an input surface, an identification module, a retrieving module, a generation module, a recognition module, and a language module. This system identifies typeset characters that can be completed with additional marks, retrieves predefined character versions, generates hybrid characters by combining these versions with handwritten input strokes, and recognizes these hybrid characters to complete the text accurately.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If handwriting recognition systems use standard character encodings and typesetting methods, then text recognition accuracy is improved, but the ability to efficiently edit and complete characters with diacritical marks deteriorates
Solution Approach 1:
The system pre-processes handwritten input by detecting strokes in proximity to existing text and automatically generating completion suggestions before the user finishes writing. This preliminary action allows the system to anticipate intended characters and provide completion options, improving both recognition accuracy and editing efficiency by reducing the need for manual gesture input.
Solution Approach 2:
The patent introduces an intermediary processing layer between handwritten input and final text output. This intermediary system analyzes stroke patterns, detects proximity relationships between strokes and existing characters, and generates hybrid character representations that combine handwritten elements with standardized character forms. This mediator enables accurate recognition of diacritical marks and special characters without requiring users to learn complex gesture systems.
2Adaptability or versatility
If the system requires users to learn specific gestures for editing text, then editing functionality is improved, but usability and ease of operation deteriorates
Solution Approach 1:
The system enables self-service text completion by automatically detecting when a user intends to add diacritical marks or complete characters based on stroke proximity and patterns. Instead of requiring users to learn specific editing gestures, the system autonomously generates completion suggestions and applies corrections, allowing users to simply write naturally while the system handles the complex editing operations automatically.
3Measurement precision
If the system processes handwritten input as digital ink without integration, then handwriting recognition accuracy is improved, but document creation capability and text integration deteriorates
Solution Approach 1:
The patent merges handwritten digital ink with typeset text by detecting strokes in proximity to existing characters and generating hybrid representations that combine both formats. The system integrates handwritten diacritical marks and completion strokes with standardized character forms, creating unified text that maintains both the accuracy of handwriting recognition and the document creation capabilities of typeset text. This merging enables seamless integration of handwritten input into editable, formatted documents.
Data Source
AI summary
A system and method for completing a character of a text of a digital document on a computing device, the computing device comprising a processor, a memory, and at least one non-transitory computer readable medium for recognizing input under control of the processor, the at least one non-transitory computer readable medium is configured to cause display (S900) of at least one typeset character of the text on a display interface of the computing device; detect a handwritten input stroke (S902) performed on the digital document in the vicinity of a typeset character; identify an first typeset character (S904) if the typeset character belongs to a list of base characters according to a language model; retrieve a predefined character version (S906) of the first typeset character from the memory; generate a hybrid character (S908) by replacing the initial typeset character by the predefined character; generate a list of character candidates (S910) with associated probabilities of recognition of the hybrid character provided by a recognition expert; select a recognized character (S912) from the character candidate list by using a language expert.