Handwriting Recognition for Mixed Diagram Elements

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional handwritten diagramming applications on computing devices are limited in handling the complexity of diagramming, forcing users to navigate menus and follow specific drawing orders, which is non-intuitive and hinders quick diagram creation, especially for mixed content diagrams with shapes and text.

Innovation Solution

A system and method that enables users to draw shapes and text naturally without pre-determined orders by accurately differentiating between shapes and text, allowing for freehand diagram creation, editing, and manipulation of digital ink while retaining the original style and relationships, using a handwriting recognition system that includes preprocessing, segmentation, recognition, and output stages to convert hand-drawn inputs into typeset or beautified digital documents.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If conventional handwritten diagramming applications use menu navigation and predefined drawing orders, then the system structure is simple, but the ease of operation deteriorates and productivity decreases

Engineering Contradiction:
Improveease of diagram creationVSAvoidsystem complexity
Core Design Contradiction:
Ease of operationVSDevice complexity

Solution Approach 1:

The system automatically performs handwriting recognition, shape identification, and text extraction without requiring user intervention. The application self-adjusts by analyzing stroke patterns and spatial relationships to differentiate between diagram elements and text, eliminating the need for manual mode switching or menu navigation.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system changes parameters such as stroke continuity, curvature, pressure patterns, and spatial distribution to automatically differentiate between shapes and text. By monitoring these parameter variations in real-time, the system adapts its recognition mode without requiring predefined drawing orders or manual configuration.

Inventive Principle:
Principle #35Parameter changes

2Productivity

If the system requires specific drawing orders and menu navigation, then device complexity is low, but productivity deteriorates

Engineering Contradiction:
Improvediagram creation speedVSAvoidrecognition system complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The system performs preliminary classification of input strokes during the drawing process itself, continuously analyzing stroke patterns and spatial relationships to predict and identify diagram elements as they are being created. This allows immediate recognition and conversion to digital format without requiring post-processing or adherence to predefined sequences.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The recognition system dynamically adapts its parameters and thresholds based on the current drawing context, stroke characteristics, and spatial relationships. The system continuously adjusts its classification criteria to accommodate different diagram types and user styles, enabling high-speed processing without sacrificing accuracy.

Inventive Principle:
Principle #15Dynamics

3Ease of operation

If the system allows freehand drawing without predefined orders, then ease of operation improves, but measurement precision and recognition accuracy deteriorate

Engineering Contradiction:
Improvenatural drawing capabilityVSAvoidshape and text differentiation accuracy
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The system adds temporal and spatial dimensions to the analysis by monitoring the sequence, speed, and spatial distribution of strokes over time. By analyzing the temporal pattern of stroke input and the spatial relationships between elements, the system can accurately differentiate between text and shapes even when drawn freely without predefined orders.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The system provides continuous feedback by analyzing stroke patterns in real-time and adjusting its recognition thresholds based on the observed drawing behavior. The feedback mechanism allows the system to learn from the user's drawing style and improve its differentiation accuracy between text and diagram elements while maintaining natural freehand drawing capability.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3365764B1System and method of handwriting recognition in diagrams
Publication Date: 2022.08.03 MYSCRIPT
  • EP3365764B1 patent drawingFigure 1A~1B
  • EP3365764B1 patent drawingFigure 2A~2B
  • EP3365764B1 patent drawingFigure 3A~3B

AI summary

A system for hand-drawing diagrams including text and non-text elements on a computing device is provided, the computing device comprising a processor and at least one non-transitory computer readable medium for detecting and recognizing hand-drawing diagram element input under control of the processor, the at least one non-transitory computer readable medium configured to cause display of a plurality of input diagram elements in interactive digital ink on a display device associated with the computing device, associate one or more of the diagram elements with one or more other of the diagram elements in accordance with a class and type of each diagram element, and cause re-display of the diagram elements based on one or more interactions with the digital ink received and in accordance with the one or more associations.