Hand-Drawn Ellipse Recognition Using Convex Hull Area Ratios

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing systems for hand-drawn graphic recognition on smart whiteboards suffer from low accuracy and robustness due to non-standardized user inputs, leading to challenges in distinguishing ellipses from other shapes.

Innovation Solution

A system and method that determines hand-drawn graphics by analyzing hand-drawn strokes through a user interface, using criteria such as convex hulls, inscribed polygons, and ratio thresholds to distinguish ellipses from non-ellipses, and performs fitting operations on polyline strokes to enhance accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional hand-drawn graphic recognition methods are used on smart whiteboards, then the system is simple and easy to operate, but the recognition accuracy is low due to non-standardized user inputs

Engineering Contradiction:
Improvehand-drawn graphic recognition accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent transforms hand-drawn strokes into standardized geometric parameters by calculating convex hulls, inscribed polygons, and area ratios. This parameter transformation converts variable, non-standardized user inputs into consistent geometric measurements that can be accurately recognized and classified, directly resolving the accuracy issue without requiring complex hardware modifications

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The recognition process is divided into distinct segments: stroke collection, convex hull calculation, inscribed polygon determination, area ratio computation, and shape classification. This segmentation breaks down the complex recognition task into manageable steps, improving accuracy through systematic processing while keeping the overall system structure clear and maintainable

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If simple area comparison is used to distinguish ellipses from other shapes, then the method is simple to implement, but the distinction accuracy is insufficient for non-standardized drawings

Engineering Contradiction:
Improveellipse recognition accuracyVSAvoidshape differentiation difficulty
Core Design Contradiction:
Measurement precisionVSDifficulty of detecting and measuring

Solution Approach 1:

The patent introduces an inscribed polygon as an intermediary element between the hand-drawn stroke and the final shape classification. By calculating the maximum inscribed polygon within the convex hull and comparing its area ratio to the convex hull area, the system creates a reliable intermediate measurement that effectively distinguishes ellipses from other shapes, even when user drawings are non-standardized

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent moves from simple 2D area comparison to a multi-dimensional analysis by incorporating convex hull geometry, inscribed polygon characteristics, and area ratio calculations. This dimensional expansion provides more discriminative power for shape classification, enabling accurate ellipse detection while maintaining a relatively simple implementation approach

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

3Reliability

If the system processes all hand-drawn strokes uniformly, then the processing is simple, but the robustness is low when dealing with different stroke patterns and termination conditions

Engineering Contradiction:
Improverecognition robustnessVSAvoidprocessing logic complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent implements dynamic processing logic that adapts to different input patterns. The system determines input termination conditions based on minimum distance thresholds between starting and ending points, and dynamically selects appropriate processing paths for single strokes versus multiple strokes. This dynamic adaptation enhances robustness across different drawing patterns while keeping the processing logic organized and manageable through clear conditional structures

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP4377887B1Systems and methods for hand-drawn graphic determination
Publication Date: 2026.04.08 ZHEJIANG DAHUA TECH CO LTD
  • EP4377887B1 patent drawingFigure 1
  • EP4377887B1 patent drawingFigure 2
  • EP4377887B1 patent drawingFigure 3

AI summary

A system (100) and a method for determining a hand-drawn graphic are provided. The method may include obtaining one or more hand-drawn strokes inputted into a drawing region (310), and in response to determining that an input termination condition is satisfied, determining a hand-drawn graphic corresponding to the one or more hand-drawn strokes (320).