Heuristic Finger Detection via 3D Geodesic Distance

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current finger detection methods based on depth images suffer from high detection errors due to interference from complex background images, particularly when using RGB cameras, and existing methods like maximum contour curvature are not robust enough.

Innovation Solution

A heuristic finger detection method that calculates the central point of a hand connected region, identifies extremely far points using 3D geodesic distances, and iteratively detects fingertips and finger regions by analyzing surrounding points within predetermined distance thresholds, effectively filtering noise and improving accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If maximum contour curvature method is used to detect fingertips from depth image, then detection process is simple, but detection accuracy is poor under complicated background

Engineering Contradiction:
Improvedetection process complexityVSAvoidfingertip detection accuracy
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent segments the hand region from the background using depth image thresholding, dividing the image into hand-connected regions and background regions. This segmentation allows subsequent fingertip detection to focus only on relevant regions, improving accuracy without proportionally increasing overall complexity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediate processing step of extracting hand-connected regions from the depth image before applying curvature detection. This intermediary step filters out background interference, allowing the simple curvature method to work effectively on cleaned data.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Ease of manufacture

If RGB camera is used to acquire images for finger detection, then device cost is low, but detection accuracy is poor due to background interference

Engineering Contradiction:
Improvedevice costVSAvoidfinger detection accuracy
Core Design Contradiction:
Ease of manufactureVSMeasurement precision

Solution Approach 1:

The patent changes the parameter used for image acquisition from RGB color space to depth space. By using depth camera to capture Z-axis distance information, the system achieves better finger detection accuracy while maintaining reasonable device cost, as depth cameras have become increasingly affordable.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent substitutes optical-based RGB imaging with depth-based mechanical measurement. Instead of relying on color and intensity information from light reflection, the system uses direct distance measurement from the depth camera, which is less susceptible to background interference.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Measurement precision

If depth image is used to detect finger movements, then detection accuracy is improved, but processing time increases

Engineering Contradiction:
Improvefinger detection accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent extracts only the essential hand-connected regions from the full depth image, discarding unnecessary background data. This extraction reduces the amount of data requiring detailed processing, thereby reducing processing time while preserving detection accuracy.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent performs preliminary processing by thresholding and extracting hand regions before detailed fingertip detection. This preliminary action prepares the data in advance, making subsequent curvature calculation and fingertip identification faster and more efficient.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3514724B1Depth map-based heuristic finger detection method
Publication Date: 2024.05.29 BOE TECHNOLOGY GROUP CO LTD
  • EP3514724B1 patent drawingFigure 1
  • EP3514724B1 patent drawingFigure 2
  • EP3514724B1 patent drawingFigure 3~4

AI summary

The present disclosure provides a heuristic finger detection method based on a depth image. The method comprises the steps of: acquiring a hand connected region from a user's depth image; calculating the central point of the hand connected region; calculating a plurality of extremely far points in the hand connected region that have extremum 3D geodesic distances from the central point; detecting fingertips and finger regions from the plurality of calculated extremely far points; and outputting fingertip positions and the finger regions. The method calculates and detects fingertips of users by means of 3D geodesic distance, without extracting boundary contours of hand regions, which improves robustness of gesture detection and reduces detection error rates. The method has the advantages of higher finger detection accuracy and fast computing speed.