Related Experiment Video
Updated: Jul 17, 2026
08:15
Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision
Published on: March 28, 2025
Hand region extraction and gesture recognition from video stream with complex background through entropy analysis
JongShill Lee1, YoungJoo Lee, EungHyuk Lee
1Dept. of Electron. Eng., Inha Univ., Inchon, South Korea.
Summary
This study introduces a novel method for hand gesture recognition using image processing, achieving over 95% accuracy in complex backgrounds for natural human-computer interaction (HCI). The technique effectively extracts hand movements for intuitive control.
Area of Science:
- Computer Vision
- Human-Computer Interaction (HCI)
- Image Processing
Background:
- Traditional hand gesture recognition is limited by markers or color-based extraction, restricting its use with varying skin tones and clothing.
- Complex backgrounds and changing environmental conditions (lighting, individual differences) pose significant challenges for current methods.
Purpose of the Study:
- To develop a robust hand gesture recognition method for natural human-computer interaction (HCI).
- To overcome the limitations of color and marker-based approaches by utilizing image differencing and entropy measurement.
Main Methods:
- Image differencing between sequential frames to isolate motion.
- Entropy measurement for adaptive hand region extraction, incorporating both color and motion information.
- Chain code contour detection and an improved centroidal profile method for gesture recognition.
Main Results:
- Achieved a hand gesture recognition rate exceeding 95% for individuals.
- Demonstrated high accuracy (90-100%) for recognizing 6 distinct hand gestures.
- The method operates effectively at 5 frames per second, suitable for real-time applications.
Conclusions:
- The proposed entropy-based method enables robust hand gesture recognition in complex backgrounds.
- It adapts to variable lighting and individual differences, offering a more natural HCI.
- The high accuracy and efficiency support its application in advanced interactive systems.
