Hand Gesture Recognition via Adaptive Skin Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing hand gesture recognition systems are either cumbersome with marker systems, computationally intensive and expensive with depth cameras, or lack robustness under varying lighting conditions.
Innovation Solution
A hand gesture recognition system utilizing a conventional video camera with image processing modules for face and hand detection, tracking, and skin color model updating to enable real-time gesture recognition without the need for proprietary hardware or markers, employing modules like face detection, hand tracking, skin segmentation, and shape feature extraction to identify gestures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If marker systems are used for hand gesture recognition, then positioning accuracy is improved, but system complexity and ease of operation deteriorate due to extensive user positioning of markers
Solution Approach 1:
The patent extracts the marker positioning requirement from the system by using skin color segmentation to automatically detect hand regions without any external markers. The hand detection module processes video frames by analyzing skin color characteristics to identify and track hand gestures, eliminating the need for users to position physical markers on their bodies.
Solution Approach 2:
The patent introduces skin color as an intermediary characteristic to bridge the gap between the user's hand and the camera. By segmenting based on skin color properties (hue, saturation, value) and comparing them against learned hand region characteristics, the system indirectly identifies hand gestures without direct marker contact.
2Difficulty of detecting and measuring
If depth cameras are used for hand gesture recognition, then gesture detection capability is improved, but computational intensity and cost worsen
Solution Approach 1:
The patent replaces expensive depth cameras with conventional video cameras that capture standard RGB images. The system achieves gesture recognition by processing these standard images through skin color segmentation and shape feature extraction, eliminating the need for costly proprietary depth sensing hardware while maintaining effective gesture detection.
Solution Approach 2:
The patent substitutes the mechanical/optical depth sensing mechanism of depth cameras with a computational approach based on color image processing. Instead of using complex depth sensors, the system uses algorithmic analysis of skin color distributions and hand shape features extracted from conventional camera images to achieve gesture recognition.
3Ease of manufacture
If conventional video cameras are used for hand gesture recognition, then cost and availability are improved, but robustness under varying lighting conditions deteriorates
Solution Approach 1:
The patent transforms the lighting sensitivity issue by changing the parameter space from absolute pixel intensity to normalized color characteristics. By converting to HSV color space and analyzing hue, saturation, and value ratios rather than raw intensity values, the system becomes invariant to lighting variations while maintaining skin color identification accuracy.
Solution Approach 2:
The patent implements feedback through iterative skin color model updating. The system continuously refines its understanding of skin color distributions by analyzing multiple video frames and adjusting its color thresholds and hand region models adaptively, enabling robust performance across different lighting conditions without requiring manual recalibration.
Data Source
AI summary
A cost-effective and computationally efficient hand gesture recognition system for detecting and/or tracking a face region and/or a hand region in a series of images. A skin segmentation model is updated with skin pixel information from the face and iteratively applied to the pixels in the hand region, to more accurately identify the pixels in the hand region given current lighting conditions around the image. Shape features are then extracted from the image, and based on the shape features, a hand gesture is identified in the image. The identified hand gesture may be used to generate a command signal to control the operation of an application or system.