Biometric Hand Analysis Using 3D Range Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current hand analysis systems require hands to be positioned flat for accurate measurements, making it difficult to achieve reliable biometric identification or verification from a distance due to variations in hand and finger orientation.
Innovation Solution
A system that uses a combination of a visible camera and range mapping sensors to capture and correct hand images for pose, transforming randomly positioned hands into a flat orientation for feature measurement, allowing for biometric analysis at standoff ranges with minimal subject cooperation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If hand geometry measurements are performed from a distance, then the ability to perform identification at a distance is improved, but measurement precision deteriorates due to variations in hand position and angle
Solution Approach 1:
The system transitions from 2D image coordinates to 3D spatial coordinates by incorporating range data. The transformation process maps hand features from the distorted 2D image space into accurate 3D space, where geometric relationships are preserved regardless of viewing angle or distance. This dimensional transformation resolves the contradiction by enabling distance-based operation while maintaining measurement precision through spatial reconstruction.
Solution Approach 2:
The system changes the parameter space by introducing range information (depth data) as an additional dimension. By combining 2D image coordinates with 3D range data, the system transforms the measurement parameters from purely 2D pixel coordinates to 3D spatial coordinates, allowing accurate hand geometry measurement even when the hand is positioned at various distances and angles from the sensor.
2Ease of operation
If the hand is held in free space rather than placed on a flat sensor, then ease of operation is improved, but reliability deteriorates due to position and orientation variations
Solution Approach 1:
By incorporating the third dimension (range/depth) into the measurement system, the patent captures the actual spatial position and orientation of the hand in free space. The 3D coordinates allow the system to mathematically transform hand features into a standardized reference frame, eliminating the need for the hand to be placed on a flat sensor while maintaining measurement reliability through spatial transformation.
Solution Approach 2:
The system uses range data to provide feedback about the hand's actual position and orientation in 3D space. This feedback is then used to compute transformation parameters that normalize the hand pose, allowing the system to compensate for variations in hand positioning and maintain consistent measurements regardless of how the hand is held in free space.
3Measurement precision
If pose correction is applied to normalize hand orientation, then measurement precision is improved, but device complexity increases due to additional processing requirements
Solution Approach 1:
The pose correction process leverages the additional 3D dimension provided by range data to simplify the normalization problem. By working in 3D space rather than 2D, the system can directly compute transformation parameters that align the hand with a standard reference pose, improving measurement precision while keeping the processing complexity manageable through efficient 3D geometric calculations.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables accurate biometric measurements of hand geometry from a distance, ensuring reliable identification and verification by normalizing hand images to a standard flat position, thereby overcoming orientation and position variations.
Implementation Method 1
receive from a range mapping sensor a range map of the human hand
Implementation Method 2
receive from a range mapping sensor a range map of the human hand
Data Source
AI summary
A system includes an image sensing device that is configured to receive a digital image of a human hand, and a range sensing device that is configured to receive range data relating a distance from the range sensing device to a plurality of points on the human hand. The system uses the range data to generate a range map of the human hand, normalizes a pose of the image of the human hand using the range data and the image, extracts one or more features of the hand from the normalized pose, and stores the extracted features into a computer storage medium.