Palm Print Recognition Through Information-Rich Region Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing palm print recognition methods struggle to extract sufficiently discriminative features from highly similar palm print images, leading to reduced recognition accuracy, particularly in scenarios like mobile payments where high accuracy is required.
Innovation Solution
The method involves acquiring a target region in a palm image with rich palm print information, dividing it into non-overlapping sub-regions, extracting features from these sub-regions, and determining a recognition result based on the combined features of these sub-regions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the entire palm print image is used for recognition, then the recognition process is simple, but the feature discriminability is insufficient for highly similar palm images
Solution Approach 1:
The patent divides the palm print image into multiple regions (first region, second region, third region, fourth region) based on anatomical landmarks such as finger bases and palm center. Each region is processed independently to extract local features, which are then combined for recognition. This segmentation enables the system to focus on discriminative areas while maintaining overall palm structure information.
Solution Approach 2:
The patent applies different processing strategies to different regions of the palm image. Each region is assigned specific weightings and processing parameters tailored to its anatomical characteristics and information density. This allows optimal feature extraction from each local area while avoiding the computational burden of processing the entire image uniformly.
2Loss of information
If the entire palm print image is processed, then all palm information is captured, but computational efficiency decreases
Solution Approach 1:
The palm image is divided into four anatomical regions that can be processed in parallel. This segmentation enables efficient computational distribution while preserving complete palm information through systematic coverage of all regions.
Solution Approach 2:
The patent processes only the necessary portions of the palm image (the four anatomical regions) with appropriate detail levels, rather than uniformly processing the entire image. This partial action approach maintains information completeness for recognition while reducing unnecessary computational overhead in less critical areas.
3Ease of manufacture
If traditional feature extraction methods are used, then the processing is straightforward, but recognition accuracy drops for highly similar palm prints
Solution Approach 1:
The patent segments the palm image into anatomical regions and extracts features from each region using optimized parameters. This segmentation enables more sophisticated feature extraction that captures subtle differences in specific areas (such as finger base patterns and palm creases) that are critical for distinguishing highly similar palm prints.
Solution Approach 2:
The patent employs different processing parameters and weightings for each anatomical region. By adjusting parameters such as feature detection sensitivity, edge detection strength, and pattern recognition thresholds for each region, the system achieves high recognition accuracy while maintaining relatively simple processing procedures.
Data Source
AI summary
A palm print recognition method includes acquiring a target palm image, determining a target region in the target palm image in which palm print information richness satisfies a preset condition, determining a plurality of target sub-regions in the target region, the plurality of target sub-regions not overlapping with each other, determining a second feature of the target region based on first features of the plurality of target sub-regions, and determining a palm print recognition result corresponding to the target palm image based on the second feature of the target region.