Biometric Verification via Sparse Vector Consensus Network
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current biometric recognition systems face challenges in security and data privacy, particularly in verification and identification tasks, as they often require transmitting full biometric data and maintaining central databases, which can lead to security breaches and data leakage.
Innovation Solution
The method employs a neural network to generate sparse vector representations of biometric images, which are then distributed to nodes in a consensus network for voting, achieving verification or identification only when consensus is reached, thereby maintaining data privacy and reducing data transfer costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If full biometric data is transmitted to central databases for verification and identification, then accurate biometric recognition can be achieved, but security breaches and data leakage risks increase
Solution Approach 1:
The patent segments the centralized biometric verification system into a distributed consensus network where multiple nodes collectively perform verification. Instead of transmitting full biometric data to a central database, the system divides the verification task across multiple decentralized nodes, each processing portions of the biometric template. This segmentation reduces security risks associated with central data storage while maintaining verification accuracy through collective decision-making.
Solution Approach 2:
The patent extracts and transmits only essential features from full biometric data—specifically, compressed biometric templates or feature vectors—rather than transmitting complete biometric images or detailed personal information. This extraction process retains sufficient information for accurate verification while minimizing the security exposure of sensitive biometric data during transmission and storage.
2Productivity
If centralized databases are used to store biometric data, then verification and identification tasks can be performed efficiently, but data privacy is compromised and data transfer costs increase
Solution Approach 1:
The patent replaces centralized database storage with a distributed network architecture where biometric verification capabilities are segmented across multiple independent nodes. Each node stores and processes portions of the verification logic and template data, eliminating the need for a central repository. This segmentation maintains verification efficiency through parallel processing while preserving data privacy by preventing any single point of data concentration.
Solution Approach 2:
The patent enables each node in the consensus network to independently perform verification operations using locally stored templates and algorithms, reducing the need for centralized data access. Nodes self-serve verification requests by processing biometric templates locally and contributing to the consensus decision, thereby eliminating data transfer to central databases while maintaining verification productivity.
3Measurement precision
If complete biometric data is transmitted across the network, then accurate verification can be achieved, but data transfer costs and network bandwidth requirements increase
Solution Approach 1:
The patent extracts only the essential biometric template or feature vector from complete biometric data for transmission across the network. Instead of transmitting full-resolution biometric images or detailed personal information, the system transmits compressed representations that contain sufficient information for accurate verification. This extraction significantly reduces data transfer costs and network bandwidth requirements while preserving verification accuracy.
Solution Approach 2:
The patent transmits partial biometric information (compressed templates or feature vectors) rather than complete biometric data, applying partial action to achieve verification. The transmitted data contains just enough information for accurate verification without the excess data transfer costs associated with transmitting complete biometric datasets. This partial transmission approach maintains verification precision while reducing energy consumption and transfer costs.
Data Source
AI summary
Image data is run through a neural network, and the neural network produces a vector representation of the image data. Random sparse sampling masks are created. The vector representation of the image data is masked with each of the random sparse sampling masks, the masking generating corresponding sparsely sampled vectors. The sparsely sampled vectors are transmitted to nodes of a consensus network, wherein a sparsely sampled vector of the sparsely sampled vectors is transmitted to a node of the consensus network. Votes from the nodes of the consensus network are received. Whether a consensus is achieved in the votes is determined. Responsive to determining that the consensus is achieved, at least one of identification and verification of the image data may be provided.