Multi-Technology Identity Verification System
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Facial recognition systems face high false acceptance and rejection rates due to factors like image quality, illumination, resolution, expression, and variability in datasets, making them less effective in security applications, particularly for individuals with darker complexions and those with low-resolution images.
Innovation Solution
A system that combines multiple independent identification technologies and platforms, such as facial recognition, object recognition, text recognition, and gait recognition, to enhance identification accuracy by using a combination of internal and external databases and machine learning models, and incorporates geolocation and confidence evaluation to verify identities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If facial recognition systems are implemented, then identification capability is provided, but false acceptance and rejection rates are high
Solution Approach 1:
The patent combines multiple independent identification technologies (facial recognition, object recognition, text recognition, gait recognition) into a unified system. Each technology operates independently and their results are aggregated through machine learning models to produce a final identification decision, thereby reducing false acceptance and rejection rates while maintaining identification capability.
2Reliability
If multiple independent identification technologies are combined, then identification accuracy is improved, but system complexity increases
Solution Approach 1:
The system divides the identification process into separate independent modules (facial recognition module, object recognition module, text recognition module, gait recognition module). Each module processes specific aspects independently and outputs results that are then aggregated by machine learning models, reducing overall system complexity while maintaining high identification accuracy.
Solution Approach 2:
Machine learning models serve as intermediaries that aggregate results from multiple independent identification technologies. These models synthesize the outputs from different recognition systems and produce the final identification decision, simplifying the integration process and managing system complexity.
3Device complexity
If facial recognition is used alone, then the system is simple, but performance accuracy is affected by image quality, illumination, and expression
Solution Approach 1:
The patent merges facial recognition with other identification technologies (object recognition, text recognition, gait recognition) to create a robust multi-technology system. This combination compensates for the weaknesses of individual technologies under varying conditions of image quality, illumination, and expression, thereby improving overall performance accuracy.
Data Source
AI summary
Techniques are described for identifying and/or authenticating entities using a combination of independent identification technologies and/or platforms. In one embodiment, a system can comprising a memory that stores computer executable components and a processor that executes the computer executable components stored in the memory. The computer executable components can comprise a reception component that receives image data captured of a person, and an identification component that employs two or more independent identification technologies and/or platforms to determine an identity of the person based on the image data. In some embodiments, the two or more independent identification technologies are selected from a group consisting of: facial recognition, object recognition, text recognition, and gait recognition.