Identity Verification via Optical Face Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current identity verification systems that rely on chip readers are limited in their applicability, as they cannot function with ID cards lacking built-in chips, restricting their use in countries like Singapore where such cards are standard.
Innovation Solution
A method and apparatus for identity verification that compares face images from documents with real-time face images using face feature extraction and similarity calculations, allowing verification without the need for chip readers, and can differentiate between authentic and fake documents by processing both face and text information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If chip readers are used for identity verification, then verification accuracy is improved, but device complexity and limitation increase
Solution Approach 1:
The patent replaces the mechanical chip reading system with an optical face recognition system. Instead of using chip readers to verify identity, the system captures facial images, extracts feature points, and compares them to verify identity. This substitution eliminates the limitation of requiring chip-enabled cards while maintaining verification capabilities through biometric recognition technology.
Solution Approach 2:
The patent introduces face feature point data as an intermediary between the captured facial image and the identity verification result. The system extracts key feature points from facial images, converts them into comparable data structures, and uses these feature point matches as the basis for determining identity authenticity, thereby enabling verification without direct chip interaction.
2Adaptability or versatility
If face comparison is performed on document images and real-time images, then adaptability is improved, but measurement precision challenges arise
Solution Approach 1:
The patent transforms the face comparison task from direct image pixel comparison to feature point coordinate comparison. By extracting key facial feature points and converting them into numerical coordinate data, the system enables precise comparison across different image types (document photos, real-time captures, various formats) while maintaining accuracy through standardized feature point matching algorithms.
Data Source
AI summary
A method and an apparatus for identity verification, an electronic device, a computer program, and a storage medium include: obtaining a first image of a document, where the first image contains a first face image; obtaining a second image containing a face of a to-be-verified person; performing face comparison on the first image and the second image to obtain a first comparison result; and obtaining an identity verification result according to the first comparison result.