Face Recognition Using Dynamic Focal Length Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current smartphone face recognition technologies face challenges in accurately distinguishing between real faces and photos, leading to potential security vulnerabilities and inconvenience in detection processes.

Innovation Solution

An electronic device equipped with a face image acquisition device that consecutively acquires multiple face images by changing focal length, determining a second focal length range based on image definitions, and judging whether the face is real by assessing the size of this range, without requiring additional hardware, thereby enhancing security and detection convenience.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional face recognition is used, then the detection process is simple, but the accuracy of distinguishing real faces from photos is insufficient

Engineering Contradiction:
Improveface recognition accuracyVSAvoiddetection process complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies the dynamics principle by changing the focal length of the camera lens dynamically during the face recognition process. The system captures multiple face images at different focal lengths (first focal length range and second focal length range) to obtain images with different depths of field. This dynamic adjustment allows the system to analyze focus differences and distinguish real faces from photos based on how different parts of the face (eyes, nose, mouth) come into focus at different focal lengths, thereby improving recognition accuracy without requiring additional hardware.

Inventive Principle:
Principle #15Dynamics

2Measurement precision

If additional hardware is added to improve detection accuracy, then the face recognition precision improves, but the device complexity increases

Engineering Contradiction:
Improveface recognition accuracyVSAvoidhardware complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies the self-service principle by utilizing the existing camera module's focal length adjustment capability to perform depth of field analysis. The system uses the camera's own optical properties and processing capabilities to distinguish real faces from photos, without requiring external sensors or additional hardware components. The camera module serves multiple functions: capturing face images, analyzing focus characteristics, and determining authenticity, thereby improving accuracy while maintaining simple hardware architecture.

Inventive Principle:
Principle #25Self-service

3Measurement precision

If multiple images are captured at different focal lengths, then the accuracy of real face detection improves, but the detection time increases

Engineering Contradiction:
Improvereal face detection accuracyVSAvoiddetection time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent applies the periodic action principle by systematically capturing multiple face images at different focal length ranges in sequence. The system first captures images at a first focal length range, then adjusts to a second focal length range and captures additional images. This periodic capture pattern allows the system to gather necessary focus analysis data while maintaining a structured and efficient detection flow, balancing accuracy improvement with time management.

Inventive Principle:
Principle #19Periodic action

Data Source

PatentEP3432212B1Electronic device and method for recognizing real face and storage medium
Publication Date: 2023.08.30 GUANGDONG OPPO MOBILE TELECOMMUNICATIONS CORP LTD
  • EP3432212B1 patent drawingFigure 1
  • EP3432212B1 patent drawingFigure 2A
  • EP3432212B1 patent drawingFigure 2B

AI summary

An electronic device and method for recognizing a real face are provided. The electronic device includes a face image acquisition device and a processor connected to the face image acquisition device. The face image acquisition device is configured to acquire a plurality of face images of a face at a plurality of focal lengths within a first focal length range. The plurality of face images correspond to a plurality of focus planes respectively. The processor is configured to determine at least one face images with a higher definition out of the plurality of the face images, and form a second focal length range with focal lengths of the at least one face images. The processor is further configured to determine that the face is a real face according to the second focal length range.