Camera Autofocus Focal Distance for Photo Spoof Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for recognizing remade photographs in the context of identity authentication, such as those used in Taobao shop certifications, are complex and require additional hardware or sophisticated software, leading to inefficiencies in distinguishing between real and manipulated images.

Innovation Solution

The method employs an auto-focus function of a photographing device to determine whether a photographed object is a photograph or a real person by analyzing the focal distance after successful auto-focusing, without the need for additional hardware, using a simple software process.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If fingerprint verification hardware is added to identify remade photographs, then identification accuracy is improved, but device complexity increases

Engineering Contradiction:
Improveidentification accuracyVSAvoiddevice complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces the mechanical/fingerprint verification hardware with an optical analysis method. The system uses the camera's existing autofocus mechanism to detect focal distance, substituting complex hardware verification with a simpler optical measurement approach that leverages the camera's built-in capabilities.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent makes the camera system self-sufficient by using its own autofocus and focal distance measurement capabilities to detect photograph remaking. No external hardware is needed - the camera uses its inherent optical properties to identify whether it is capturing a real person or a photograph.

Inventive Principle:
Principle #25Self-service

2Measurement precision

If video live detection with face detection algorithm is implemented, then identification capability is improved, but software complexity and processing time increase

Engineering Contradiction:
Improveidentification capabilityVSAvoidsoftware complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts the essential identification capability from complex software algorithms and reduces it to a simple focal distance measurement. Instead of implementing full face detection and action recognition software, the system isolates and uses only the focal distance parameter, which can be obtained from the existing autofocus mechanism.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent changes the identification parameter from complex multi-dimensional data (face features, actions, expressions) to a single scalar parameter (focal distance). This parameter change simplifies the software requirements while maintaining identification effectiveness, as focal distance naturally differs between real persons and photographs.

Inventive Principle:
Principle #35Parameter changes

3Measurement precision

If additional hardware facilities are installed for photograph verification, then detection accuracy is improved, but manufacturing cost and device complexity increase

Engineering Contradiction:
Improvedetection accuracyVSAvoidmanufacturing cost
Core Design Contradiction:
Measurement precisionVSEase of manufacture

Solution Approach 1:

The patent makes the camera's autofocus system serve dual purposes: normal focusing for photography and focal distance measurement for photograph verification. This multi-functionality eliminates the need for dedicated verification hardware, reducing manufacturing costs while maintaining detection accuracy.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system uses the camera's own autofocus mechanism to provide verification functionality, making the device self-sufficient. The camera measures focal distance using its built-in sensors and processing, eliminating the need for external verification hardware and reducing overall system cost.

Inventive Principle:
Principle #25Self-service

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach simplifies the recognition of remade photographs, reducing complexity and enhancing efficiency by leveraging the difference in focal distances between real and manipulated images, thereby improving the authentication process.

Implementation Method 1

activate a photographing device of a terminal to auto-focus a photographed object

Methodology Applied
Scientific EffectAuto-focusing: Focusing

Data Source

PatentEP3422698B1Method, device, mobile terminal, and camera for identifying photograph subject
Publication Date: 2024.03.27 ALIBABA GROUP HOLDING LTD
  • EP3422698B1 patent drawingFigure 1
  • EP3422698B1 patent drawingFigure 2
  • EP3422698B1 patent drawingFigure 3

AI summary

A method, an apparatus, a mobile terminal, and a camera for recognizing a photographed object are disclosed in the present application. The recognition method includes starting a photographing device of a terminal to auto-focus a photographed object in response to receiving a photographing instruction; obtaining a post-focusing focal distance of the photographing device after the photographing device successfully focuses on the photographed object; and determining whether an object type of the photographed object is a photograph based on the post-focusing focal distance. The present application resolves the technical problem of the high complexity associated with the schemes for recognizing a remade photograph in existing technologies.