Mirror Image Verification for Mobile Device Integrity Checks

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing computer vision technologies are inadequate for rapidly and precisely verifying the integrity of mobile devices, particularly in scenarios where direct image capture and device operation are not possible, leading to inefficiencies and vulnerabilities to fraud.

Innovation Solution

Utilizing machine learning models, such as neural networks, to analyze images of mobile devices captured through a mirror, combined with image processing techniques to determine device integrity, enabling real-time or near real-time verification without human intervention.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If manual image review is used to verify device integrity, then fraud detection capability is improved, but processing time and operational efficiency deteriorate

Engineering Contradiction:
Improvefraud detection capabilityVSAvoidprocessing speed
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent replaces manual mechanical review processes with automated machine learning-based image analysis systems. The ML models automatically detect device damage, screen cracks, and authenticity markers in submitted images, eliminating the need for human reviewers while maintaining high accuracy in fraud detection and enabling rapid processing of multiple device verification requests simultaneously.

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

Solution Approach 2:

The patent introduces machine learning models as intermediary systems between image submission and verification decisions. These ML models act as intelligent mediators that analyze images for device integrity, detect fraud indicators, and provide verification results without requiring direct human intervention, thus bridging the gap between automated processing and reliable fraud detection.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Productivity

If automated image processing is implemented to increase processing speed, then productivity is improved, but measurement precision and reliability deteriorate

Engineering Contradiction:
Improveverification speedVSAvoidintegrity assessment accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent implements preliminary training of machine learning models using extensive datasets of device images with various damage conditions and fraud scenarios. This preliminary action prepares the ML systems with high precision detection capabilities before actual verification operations begin, ensuring that automated processing maintains accuracy comparable to or exceeding manual review standards while achieving rapid processing speeds.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent employs multiple machine learning models with different architectural parameters and detection thresholds optimized for various verification tasks. By adjusting model parameters such as confidence thresholds, detection sensitivity, and analysis depth based on specific verification requirements, the system maintains high measurement precision while achieving fast processing through parallel model execution and selective analysis.

Inventive Principle:
Principle #35Parameter changes

3Reliability

If multiple verification checks are performed to improve reliability, then fraud detection capability is improved, but device complexity and processing steps increase

Engineering Contradiction:
Improveverification robustnessVSAvoidsystem architecture complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent segments the verification system into multiple specialized machine learning models, each trained to detect specific types of device damage or fraud indicators (e.g., screen cracks, bezel damage, authenticity markers). This segmentation allows the system to perform comprehensive multi-faceted verification through coordinated model execution while keeping individual model architectures relatively simple and manageable, avoiding monolithic system complexity.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP4032017B1Method, apparatus, and computer program product for utilizing machine learning to process an image of a mobile device to determine a mobile device integrity status
Publication Date: 2025.11.12 ASSURANT INC
  • EP4032017B1 patent drawingFigure 1~2
  • EP4032017B1 patent drawingFigure 3
  • EP4032017B1 patent drawingFigure 4A

AI summary

A system, apparatus, method and computer program product are provided for determining a mobile device integrity status. Images of a mobile device captured by the mobile device and using a reflective surface are processed with various trained models, such as neural networks, to verify authenticity, detect damage, and to detect occlusions. A mask may be generated to enable identification of concave occlusions or blocked corners of an object, such as a mobile device, in an image. Images of the front and/or rear of a mobile device may be processed to determine the mobile device integrity status such as verified, not verified, or inconclusive. A user may be prompted to remove covers, remove occlusions, and/or move the mobile device closer to the reflective surface. A real-time response relating to the mobile device integrity status may be provided. The trained models may be trained to improve the accuracy of the mobile device integrity status.