Vehicle Occupant Face Morphing for Privacy-Compliant State Assessment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle service providing systems face challenges in protecting personal information, such as facial images, while maintaining the ability to accurately assess the state of vehicle occupants, due to legal restrictions on transmitting personally identifiable information.
Innovation Solution
The system employs an in-vehicle controller to execute processes like segmentation, point group generation, patch generation, and morphing to create a processed face image that reflects the occupant's state, which is then transmitted to a server apparatus for analysis, ensuring personal information protection without compromising service quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the in-vehicle camera captures and transmits the actual face image of the occupant to the server apparatus, then the server can accurately assess the occupant's state, but this violates legal restrictions on transmitting personally identifiable information
Solution Approach 1:
The system extracts only the necessary state information from the face image while removing personally identifiable information. The vehicle controller performs segmentation to separate the face region, generates point group information representing facial features, and creates a morphed face image that preserves state characteristics (such as expressions indicating illness or fatigue) while eliminating unique identifying features. This extracted state information is then transmitted to the server for assessment.
Solution Approach 2:
The system creates a morphed face image that serves as a copy of the original face image but with modified properties. This morphed image retains the occupant's state characteristics (expressions, emotions) while replacing unique identifying features with generic or anonymized features. The server receives this copy instead of the original, enabling state assessment without accessing personal information.
2Object-affected harmful factors
If the system processes face images through segmentation, point group generation, and morphing processes in the vehicle controller, then personal information is protected, but the system complexity increases
Solution Approach 1:
The vehicle controller divides the face image processing into distinct segments: first segmenting the face region from the captured image, then generating point group information representing key facial features, and finally performing morphing to create the anonymized image. This segmentation of the processing task allows for systematic implementation of privacy protection while managing computational complexity through modular processing steps.
Data Source
AI summary
A vehicle of a system that provides a service includes an in-vehicle camera configured to capture an image of an occupant of the vehicle, and a vehicle controller. A sever apparatus includes an output unit and a server controller. The vehicle controller segments a captured region of a face of the occupant from the captured image by the in-vehicle camera, and generates point group information and patch information for the face regarding a segmented face image of the occupant. The vehicle controller executes a morphing process on a sample face image with use of the point group information and the patch information regarding the face image of the occupant, to generate a morphed face image to be transmitted to the server apparatus instead of the face image of the occupant.