Dual-camera vehicle authorization with two-stage facial recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle authorization systems face challenges in ensuring accurate user identification, particularly when images are captured in low-light environments or at close distances, leading to low resolution images and potential unauthorized access.
Innovation Solution
A dual-image capturing system with both exterior and interior cameras, where the first image is used for initial door unlocking and the second image, taken inside the vehicle, is compared with a stored image using facial recognition and auxiliary feature identification to ensure secure engine start, with adjustable thresholds and weighted matching for improved accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single image capturing device is used for user identification, then the system complexity is low, but the identification success rate decreases in low-light or close-distance environments
Solution Approach 1:
The identification system is segmented into two distinct stages: first identification (door unlocking) and second identification (engine starting). Each stage uses an image capturing device optimized for its specific function, with the first device positioned for external capture and the second device positioned for internal capture. This segmentation allows each device to operate in its optimal environment, improving overall identification reliability while managing system complexity through functional division.
Solution Approach 2:
The system transitions from a single-dimension identification approach to a two-dimension approach by capturing images from both outside and inside the vehicle. The first image is captured from the external environment and the second image is captured from the internal environment, creating a multi-dimensional verification process that significantly improves identification accuracy in varying environmental conditions.
2Ease of operation
If image capture distance is reduced for close access, then user convenience is improved, but image resolution and quality deteriorate
Solution Approach 1:
Different image capturing devices are positioned with different optical characteristics suited to their specific locations and functions. The first image capturing device is configured for external capture at appropriate distances, while the second image capturing device is positioned inside the vehicle to capture images at close distance without sacrificing quality. This local optimization of capture quality according to specific location requirements resolves the contradiction between close access convenience and image quality.
3Reliability
If a two-stage identification process is implemented, then unauthorized access is reduced, but the time required for vehicle operation increases
Solution Approach 1:
The first identification and door unlocking process serves as a preliminary action that prepares the system for the second identification. By completing the first verification stage and unlocking the door in advance, the system enables the second identification to proceed more efficiently. The preliminary action of uploading the first image to the database also prepares reference data for faster comparison during the second stage, reducing the overall time impact of the two-stage process.
Solution Approach 2:
The system implements feedback mechanisms where the result of the first identification influences the second identification process. The first captured image is uploaded to the database and used as reference data for the second comparison. This feedback loop allows the system to leverage information from the first stage to accelerate and optimize the second stage, reducing the cumulative time penalty of dual verification while maintaining high security standards.
Data Source
AI summary
A system for authorizing a user to operate a vehicle comprises first and second image capturing devices disposed inside and outside the vehicle to capture first and second images of the user, respectively; an identification device to compare the first image with a first stored user image, generate a first output, and upload the first image to the database as a second stored user image if a matching degree of the first image with the first stored user image is greater than a first threshold. The identification device further compares the second image with the second stored user image and generates a second output if a matching degree of the second image and the second stored user image is larger than a second threshold. The system further comprises first and second operation execution devices to perform first and second vehicle operations upon the first and second outputs, respectively.