Biometric System Dual Image Capture for Transit Check-in
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Self-service terminals and kiosks for transit check-in face challenges with inaccurate and slow biometric identification due to poor image quality and high processing demands, particularly in verifying multiple subjects within a region.
Innovation Solution
A biometric system utilizing two image capture devices, one for defining a virtual region and another for capturing high-resolution images, with 3D segmentation and image scoring to ensure accurate identification, allowing access only if images meet predetermined quality thresholds.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single high-resolution image capture device is used, then image quality is improved, but processing time increases and identification accuracy decreases due to poor image quality
Solution Approach 1:
The patent divides the image capture function into two separate devices: a first image capture device for capturing low-resolution images to identify subjects within a region, and a second image capture device for capturing high-resolution images of identified subjects. This segmentation allows the system to first quickly locate subjects using low-resolution images, then capture high-resolution images only of relevant subjects, thereby reducing overall processing time while maintaining high image quality for identification.
Solution Approach 2:
The system performs preliminary subject identification using low-resolution images from the first capture device before capturing high-resolution images. This preliminary action filters out non-relevant subjects, so that the second capture device only needs to capture images of identified subjects, significantly reducing processing time and computational resources while ensuring high image quality for actual identification.
2Measurement precision
If high-resolution images are captured for all subjects in a region, then identification accuracy is improved, but processing power requirements increase significantly
Solution Approach 1:
The patent segments the image capture process into two stages with two different capture devices. The first device captures low-resolution images of all subjects to enable quick identification of who is in the region. The second device then captures high-resolution images only of the identified subjects. This segmentation reduces the total number of high-resolution images that need to be processed, thereby reducing processing power requirements while maintaining identification accuracy for the actual subjects of interest.
Solution Approach 2:
The system applies different image quality levels to different subjects based on their relevance. Identified subjects receive high-resolution imaging for accurate identification, while non-identified subjects are only processed at low resolution. This local quality approach ensures high identification accuracy for relevant subjects while minimizing overall processing power consumption by not applying high resolution to all subjects uniformly.
3Measurement precision
If manual check-in with attendants is used, then verification accuracy is improved, but labor costs and queue times increase
Solution Approach 1:
The patent implements a self-service biometric verification system where passengers automatically have their images captured and processed by the dual capture device system. The first capture device identifies subjects within the region, and the second capture device captures verification images, all without requiring manual intervention. This self-service approach maintains verification accuracy through high-resolution imaging while dramatically increasing check-in speed and reducing queue times compared to manual attendant-based systems.
Data Source
AI summary
A device for verifying a subject includes: a device body comprising a processor and a biometric system; wherein the biometric system comprises a first image capture device and a second image capture device, in which the first image capture device is configured to define a spatial region and the second image capture device is configured to capture an image of a subject within said spatial region, and the processor is configured to conduct an identification process on the captured image of the subject within the spatial region.