Facial Image Quality Assessment for Mobile Authentication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Facial recognition authentication in mobile devices often fails due to poor quality image captures, leading to false negatives or false positives, as the device may struggle to accurately identify facial landmarks when the face is positioned outside the range of accounted poses, resulting in alignment issues such as pitch, yaw, and roll.
Innovation Solution
A method is implemented in mobile computing devices to assess the quality of captured facial images by generating confidence scores for facial detection, landmark detection, and geometric consistency, which are combined to determine an image quality score, allowing the device to reject low-quality images and alert users to improve their capture pose.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If facial recognition authentication is performed using captured images, then user access functionality is enabled, but authentication failures occur due to poor image quality
Solution Approach 1:
The system performs preliminary quality assessment of captured images by generating confidence scores for facial detection, landmark detection, and geometric consistency before proceeding with authentication. This preliminary action filters out poor-quality images that would cause authentication failures, ensuring only high-quality images are used for the actual authentication process.
Solution Approach 2:
The system generates feedback in the form of quality scores and confidence metrics that indicate whether a captured image is suitable for authentication. This feedback mechanism allows the system to identify and reject images with poor geometric consistency or inadequate landmark detection, thereby improving authentication reliability by preventing use of low-quality images.
2Ease of operation
If the device accepts images with various poses for authentication, then ease of operation is improved, but alignment issues such as pitch, yaw, and roll cause authentication failures
Solution Approach 1:
The system changes the parameter assessment by introducing geometric consistency scoring that specifically evaluates pitch, yaw, and roll angles of captured images. Rather than simply accepting or rejecting images based on pose, the system quantifies the degree of alignment and uses this metric to determine suitability for authentication, allowing flexible pose acceptance within acceptable geometric parameters.
3Measurement precision
If the device uses multiple confidence scores and geometric consistency checks, then image quality assessment accuracy is improved, but computational complexity increases
Solution Approach 1:
The quality assessment process is segmented into distinct independent components: facial detection confidence scoring, landmark detection confidence scoring, and geometric consistency scoring. Each component operates separately and contributes to the overall quality assessment, allowing the system to achieve comprehensive evaluation through modular processing rather than a single complex algorithm.
Data Source
AI summary
In general, this disclosure describes techniques for assessing image quality of captured facial images. An example method includes capturing an image, generating a facial detection confidence score based in part on a likelihood that a representation of at least a portion of a face is included in the image, generating a facial landmark detection confidence score based at least in part on a likelihood that representations of facial landmarks are accurately identified in the image, and generating a geometric consistency score based at least in part on a difference between a point of intersection between a nose base and a line segment that passes through each eye and a midpoint of the line segment. The method also includes generating an image quality score based in part on a combination of the confidence scores, and the consistency score, and classifying an image quality based on the image quality score.