Face Recognition Frame Selection for Access Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing access control systems face challenges in reliably and cost-effectively using facial recognition for secure entry, due to issues like poor camera quality, bad lighting, and the need for expensive upgrades.
Innovation Solution
A method for access control that involves receiving multiple frames of a video stream, generating scores for each frame based on factors like the presence of individuals, pose, and blur, and then selecting a subset of frames for face recognition matching against a database, with the option to actuate security mechanisms based on the results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If expensive camera upgrades and additional lighting equipment are used to improve image quality, then facial recognition reliability is improved, but system cost increases
Solution Approach 1:
The system performs preliminary assessment of image quality metrics (sharpness, lighting conditions, pose angles) before committing to facial recognition processing. This allows the system to identify suitable frames in advance and process only those that meet quality thresholds, avoiding the need for expensive hardware upgrades while maintaining recognition reliability
Solution Approach 2:
The system changes the parameter of frame selection by using multiple video frames instead of relying on a single high-quality image. By capturing and evaluating multiple frames over time, the system can select the optimal frame for recognition without requiring expensive camera upgrades or additional lighting equipment
2Reliability
If multiple video frames are processed for face recognition, then recognition accuracy is improved, but processing time increases
Solution Approach 1:
The system performs preliminary filtering of video frames based on quality metrics (sharpness, lighting, pose) before submitting them to the facial recognition algorithm. This preliminary action ensures that only high-quality frames are processed, maintaining recognition accuracy while minimizing the number of frames that need full processing
Solution Approach 2:
The system processes a selective subset of video frames rather than all frames. By applying quality thresholds and selecting only frames that meet specific criteria, the system achieves sufficient recognition accuracy without the excessive processing time that would result from analyzing every frame in the video stream
3Measurement precision
If strict quality thresholds are applied to select frames, then recognition precision is improved, but the number of usable frames decreases
Solution Approach 1:
The system performs preliminary evaluation of multiple frame quality metrics (sharpness, lighting conditions, pose angles, occlusion levels) and selects frames that meet predetermined thresholds. This preliminary selection process ensures high recognition precision while maintaining a sufficient quantity of usable frames by evaluating multiple candidates before final selection
Data Source
AI summary
Aspects of the invention provide a method of face recognition that includes receiving plural frames of a video stream imaging a candidate individual, e.g., in the field of view of a camera, and generating for each of those frames a score of the image and/or of the candidate therein. This can include a score (or count) indicative of the number of individuals present in the frame, a pose of the candidate individual (e.g., face-on or otherwise), blur in the image, and so forth. The method further includes selecting, based on the respective scores of the frames, a subset of the frames for matching by a face recognizer against a set of one or more images of designated individuals. That set may be of individuals approved for access, individuals to be prevented for access, or otherwise. An output can be generated based on such matching by the face recognizer.