Pre-record Data Storage for Face Comparison Timing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In security monitoring systems that use face comparison for image recording triggers, the time taken for face comparison processes can result in missed recordings if individuals move out of the camera frame, and pre-recording for fixed times may capture unintended faces, making it difficult to identify the intended subject during playback.
Innovation Solution
A pre-record data storage device and method that employs a first recorder for initial image capture, a second recorder with greater capacity for storing pre-recorded data, a face comparison processor, and a recording controller to start and finish pre-recording based on face detection and matching times, ensuring accurate recording and easy identification of the compared face during playback.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If pre-recording is applied for a fixed pre-recording time to ensure the intended individual is captured during face comparison process, then the intended individual's behavior can be recorded, but other faces may also be recorded making it difficult to identify which face was to be compared
Solution Approach 1:
The system starts pre-recording in advance when a face is detected, before the face comparison process completes. This preliminary action ensures that the intended individual's behavior during the comparison period is captured, solving the problem of missed recordings while maintaining the ability to identify the correct face through timestamp correlation.
2Measurement precision
If face comparison process is conducted by comparing a plurality of faces with face information, then accurate identification can be achieved, but the process takes several seconds to a dozen seconds causing the object to move out of screen
Solution Approach 1:
The system performs preliminary face detection and starts pre-recording before the lengthy face comparison process begins. This allows the comparison of multiple faces for accurate identification to proceed without time pressure, while the pre-recording ensures the intended subject remains captured even if they move during the several seconds to dozen seconds comparison period.
Solution Approach 2:
The system creates a copy of the face detection result and uses it as a trigger for pre-recording, separate from the main face comparison process. This copying mechanism allows the time-consuming comparison to proceed independently while the recording is already in progress, eliminating the time loss issue.
3Reliability
If fixed time pre-recording is used to capture the intended individual, then recording coverage is improved, but storage capacity is wasted on unrelated faces
Solution Approach 1:
The system extracts and stores only the specific segment of pre-recorded data that corresponds to the face comparison period by using the face detection timestamp as an index. This extracts the relevant information from the pre-recording buffer, ensuring storage capacity is used efficiently for only the necessary data while maintaining complete coverage of the intended individual's behavior.
Data Source
AI summary
A pre-record data storage device includes: a first recorder; a second recorder having a capacity larger than that of the first recorder; a face comparison processor that executes a face comparison process on a person's face detected from an image obtained by photographing the person; and a recording controller. The recording controller allows the first recorder to start pre-recording of the image of the person from a face detection time when the person's face is detected, and to finish the pre-recording at a matched time when matching is confirmed as a result of the face comparison process of the face comparison processor. The recording controller stores pre-recorded data from the face detection time to the matched time in the second recorder.