Event Data Tracking With Motion Prediction for Mobile Authentication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing event cameras only output event data when luminance changes exceed a certain level, limiting their dynamic range and requiring high memory capacity for processing, and existing authentication methods using event cameras are not efficient for mobile objects.
Innovation Solution
A data processing method that includes detecting movement vectors in event data from two-dimensionally arrayed capturing pixels, predicting future observation positions, and updating these positions to efficiently process and decode data from event cameras, particularly for mobile objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If event cameras only output data when luminance changes exceed a certain level, then the data output is reduced, but the dynamic range is limited and measurement precision deteriorates
Solution Approach 1:
The patent changes the parameter of luminance change threshold by using relative luminance change detection instead of fixed threshold detection. This allows the system to adapt to different lighting conditions and maintain measurement precision across a wide dynamic range while still reducing data output quantity by only triggering events when meaningful changes occur.
2Loss of information
If all event data from capturing pixels are processed, then complete information is obtained, but memory capacity requirements increase
Solution Approach 1:
The patent extracts only the essential information from event data by identifying and tracking the event occurrence region (connected component of pixels with same polarity) and its movement characteristics. Instead of storing all raw event data, the system extracts movement vectors and region positions, significantly reducing memory requirements while preserving the critical information needed for authentication.
Solution Approach 2:
The patent segments the event data processing by separating pixels into different polarity groups (positive and negative events) and processing each group independently to identify connected components. This segmentation allows efficient tracking of distinct event regions and reduces the overall data processing burden.
3Measurement precision
If authentication is performed using stationary event camera methods, then authentication accuracy is maintained, but efficiency deteriorates for mobile objects
Solution Approach 1:
The patent introduces dynamic tracking of event occurrence regions by calculating movement vectors between consecutive frames. This dynamic approach allows the system to follow moving objects while maintaining authentication accuracy, unlike stationary methods that would lose track of moving targets and reduce efficiency.
Solution Approach 2:
The patent performs preliminary identification of event occurrence regions and their movement patterns before authentication processing. By pre-processing the event data to extract and track region movements, the system prepares the data in an optimized format that speeds up the subsequent authentication process, improving overall efficiency without sacrificing accuracy.
Data Source
AI summary
A data processing method including: detecting a movement vector MV indicating a time change in a position C of an event occurrence region E constituted by a connected pixel group of capturing pixels that output event data e having the same polarity of a change among a plurality of pieces of the event data e output from the capturing pixels; predicting observation positions P1 to P5 of the event data e output from the capturing pixel at a future observation time, based on the movement vector MV; and updating the observation positions P1 to P5 of the event data e at future observation times 0T to 4T, based on the event data e observed at the predicted observation times 0T to 4T and in the predicted observation positions P1 to P5.