Event Data Stream Frames for Faster Optical Flow Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional optical flow calculation methods using image sensors generate redundant data, leading to high computational costs and processing speed restrictions, while event-based methods lack accuracy due to independent pixel operation and sparse event detection, especially for rapidly moving objects.
Innovation Solution
A method for processing event data flow that calculates time-difference information from event timestamps to determine pixel values, generating image frames that represent movement changes, thereby improving optical flow calculation speed and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If event data is stored in a database and processed by querying and sorting, then data can be retrieved and analyzed, but processing time increases significantly as data volume grows
Solution Approach 1:
The patent applies preliminary action by pre-aggregating event data into summary records at multiple time granularities (e.g., per second, per minute, per hour) and storing them in a time-series database. This preliminary processing eliminates the need to scan and process raw event data from scratch during queries, significantly reducing processing time for historical data analysis while maintaining the ability to handle large data volumes.
2Loss of information
If all event data is processed and stored for analysis, then complete information is available, but memory and storage resources are consumed excessively
Solution Approach 1:
The patent extracts and stores only the essential aggregated information from raw event data at multiple time granularities. Instead of storing all raw event records, it extracts summary statistics (count, sum, average, min, max) for each time period and stores these compressed representations. This extraction maintains information completeness for analysis while dramatically reducing storage resource consumption.
Solution Approach 2:
The patent segments event data into multiple time granularities (second-level, minute-level, hour-level summaries) and stores them separately in a time-series database. This segmentation allows the system to store and query only the relevant aggregated data for each time period, reducing overall storage requirements while preserving the ability to analyze complete event information when needed.
3Speed
If event data is processed in real-time, then timely responses are achieved, but system complexity and processing overhead increase
Solution Approach 1:
The patent implements dynamic processing by allowing the system to adjust its behavior based on data arrival patterns and query requirements. The time-series database can efficiently handle both real-time queries (using pre-computed summaries) and historical analysis queries. The system dynamically selects appropriate data granularities and processing strategies based on the specific query needs, maintaining fast response times while managing system complexity through flexible query optimization.
Data Source
Figure 1
Figure 2A~2B
Figure 3~4
AI summary
The present disclosure provides a method for processing an event data flow and a computing device. The method includes: reading a plurality of pieces of event data with a first duration sequentially from the event data flow; with respect to each piece of event data with the first duration, analyzing the event data to acquire time-difference information about each event within the first duration; and generating an image frame presenting a change in movement within the first duration in accordance with the time-difference information about each event within the first duration.