Dynamic Vision Sensor Event Correction via Edge Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Dynamic vision sensors (DVS) often omit event data, leading to malfunctions in imaging systems, particularly in motion imaging tasks, due to the use of still-image sequencing, resulting in redundant and unusable data.
Innovation Solution
A method and system for correcting DVS events by generating edge data from image pixels and using it to identify and supplement omitted events, enhancing the accuracy of event data through edge information-based compensation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If DVS uses still-image sequencing to create motion, then data usage is reduced, but redundant and unusable data is created
Solution Approach 1:
The patent segments the image sensor array into two distinct types of pixels: DVS pixels for motion detection and image pixels for still image capture. This segmentation allows each pixel type to perform its specialized function optimally, with DVS pixels capturing motion events and image pixels providing reference frames, thereby eliminating redundant data while preserving motion information.
Solution Approach 2:
The patent introduces edge data as an intermediary element that bridges the gap between image pixels and DVS pixels. The edge detection circuit processes image data to generate edge information, which then guides the event generation process, ensuring that motion detection is based on meaningful edge changes rather than raw pixel sequencing, thus reducing redundancy.
2Adaptability or versatility
If DVS provides event data, then motion imaging is enabled, but omitted events cause system malfunction
Solution Approach 1:
The patent implements a feedback mechanism where the edge detection circuit continuously monitors image data and provides feedback to the event generation process. When edge changes are detected in regions where no DVS events were generated, the system compensates by generating supplemental events, ensuring complete motion information and preventing system malfunction.
Solution Approach 2:
The patent performs preliminary edge detection and analysis before final event generation. By pre-processing image data to identify potential motion regions and edge changes, the system can proactively generate events that might otherwise be omitted, ensuring comprehensive motion capture and system reliability.
3Productivity
If DVS captures pixel-level changes, then motion data efficiency is improved, but omitted events reduce accuracy
Solution Approach 1:
The patent merges the capabilities of DVS pixels and image pixels into a unified sensing system. DVS pixels provide efficient pixel-level motion changes, while image pixels provide reference information for edge detection. By combining these two data sources and processing them through the edge detection circuit, the system achieves both high efficiency and high accuracy in motion data capture.
Solution Approach 2:
The patent uses edge data as an intermediary to enhance event data accuracy. The edge detection circuit processes image pixel data to create edge information that supplements DVS event data, filling in gaps and correcting omissions to improve measurement precision while maintaining the efficiency benefits of pixel-level change detection.
Data Source
AI summary
A method of correcting dynamic vision sensor (DVS) events is described. The method generates event data including events representing motion information of an object included in an image. Additionally, the method generates image data capturing an image and generates edge data representing edge information of the image based on the image data. The method also generates omitted events of the event data based on the edge data. Accuracy of the event data and performance of machine vision devices and systems operating based on the event data are enhanced by supplementing the omitted events of the event data provided from the DVS, using the edge information.