Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

A Denoising Method for Address Event Stream Data for Dynamic Vision Sensors

A visual sensor and event flow technology, applied in the field of computer vision, can solve the problems of reduced data information integrity, loss of time attributes, and reduced denoising efficiency, so as to ensure integrity, improve denoising efficiency, and avoid conversion to two-dimensional The effect of the frame image

Active Publication Date: 2021-05-04
XIDIAN UNIV
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

But the defect is that this method converts the three-dimensional address event stream data output by the dynamic visual sensor into a two-dimensional frame image, and uses the two-dimensional frame image as the image to be denoised, not only because of the loss of the address event stream data. The time attribute leads to a reduction in the integrity of the data information, and at the same time reduces the denoising efficiency

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A Denoising Method for Address Event Stream Data for Dynamic Vision Sensors
  • A Denoising Method for Address Event Stream Data for Dynamic Vision Sensors
  • A Denoising Method for Address Event Stream Data for Dynamic Vision Sensors

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0036] Below in conjunction with accompanying drawing and specific embodiment, the present invention is described in further detail:

[0037] Refer to attached figure 1 , the implementation steps of the present invention are as follows:

[0038] Step 1) Construct a probabilistic undirected graph model:

[0039] (1a) Assuming that the address event stream data X output by the dynamic visual sensor includes M events, M=5000000 in the present invention, the address event stream data of X after denoising is Y, and each event x in X i,j,t Event x in its spatiotemporal neighborhood i±Δi,j±Δj,t 、x i,j,t±Δt 、x i±Δi,j±Δj,t±Δt has correlation, x i,j,t Event y corresponding to Y i,j,t Has correlation, x in the present invention i,j,t ∈{+1,-1},y i,j,t ∈{+1,-1}, where i, j and t represent the row address, column address and activation time of a single event, i=1,2,...,768, j=1,2,...,640,t ≥1, Δi, Δj and Δt are the offset of row address, column address and activation time respectiv...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The present invention proposes a method for denoising address event stream data of a dynamic visual sensor, which mainly solves the deficiency of denoising address event stream data output by a dynamic visual sensor in the prior art, including the following steps: constructing a probabilistic undirected graph model ; Obtain the energy function of the probabilistic undirected graph model; divide the address event flow data into event areas; perform denoising processing on each event area. The present invention uses the correlation between events in the address event flow data output by the dynamic visual sensor to construct a probability undirected graph model including event row address information, column address information and activation time information, and then divides the address event flow data into Multiple event areas are thus denoised, aiming to ensure the integrity of the address event stream data information output by the dynamic visual sensor and at the same time improve the denoising efficiency of the address event stream data, which can be used to denoise the address event stream data output by the dynamic visual sensor noise.

Description

technical field [0001] The invention belongs to the technical field of computer vision, and relates to a method for denoising address event stream data of a dynamic visual sensor, which can be used for denoising the address event stream data output by the dynamic visual sensor. Background technique [0002] With breakthroughs and leaps in the field of science and technology, computer vision research continues to deepen, and image and video processing have achieved great results in recent years. Various cameras can be seen everywhere in life. The sensor used by the camera is the key to the image quality of the camera. The essential. At present, traditional image sensors such as CCD and CMOS sensors have been widely used in daily life. Simply put, CCD image sensors mainly complete the generation and transfer of signal charges through photosensitive element arrays and charge transfer devices, thereby converting optical signals The CMOS image sensor integrates components with s...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06F17/15
Inventor 吴金建马传威石光明
Owner XIDIAN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com