Real-time feature extraction method used for AER (Address-Event Representation) image sensor

An image sensor and real-time feature technology, which is applied to instruments, character and pattern recognition, computer components, etc., can solve problems such as difficult to achieve real-time image processing

Inactive Publication Date: 2017-02-22
TIANJIN UNIV
View PDF1 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] refer to figure 1 , the upper part of the figure shows the image acquisition and transmission process of the traditional image sensor, which takes "frame" as the unit, and the image will be output only when the period of one frame is reached, so it is difficult to realize the real-time processing of the image

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
  • Real-time feature extraction method used for AER (Address-Event Representation) image sensor
  • Real-time feature extraction method used for AER (Address-Event Representation) image sensor
  • Real-time feature extraction method used for AER (Address-Event Representation) image sensor

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0018] The real-time feature extraction method proposed by the present invention has a structure such as image 3 Shown. This structure is a two-layer structure. The output event of the AER image sensor is used as the input of the feature extraction module. After the convolution operation of the convolutional layer (S1), and then the maximization operation of the class maximization operation (C1) layer, you can Extract the corresponding feature information.

[0019] The S1 layer implements the event convolution operation. This layer contains N1 convolution modules, and each convolution module is configured with a convolution kernel. In the present invention, the Gabor function is used to calculate and generate the convolution kernel. The calculation of the Gabor function is as Formula (1) shows:

[0020]

[0021] μ 0 =μcosθ+νsinθ

[0022] ν 0 =-μsinθ+νcosθ (1)

[0023] Where μ and ν represent the horizontal and vertical coordinates of the generated convolution kernel, λ represents t...

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 invention relates to the field of AER (Address-Event Representation) image sensor processing. In order to realize high speed and instantaneity, extracted features contain good scale and angle information. Therefore, the invention adopts the technical scheme that a real-time feature extraction method used for an AER image sensor is characterized in that the output event of the AER image sensor is taken as the input of feature extraction, corresponding feature information can be extracted through the convolution operation of a convolutional layer S1 and the maximum operation of a similar maximum operation C1 layer, wherein the layer S1 capable of realizing the convolution operation of the event comprises N1 pieces of convolutional modules, a convolution kernel is configured for each convolutional module, and a Gabor function is adopted to calculate and generate the convolution kernel. The method is mainly applied to the image processing field of the AER image sensor.

Description

Technical field [0001] The invention relates to the field of image processing of AER image sensors, and in particular to a real-time feature extraction method for special image convolution using AER as an information carrier. Specifically, it involves a real-time feature extraction method for AER image sensor. Background technique [0002] AER (Address-Event Representation, AER, address-event representation) image sensor can effectively reduce data redundancy, has the characteristics of ultra-high speed, high real-time, etc., research on the special image back-end processing chip compatible with AER image sensor, can Get rid of the limitation of "frame", take events as the research object, and receive and calculate the event stream generated by front-end sensors in real time. [0003] reference figure 1 , The upper part of the figure shows the image acquisition and transmission process of traditional image sensors. It uses "frames" as the unit, and images are output only when the ...

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 Applications(China)
IPC IPC(8): G06K9/46
CPCG06V10/462
Inventor 姚素英卢成业高志远聂凯明高静史再峰
Owner TIANJIN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap