Skip to main navigation Skip to search Skip to main content

Spiking cortical model for geometry invariant and antinoise texture retrieval

  • Ruijia Yang
  • , Congyi Lyu
  • , Yunhui Liu
  • , Weiguo Zhou
  • , Chen Chen
  • , Xin Jiang
  • , Peng Li
  • , Haoyao Chen
  • , Ruishuo Xu
  • , Yukun Wang
  • Harbin Institute of Technology Shenzhen
  • Chinese University of Hong Kong
  • Hebei University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In recent years, CBIR (content-based image retrieval) becomes a new hotspot. In the technology, image querying is achieved based on the characteristics of the color, shape, texture, spatial position of the object or the combination of these features. As the images are the most intuitive contents in the multimedia, content-based image retrieval is a very important problem in the multimedia information processing. Spiking cortical model (SCM) used in this paper is a neural network algorithm that generates a series of binary pulse images when excited by the grayscale or color images. And it has a superior performance in the feature extraction and the texture retrieval of images due to the properties of anti-noise and the geometry invariant of rotation, scale and translation. In order to improve the speed of texture retrieval, SCM is modeled based on FPGA in this paper.

Original languageEnglish
Title of host publication2017 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages645-650
Number of pages6
ISBN (Electronic)9781538620342
DOIs
StatePublished - 2 Jul 2017
Externally publishedYes
Event2017 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2017 - Okinawa, Japan
Duration: 14 Jul 201718 Jul 2017

Publication series

Name2017 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2017
Volume2017-July

Conference

Conference2017 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2017
Country/TerritoryJapan
CityOkinawa
Period14/07/1718/07/17

Fingerprint

Dive into the research topics of 'Spiking cortical model for geometry invariant and antinoise texture retrieval'. Together they form a unique fingerprint.

Cite this