Robust optical flow detection based on the distance transform with the CNN nonlinear circuits

Hyongsuk Kim, Tamas Roska, Hongrak Son, Leon O. Chua

Research output: Contribution to journalConference article

Abstract

A robust optical flow computation algorithm utilizing the trajectories of feature points has been developed. For some applications of optical flows, correct optical flows though they are not so many are more useful than unreliable ones at every pixel point. The proposed algorithm is for detecting the optical flows only at the feature points. The optical flow vectors are extracted from the trajectory segments of feature points on which distance information is developed through the Distance Transform. A multi-layer CNN structure and nonlinear templates for the proposed algorithm are suggested and examined. The simulation result shows that the proposed algorithm is robust against noise even without any preprocessing.

Original languageEnglish
Pages (from-to)743-752
Number of pages10
JournalIEEE Workshop on Signal Processing Systems, SiPS: Design and Implementation
Publication statusPublished - Dec 1 2000
Event2000 IEEE Workshop on Signal Processing Systems (SIPS 2000) - Lafayette, LA, USA
Duration: Oct 11 2000Oct 13 2000

ASJC Scopus subject areas

  • Engineering(all)

Fingerprint Dive into the research topics of 'Robust optical flow detection based on the distance transform with the CNN nonlinear circuits'. Together they form a unique fingerprint.

  • Cite this