Interpretation of optical flow through neural network learning

Author

Miyauchi, Minami ; Seki, Masatoshi

Author_Institution

Sch. of Manage. & Inf., Sanno Coll., Kanagawa, Japan

fYear

1992

fDate

16-20 Nov 1992

Firstpage

1247

Abstract

This study proposes a motion interpretation network which allows optical flow interpretation and describes motions on a plane through the use of a neural network with complex back propagation learning. A network for optical flow normalization is proposed for the interpretation of diverse flow patterns, such as real image optical flow. Using test patterns, the generalization capacity of the proposed network is investigated. The ability is confirmed experimentally

Keywords

backpropagation; computer vision; generalisation (artificial intelligence); motion estimation; optical neural nets; back propagation learning; generalization capacity; motion interpretation network; neural network; optical flow normalization; test patterns; Back; Biomedical optical imaging; Image motion analysis; Motion estimation; Neural networks; Optical computing; Optical fiber networks; Optical network units; Optical noise; Optical propagation;

fLanguage

English

Publisher

ieee

Conference_Titel

Singapore ICCS/ISITA '92. 'Communications on the Move'

Print_ISBN

0-7803-0803-4

Type

conf

DOI

10.1109/ICCS.1992.255060

Filename

255060

Link To Document

https://search.isc.ac/dl/search/defaultta.aspx?DTC=49&DC=3432061