Hough transform network: learning conoidal structures in a connectionist framework

Author

Basak, Jayanta ; Das, Anirban

Author_Institution

Machine Intelligence Unit, Indian Stat. Inst., Calcutta, India

Volume

13

Issue

2

fYear

2002

fDate

3/1/2002 12:00:00 AM

Firstpage

381

Lastpage

392

Abstract

A two-layer neural-network model is designed which accepts image coordinates as the input and learns the parametric form of conoidal shapes (lines/circles/ellipses) adaptively. It provides an efficient representation of visual information embedded in the connection weights and the parameters of the processing elements. It not only reduces the large space requirements of the classical Hough transform (HT), but also represents parameters with a higher precision. The performance of the methodology is compared with other existing algorithms and has been found to excel over those algorithms in many cases

Keywords

Hough transforms; learning (artificial intelligence); matrix algebra; neural net architecture; pattern recognition; Hough transform network; connection weights; connectionist framework; conoidal structures; image coordinates; learning; local receptive fields; processing elements; shell clustering; two-layer neural-network model; visual information; Clustering algorithms; Data compression; Image edge detection; Image segmentation; Intelligent networks; Machine intelligence; Motion detection; Pixel; Shape; Vehicle detection;

fLanguage

English

Journal_Title

Neural Networks, IEEE Transactions on

Publisher

ieee

ISSN

1045-9227

Type

jour

DOI

10.1109/72.991423

Filename

991423