COMPARISON OF HOUGH TRANSFORM AND SIGNAL DETECTION THEORY APPROACHES FOR CURVE DETECTION.

Published

Journal Article

A signal detection theory approach is used for curve detection in digital images corrupted by random noise. This approach is shown to be a refinement of the frequently used Hough transform, and it therefore results in improved performance, both in deciding the presence or absence of a curve in the image and in determining the location in the image of an existing curve. The performances of these two approaches are compared for various signal-to-noise ratios and found to be significantly different for some values of signal-to-noise ratio.

Duke Authors

Cited Authors

  • Hunt, DJ; Nolte, LW; Ruedger, WH

Published Date

  • January 1, 1987

Published In

  • Proceedings of the Annual Southeastern Symposium on System Theory

Start / End Page

  • 520 - 524

Citation Source

  • Scopus