The generalized radon transform: Sampling, accuracy and memoryconsiderations

Journal Article · Thu Sep 23 04:00:00 EDT 2004 · Pattern Recognition

OSTI ID:860965

Luengo Hendriks, Cris L; van Ginkel, Michael; Verbeek, Piet W; van Vliet, Lucas J

The generalized Radon (or Hough) transform is a well-known tool for detecting parameterized shapes in an image. The Radon transform is a mapping between the image space and a parameter space. The coordinates of a point in the latter correspond to the parameters of a shape in the image. The amplitude at that point corresponds to the amount of evidence for that shape. In this paper we discuss three important aspects of the Radon transform. The first aspect is discretization. Using concepts from sampling theory we derive a set of sampling criteria for the generalized Radon transform. The second aspect is accuracy. For the specific case of the Radon transform for spheres, we examine how well the location of the maxima matches the true parameters. We derive a correction term to reduce the bias in the estimated radii. The third aspect concerns a projection-based algorithm to reduce memory requirements.

• Research Organization: Ernest Orlando Lawrence Berkeley NationalLaboratory, Berkeley, CA (US)
• Sponsoring Organization: USDOE; Dutch Ministry of Economic Affairs. IOP grantIBV98006, Unilever Research and Development Vlaardingen TheNetherlands
• DOE Contract Number: AC02-05CH11231
• OSTI ID: 860965
• Report Number(s): LBNL--56367; BnR: 400412000
• Journal Information: Pattern Recognition, Journal Name: Pattern Recognition Journal Issue: 2 Vol. 38; ISSN PTNRA8; ISSN 0031-3203
• Country of Publication: United States
• Language: English

Related Subjects

99 GENERAL AND MISCELLANEOUS
ACCURACY
ALGORITHMS
AMPLITUDES
RADON
Radon transform Hough transform parameterized shape detectionmulti-dimensional image error estimation efficient implementation imageanalysis
SAMPLING
SHAPE