Results 1 
5 of
5
Finding Lines under Bounded Error
 Pattern Recognition
, 1993
"... A new algorithm for finding lines in images under a bounded error noise model is described. The algorithm is based on a hierarchical and adaptive subdivision of the space of line parameters, but, unlike previous adaptive or hierarchical line finders based on the Hough transform, measures errors in i ..."
Abstract

Cited by 14 (7 self)
 Add to MetaCart
A new algorithm for finding lines in images under a bounded error noise model is described. The algorithm is based on a hierarchical and adaptive subdivision of the space of line parameters, but, unlike previous adaptive or hierarchical line finders based on the Hough transform, measures errors in image space and thereby guarantees that no solution satisfying the given error bounds will be lost. In addition, the algorithm can find interpretations of all the lines in the image that satisfy the constraint that each image feature supports at most one line hypothesisa constraint that is often useful to impose in practice. The algorithm can be extended to compute the probabilistic Hough transform and the generalized Hough transform a variety of statistical error models efficiently.
Block Decomposition and Segmentation for Fast Hough Transform Evaluation
 Pattern Recognition
, 1999
"... The decomposition of binary images using rectangular blocks of foreground pixels as primitives is considered. Based on this type of decomposition, a fast method for evaluating the Hough transform is introduced. A complexity analysis of the proposed block Hough transform algorithm sets constraints on ..."
Abstract

Cited by 13 (8 self)
 Add to MetaCart
The decomposition of binary images using rectangular blocks of foreground pixels as primitives is considered. Based on this type of decomposition, a fast method for evaluating the Hough transform is introduced. A complexity analysis of the proposed block Hough transform algorithm sets constraints on the complexity of algorithms used for block decomposition, so that the total decomposition and Hough transform application time is much less than the time consumed by the usual point Hough transform. Using this analysis, we propose two algorithms for the decomposition and segmentation of binary images into rectangular blocks. A combination of these methods leads to significant acceleration in the identification of linear features, which is demonstrated in various image processing experiments.
Finding The 3d Orientation Of A Line Using Hough Transform And A Stereo Pair
, 2000
"... In human vision system, the differences between the left and right images are used to recover 3D properties of a scene. Similarly on an artificial vision system, the differences between the two images can be used for the extraction of many useful 3D characteristics such as the depth, the surface nor ..."
Abstract

Cited by 1 (1 self)
 Add to MetaCart
(Show Context)
In human vision system, the differences between the left and right images are used to recover 3D properties of a scene. Similarly on an artificial vision system, the differences between the two images can be used for the extraction of many useful 3D characteristics such as the depth, the surface normal and the exact position of a point. In this report we derive the formulation for the computation of a line in space based on its projections on a stereo pair of images and on the angle of converge of the two cameras. Experiments have been carried out that shows using our setup the proposed technique is rather unstable. Solutions for future implementations are proposed, which are rising the estimation range of the Hough accumulation array, and increasing the length of the baseline.
IDIAP Finding Lines under Bounded Error
, 1993
"... A new algorithm for nding lines in images under a bounded error noise model is described. The algorithm is based onahierarchical and adaptive subdivision of the space of line parameters, but, unlike previous adaptive or hierarchical line nders based on the Hough transform, measures errors in image s ..."
Abstract
 Add to MetaCart
A new algorithm for nding lines in images under a bounded error noise model is described. The algorithm is based onahierarchical and adaptive subdivision of the space of line parameters, but, unlike previous adaptive or hierarchical line nders based on the Hough transform, measures errors in image space and thereby guarantees that no solution satisfying the given error bounds will be lost. In addition, the algorithm can nd interpretations of all the lines in the image that satisfy the constraint that each image feature supports at most one line hypothesis{a constraint that is often useful to impose in practice. The algorithm can be extended to compute the probabilistic Hough transform and the generalized Hough transform a variety of statistical error models e ciently. 1
Architectures for the Hough Transform: a Survey
, 1996
"... This survey reviews the implementations of the Hough transform on parallel systems and on special purpose devices. The Hough transform continuously receives much attention because of its usefulness both as a tool in industrial applications and as a step in building perceptual representations for com ..."
Abstract
 Add to MetaCart
(Show Context)
This survey reviews the implementations of the Hough transform on parallel systems and on special purpose devices. The Hough transform continuously receives much attention because of its usefulness both as a tool in industrial applications and as a step in building perceptual representations for computer vision tasks. Its computational complexity has motivated many efforts towards fast implementations: all parallel systems, built or just conceived in the last twenty years, have been used as a test bed for parallelization strategies. Furthermore, practical, dedicated real time solutions have emerged to meet the needs of online inspection. This paper covers SIMD, MIMD and special purpose, dedicated implementations. The analysis of asynthotic computational complexity is paired with more practical considerations on the feasibility of each solution.