Hough Transform: Underestimated Tool In The Computer Vision Field

Abstract

We discuss Hough Transform, some of its key properties, a scheme of fast and complete calculation of the Hough Transform (similar to the Fast Fourier Transform), and an efficient implementation of this scheme on SIMD processors. We also demonstrate an application of the Fast Hough Transform in computer vision by the example of an automatic page orientation detection unit incorporated in an intelligent character recognition system. Both 2D (scanner) and 3D (camera) cases of page acquisition are considered.

Efficient and effective regularised ART for computed tomography

“…However, to further compete with FBP in terms of computational efficiency, it ideally should be faster still. In order to achieve that, we are planning to optimise the speed of this procedure by incorporating a fast Hough transform calculation [17] which will reduce the complexity from O(n 3 ) using rotations to O(n 2 logn).…”

A low computational approach for price tag recognition

“…But at the recognition stage such inclined (or even worse, truncated) zone of price will lead to recognition errors. To calculate the image incline angle the fast Hough transform was used as described in [23]. If the angle is larger than the given threshold value the bounding rectangle coordinates are corrected.…”

On the use of FHT, its modification for practical applications and the structure of Hough image

“…For the following algorithm description to be clear we need some definitions that were first introduced in [11] and further developed in [2] and [7]. We consider an image with sizes and line .…”

“…≤ Let's consider the case where our line is "mostly vertical". In [2] the following parametrization is proposed: p , where is the intersection of with the abscissa axis 0X, and . The sign of is…”