Performance of the Hough transform on a distributed memory multiprocessor

Abstract

The Hough transform is a projection-based transform which can be used to detect shapes in images. One of the disadvantages of the transform is its requirement for large amounts of computing power. Parallel machines have given programmers the potential for incredible computing power. To obtain maximum performance from parallel machines, parallel algorithms should be designed to reflect the architecture of the parallel machine. The work reported in this paper compares the performance obtained in running several parallel versions of the Hough transform on a Fujitsu AP1000 distributed memory multiprocessor. ᭧ 1999 Published by Elsevier Science B.V.

Real-time detection of lines using parallel coordinates and OpenGL

“…The Hough transform also provides space for parallel implementations using special hardware, such as a distributed memory multiprocessor [Underhill et al 1999], graphics hardware [Strzodka et al 2003], pyramid multiprocessors [Atiquzzaman 1994], or reconfigurable architectures [Pavel and Akl 1996]. For more information about different existing modifications and implementations of the Hough transform, one can see a comprehensive overview by Illingworth and Kittler [1988].…”

On the computation of the Circle Hough Transform by a GPU rasterizer

“…Many parallel architectures and sophisticated machines have been used to implement the Hough transform in the past, such as distributed memory multiprocessors Underhill et al, 1999), reconfigurable architectures (Pan et al, 1999;Pavel and Akl, 1996), special purpose hardware (Bruguera et al, 1996) and SIMD machines (Kumar et al, 1994). None of these other methods compare to the low cost, high performance and wide availability offered by the GPU.…”

Μεθοδολογία Ανάπτυξης Μεταγλωττιστών Με Εκμετάλλευση Της Δομής Του Λογισμικού Και Του Μοντέλου Του Υλικού

“…Σχετικά με τον μετασχηματισμό του Hough [35], υπάρχουν πολλές σχετικές εργασίες, [14], [27], [84], [87], [72], [71]. Το [14] παρουσιάζει διάφορες παράλληλες εκδόσεις του μετασχηματισμού σε έναν πολυεπεξεργαστή Fujitsu AP1000. Το [27] παρουσιάζει τη παραλληλοποίηση του μετασχηματισμού σε επίπεδο εντολών και μπλοκ.…”