A parallel one dimensional resampling algorithm

Abstract

Separable algorithms are useful in casting image warping into a framework that is amenable t o hardware implementation. Growing interest in this area has gained impetus from the widespread profileration o,f advanced workstations and digital signal processors. Examples include realtime hardware for video eflects among others. The central benefit of separable algorithms is the reduction in complexity of one dimensional resampling algorithms. When the input is resiiricted to be one-dimensional, eficient solutions are m a d e possible for the image reconstruction and antialiasing components of resampling. Fant presented such a solution [2] that is well suited f o r hardware implementation. Fant 's algorithm, however, is inherently serial. I n this paper, we present an alternative but equivalent (in the sense of producing exactly the same output scan lines) algorithm, that is parallel. The new algorithm was inspired b y Wolberg's algorithm [4]. The new algorithm adds some overhead to the hardware implementation of Fant's algorithm t o be amble t o achieve parallelism.

<title>Complexity analysis of two-pass algorithm and elliptical weighted average filter for VLSI implementation of perspective texture warping</title>

“…Temp(x,y) R(x,y) Wolberg's modified implementation of Fants algorithm [4] was used for complexity measurements, presented in this paper.…”