A Fast Algorithm for Computing a Histogram on Reconfigurable Mesh

  • Authors:
  • Ju-wook Jang;Heonchul Park;Viktor K. Prasanna

  • Affiliations:
  • -;-;-

  • Venue:
  • IEEE Transactions on Pattern Analysis and Machine Intelligence
  • Year:
  • 1995

Quantified Score

Hi-index 0.14

Visualization

Abstract

The reconfigurable mesh captures salient features from a variety of sources, including the CAAPP, the CHiP, the polymorphic-torus network and the bus automaton. It consists of an array of processors interconnected by a reconfigurable bus system. The bus system can be used to dynamically obtain various interconnection patterns between the processors. In this paper, we present a fast algorithm for computing the histogram of an N脳N image with h grey levels in$$O\!\!\left({\bf min}\!\left\{\sqrt h + {\bf log^*} (N/h), N\right\}\!\right)$$time on an N脳N reconfigurable mesh assuming each PE has a constant amount of local memory. This algorithm runs on the PARBUS and MRN/LRN models. In addition, histogram modification can be performed in $O(\sqrt h)$ time on the same model.A variant of our algorithm runs in$$O\!\!\left({\bf min}\!\left\{\sqrt {\bf h} + {\bf log\ log} (N/h), N\right\}\!\right)$$time on an N脳N RMESH in which each PE has constant storage. This result improves the known time and memory bounds for histogramming on the RMESH model.