arrow
Return

Accelerating Image Processing Using Reduced Precision Calculation Convolution Engines

delete2023-05-09
delete1
delete
OA
AI
N
Narayan Pokhrel
S
Sakari Snäll
O
Olli I. Heimo *
U
Uruj Sarwar
A
Antti Airola
T
Tero Säntti
DOI:10.1007/s11265-023-01869-5delete
deleteOriginal
deleteShare
deleteSave
Abstract

Abstract

En 中文
In this paper a method of accelerating image processing using convolution engines with reduced precision calculation is presented. The convolution engines are designed to be used with the Pulpissimo platform with RISC-V System-on-Chip. The aim is to move the calculation to the edge. The proposed linear convolution engines operate on 8-bit data set and the logarithmic convolution engine operates on 4-bit reduced precision data. The data reduction is done by using a logarithmic number space. Diminishing the size of the data to be processed reduces the amount of required memory, requirement for memory bandwidth, required computation, and required hardware area while simultaneously increasing the performance. This performance could benefit modern AI and image processing applications, especially in mobile and other battery-operated devices. The results show that the computation in the linear convolution engine is 91 times faster and computation in the logarithmic convolution engine is 122 times faster than in the RISC-V core with plain RISC-V instructions.
Keywords:
Convolution Engine
Image Processing
Reduced Precision Calculation
FPGA
System-on-Chip
Artificial Intelligence

Journal

Signal Image and Video Processing cover
Signal Image and Video Processing
IF:
2.1
Papers:
908
Citations:
4.8K

Organization

U
University of Turku
Scholars:
815
Papers: 297
Citations: 8.8K
Cited Papers

Cited Papers

Medical Students’ Beliefs About Screening for Intimate Partner Violence
err2015-02-06
err0
PREAI
errOluwatoni E. Aluko; Kenneth H. Beck; Donna E. Howard
errShare
errSave
Neologistic Jargon: A Clinicopathological Study
err1970-12-01
err0
errOAAI
errAndrew Kertesz; D. Frank Benson
errShare
errSave
Air Current Applied to the Face Improves Exercise Performance in Patients with COPD
errLung
IF0
err2015-08-09
err0
errOAAI
errNathaniel Marchetti; Matthew R. Lammi; John M. Travaline; David Ciccolella; Brian Civic; Gerard J. Criner
errShare
errSave
Perception of symptoms and quality of life – comparison of patients’ and physicians’ views in the COPD MIRROR study
err2017-07-01
err0
errOAAI
errBartolome Celli; Francesco Blasi; Mina Gaga; Dave Singh; Claus Vogelmeier; Valeria Pegoraro; Nicoletta Caputo; Alvar Agusti
errShare
errSave
Immediate Effect of Fan Therapy in Terminal Cancer With Dyspnea at Rest: A Meta-Analysis
err2019-09-04
err0
PREAI
errJun Kako; Masamitsu Kobayashi; Yasufumi Oosono; Kohei Kajiwara; Mika Miyashita
errShare
errSave
Fan-on-Face Therapy in Relieving Dyspnea of Adult Terminally Ill Cancer Patients: A Meta-Analysis
err2020-07-05
err0
errOAAI
errMarvin Jonne L. Mendoza; Frederic Ivan L. Ting; John Paulo B. Vergara; Danielle Benedict L. Sacdalan; Jennifer Sandoval-Tan
errShare
errSave
Alterações de linguagem nas fases iniciais da doença de Alzheimer
err2005-06-01
err0
errOAAI
errKarin Zazo Ortiz; Paulo Henrique Ferreira Bertolucci
errShare
errSave
Influence of age and schooling in written discourse of healthy adults
err2020-06-08
err0
errOAAI
errLarissa Zanichelli; Rochele Paz Fonseca; Karin Zazo Ortiz
errShare
errSave
researcher View more