Approximate Circuits in Low-Power Image and Video Processing: The Approximate Median Filter

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F17%3APU126444" target="_blank" >RIV/00216305:26230/17:PU126444 - isvavai.cz</a>

Výsledek na webu

<a href="https://www.radioeng.cz/fulltexts/2017/17_03_0623_0632.pdf" target="_blank" >https://www.radioeng.cz/fulltexts/2017/17_03_0623_0632.pdf</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.13164/re.2017.0623" target="_blank" >10.13164/re.2017.0623</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Approximate Circuits in Low-Power Image and Video Processing: The Approximate Median Filter

Popis výsledku v původním jazyce

Low power image and video processing circuits are crucial in many applications of computer vision. Traditional techniques used to reduce power consumption in these applications have recently been accompanied by circuit approximation methods which exploit the fact that these applications are highly error resilient and, hence, the quality of image processing can be traded for power consumption. On the basis of a literature survey, we identified the components whose implementations are the most frequently approximated and the methods used for obtaining these approximations. One of the components is the median image filter. We propose, evaluate and compare two approximation strategies based on Cartesian genetic programming applied to approximate various common implementations of the median filter. For filters developed using these approximation strategies, trade-offs between the quality of filtering and power consumption are investigated. Under conditions of our experiments we conclude that better trade-offs are achieved when the image filter is evolved from scratch rather than a conventional filter is approximated.

Název v anglickém jazyce

Approximate Circuits in Low-Power Image and Video Processing: The Approximate Median Filter

Popis výsledku anglicky

Low power image and video processing circuits are crucial in many applications of computer vision. Traditional techniques used to reduce power consumption in these applications have recently been accompanied by circuit approximation methods which exploit the fact that these applications are highly error resilient and, hence, the quality of image processing can be traded for power consumption. On the basis of a literature survey, we identified the components whose implementations are the most frequently approximated and the methods used for obtaining these approximations. One of the components is the median image filter. We propose, evaluate and compare two approximation strategies based on Cartesian genetic programming applied to approximate various common implementations of the median filter. For filters developed using these approximation strategies, trade-offs between the quality of filtering and power consumption are investigated. Under conditions of our experiments we conclude that better trade-offs are achieved when the image filter is evolved from scratch rather than a conventional filter is approximated.

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

<a href="/cs/project/GA16-17538S" target="_blank" >GA16-17538S: Přibližná ekvivalence pro aproximativní počítání</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2017

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

Radioengineering

ISSN

1210-2512

e-ISSN

—

Svazek periodika

26

Číslo periodika v rámci svazku

3

Stát vydavatele periodika

CZ - Česká republika

Počet stran výsledku

10

Strana od-do

623-632

Kód UT WoS článku

000411297800001

EID výsledku v databázi Scopus

2-s2.0-85029600112