Feature extraction using reconfigurable hardware

Wiesław Pamula

doi:10.1007/978-3-642-15907-7_20

Feature extraction using reconfigurable hardware

Wiesław Pamula

Katedra Systemów Transportowych, Inżynierii Ruchu i Logistyki (RT/RT4)

Wyniki badań: Rozdział w książce/raport/materiał konferencyjny › Wkład w konferencję › recenzja

3 Cytowania z bazy Scopus

Abstrakt

Feature extraction is an important stage in image processing for object classification, tracking or identification. Real time processing adds stringent constraints on the efficiency of this task. The paper presents a discussion of a reconfigurable hardware processing architecture, based on components, for performing feature calculations using convolutions, morphology operators and local statistics. Special attention is directed to pipelining calculations, fast determination of minimum, median and maximum of values. The architecture is optimised for video streams, which provide the image contents using horizontal scanning. An implementation using a low cost FPGA is presented proving the feasibility of this approach.

Język oryginału	angielski
Tytuł publikacji goszczącej	Computer Vision and Graphics - International Conference, ICCVG 2010, Proceedings
Strony	158-165
Liczba stron	8
Edycja	PART 2
Identyfikatory DOI	https://doi.org/10.1007/978-3-642-15907-7_20
Status publikacji	Opublikowano - 2010
Wydarzenie	International Conference on Computer Vision and Graphics, ICCVG 2010 - Warsaw, Polska Czas trwania: 20 wrz 2010 → 22 wrz 2010

Seria publikacji

Nazwa	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Numer	PART 2
Tom	6375 LNCS
ISSN (drukowany)	0302-9743
ISSN (elektroniczny)	1611-3349

Konferencja

Konferencja	International Conference on Computer Vision and Graphics, ICCVG 2010
Kraj/Terytorium	Polska
Miejscowość	Warsaw
Okres	20/09/10 → 22/09/10

Obszary tematyczne ASJC Scopus

Informatyka teoretyczna
Informatyka ogólna

Uzyskaj dostęp do dokumentu

10.1007/978-3-642-15907-7_20

Inne pliki i linki

Link do publikacji w Scopus

Cytowanie

Pamula, W. (2010). Feature extraction using reconfigurable hardware. W Computer Vision and Graphics - International Conference, ICCVG 2010, Proceedings (PART 2 red., str. 158-165). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Tom 6375 LNCS, Nr PART 2). https://doi.org/10.1007/978-3-642-15907-7_20

@inproceedings{2fe82de7fc9949609f9de7b0ab1a643c,

title = "Feature extraction using reconfigurable hardware",

abstract = "Feature extraction is an important stage in image processing for object classification, tracking or identification. Real time processing adds stringent constraints on the efficiency of this task. The paper presents a discussion of a reconfigurable hardware processing architecture, based on components, for performing feature calculations using convolutions, morphology operators and local statistics. Special attention is directed to pipelining calculations, fast determination of minimum, median and maximum of values. The architecture is optimised for video streams, which provide the image contents using horizontal scanning. An implementation using a low cost FPGA is presented proving the feasibility of this approach.",

author = "Wies{\l}aw Pamula",

year = "2010",

doi = "10.1007/978-3-642-15907-7\_20",

language = "English",

isbn = "3642159060",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

number = "PART 2",

pages = "158--165",

booktitle = "Computer Vision and Graphics - International Conference, ICCVG 2010, Proceedings",

edition = "PART 2",

note = "International Conference on Computer Vision and Graphics, ICCVG 2010 ; Conference date: 20-09-2010 Through 22-09-2010",

}

Pamula, W 2010, Feature extraction using reconfigurable hardware. w Computer Vision and Graphics - International Conference, ICCVG 2010, Proceedings. PART 2 wyd., Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), nr PART 2, tom 6375 LNCS, str. 158-165, International Conference on Computer Vision and Graphics, ICCVG 2010, Warsaw, Polska, 20/09/10. https://doi.org/10.1007/978-3-642-15907-7_20

Feature extraction using reconfigurable hardware. / Pamula, Wiesław.
Computer Vision and Graphics - International Conference, ICCVG 2010, Proceedings. PART 2. red. 2010. str. 158-165 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Tom 6375 LNCS, Nr PART 2).

Wyniki badań: Rozdział w książce/raport/materiał konferencyjny › Wkład w konferencję › recenzja

TY - GEN

T1 - Feature extraction using reconfigurable hardware

AU - Pamula, Wiesław

PY - 2010

Y1 - 2010

N2 - Feature extraction is an important stage in image processing for object classification, tracking or identification. Real time processing adds stringent constraints on the efficiency of this task. The paper presents a discussion of a reconfigurable hardware processing architecture, based on components, for performing feature calculations using convolutions, morphology operators and local statistics. Special attention is directed to pipelining calculations, fast determination of minimum, median and maximum of values. The architecture is optimised for video streams, which provide the image contents using horizontal scanning. An implementation using a low cost FPGA is presented proving the feasibility of this approach.

AB - Feature extraction is an important stage in image processing for object classification, tracking or identification. Real time processing adds stringent constraints on the efficiency of this task. The paper presents a discussion of a reconfigurable hardware processing architecture, based on components, for performing feature calculations using convolutions, morphology operators and local statistics. Special attention is directed to pipelining calculations, fast determination of minimum, median and maximum of values. The architecture is optimised for video streams, which provide the image contents using horizontal scanning. An implementation using a low cost FPGA is presented proving the feasibility of this approach.

UR - https://www.scopus.com/pages/publications/78049279821

U2 - 10.1007/978-3-642-15907-7_20

DO - 10.1007/978-3-642-15907-7_20

M3 - Conference contribution

AN - SCOPUS:78049279821

SN - 3642159060

SN - 9783642159060

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 158

EP - 165

BT - Computer Vision and Graphics - International Conference, ICCVG 2010, Proceedings

T2 - International Conference on Computer Vision and Graphics, ICCVG 2010

Y2 - 20 September 2010 through 22 September 2010

ER -

Feature extraction using reconfigurable hardware

Abstrakt

Seria publikacji

Konferencja

Obszary tematyczne ASJC Scopus

Uzyskaj dostęp do dokumentu

Inne pliki i linki

Fingerprint

Cytowanie