Automatic Grain Size Determination in Microstructures Using Computer Vision Algorithm Based on Support Vector Machine (SVM)

The properties of materials are identified as the results of its microstructures characteristics. Consequently the task of analysis of microstructure is very important in engineering. There are several methods such as the visual inspection and the semi-automatic inspection by image analysis. Visual...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Advanced Materials Research Ročník 1143; s. 194 - 199
Hlavní autori: Marin, Mihaela, Marin, Florin Bogdan, Alexandru, Petrică, Potecaşu, Florentina
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Trans Tech Publications Ltd 01.02.2017
Predmet:
Algorithms
Automation
Grain size
Image analysis
Microstructure
Support vector machines
Tasks
Visual inspection
Automated Measurement
Computer Vision
Microstructure
Grain Measurement
ISBN:9783035710694, 3035710694
ISSN:1022-6680, 1662-8985, 1662-8985
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:The properties of materials are identified as the results of its microstructures characteristics. Consequently the task of analysis of microstructure is very important in engineering. There are several methods such as the visual inspection and the semi-automatic inspection by image analysis. Visual inspection by an operator is subject to human error and can take important time. The semiautomatic method using specific algorithm and technique improves performance of the work, however still needs specific knowledge concerning image filters and image analysis technique. This research’s objectives are to present automatic image analysis algorithm to measure grains in microstructure images.
Bibliografia:Selected, peer reviewed papers from the 7th edition of the International Conference on Material Science&Engineering, UgalMat 2016, May 19-21, 2016, Galati, Romania
ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISBN:9783035710694
3035710694
ISSN:1022-6680
1662-8985
1662-8985
DOI:10.4028/www.scientific.net/AMR.1143.194