Recognition of welding defects in radiographic images by using support vector machine classifier

Xin Wang, Brian Stephen Wong, Ching Seong Tan

Research output: Journal PublicationArticlepeer-review

13 Citations (Scopus)


Radiographic testing method is often used for detecting defects as a non-destructive testing method. In this paper, an automatic computer-aided detection system based on Support Vector Machine (SVM) was implemented to detect welding defects in radiographic images. After extracting potential defects, two group features: texture features and morphological features are extracted. Afterwards SVM criteria and receiver operating characteristic curves are used to select features. Then Top 16 best features are used as inputs to a designed SVM classifier. The behavior of the proposed classification method is compared with various other classification techniques: k-means, linear discriminant, k-nearest neighbor classifiers and feed forward neural network. The results show the efficiency proposed method based on the support vector machine.

Original languageEnglish
Pages (from-to)295-301
Number of pages7
JournalResearch Journal of Applied Sciences, Engineering and Technology
Issue number3
Publication statusPublished - 2010
Externally publishedYes


  • Image processing
  • Radiographic testing
  • Support vector machine
  • Welding defects

ASJC Scopus subject areas

  • General Computer Science
  • General Engineering


Dive into the research topics of 'Recognition of welding defects in radiographic images by using support vector machine classifier'. Together they form a unique fingerprint.

Cite this