Comparative analysis of SVM and ANN classifiers for defective and non-defective fabric images classification

Abstract

The present work gives a comparative analysis of two different classifiers, namely, Support Vector Machine (SVM) and Artificial Neural Network (ANN) to classify defective and non-defective fabric images. The image dataset is prepared by considering all the varieties of fabric materials. The morphological operations, namely, erosion and dilation are used. A total of twelve morphological features are considered for fabric image analysis. Significant morphological features are selected by adopting Feed Backward Selection Technique (FBST) that is applied in the feature reduction process. The overall classification accuracies of 94% and 86.5% are obtained using SVM and ANN classifiers respectively. The SVM classifier is found to give better classification rate than ANN classifier. The work finds applications in apparel industry, quality analysis, cost estimation, online purchase of fabric etc.

Keywords:

• Morphology
• fabric images
• defective fabric
• non-defective fabric
• SVM
• ANN

Disclosure statement

No potential conflict of interest was reported by the authors.