Novel approaches for detecting fabric fault using Artificial Neural Network with K-fold validation

Abstract

In this paper we have proposed a novel method to detect the defects in woven fabric based on the abrupt changes in the intensity of fabric image due to the defects and have constructed a classification model to properly identify the defects. We have also improved an existing method based on histogram processing for the classifier. In classification model we have implemented Artificial Neural Network (ANN). Both of our newly proposed method and improved technique have outperformed the existing methods. We have implemented K-validation to estimate the performance of our classification model. Additionally we have analyzed the performance of our classification model for different experimental parameters. Finally we have presented a comparative analysis of these techniques.