Volume : IV, Issue : XII, December - 2014

ARTIFICIAL NEURAL NETWORK BASED DEFECT IDENTIFIER

Dr. G. M. Nasira, Dr. B. Nagarajan, Mrs. P. Banumathi

Abstract :

The purpose of this paper is to automate the detection of weaving defects in plain–woven faic by computerized software. The faic defects identification system requires efficient and robust defect detection algorithms. Due to large number of faic defect classes, the automatic faic defects identification system is very challenging. When we consider reduction of labor cost and its benefits, the automatic faic defects identification system is very economical. Various techniques have been developed to detect faic defects. Based on the features of faic, the defect detection techniques have been characterized into three categories. They are statistical, structural and model based techniques. This paper presents an statistical based approach to the faic defect identification from the images of textile industry. Textile industry needs to produce less defective textiles for minimizing production cost and time consumption and increase the accuracy. The images are acquired, preprocessed, statistical feature is extracted. The Artificial Neural Network is used as identification model. The extracted feature is given as input to the artificial neural network, it identifies the defect. The proposed method shows a better performance when compared with the existing methods.

Keywords :

Article: Download PDF   DOI : 10.36106/ijar  

Cite This Article:

Dr.G.M.Nasira, Dr.B.Nagarajan,Mrs.P.Banumathi Artificial Neural Network Based Defect Identifier Indian Journal of Applied Research, Vol.4, Issue : 12 December 2014

Number of Downloads : 647

References :