Fabric defect detection based on GLCM and Gabor filter: A comparison

Abstract

Fabric defect detection has been an active area of research since a long time and still a robust system is needed which can fulfill industrial requirements. A robust automatic fabric defect detection system (FDDS) would results in quality products and more revenues. Many different approaches and method have been tried to implement FDDS. Most of them are based on two approaches, one is statistical like gray level co-occurrence (GLCM) and other is transform based like Gabor filter. This paper presents a new scheme for automated FDDS implementation using GLCM and also compare it with Gabor filter approach. GLCM texture statistics are extracted and plotted against the inter-pixel distance of GLCM as signal graph. The non-defective fabric image information is compared with the test fabric image. In Gabor filter based approach, a bank of Gabor filter with different scales and orientations is generated and fabric images are filtered with convolution mask. The generated magnitude responses are compared for defect decision. In our implementation of both approaches in same environment, the GLCM approach produces higher defect detection accuracies than Gabor filter approach and more computationally efficient.

Introduction

In general, an image of woven fabric sample can be regarded as a typical textured image. The detection of local fabric defects is one of the most captivating problems in computer vision and has received much attention over the years. In the textile industry, careful inspections for woven fabrics have to be carried out because fabric defects may reduce the profit of a company by 45% or 65% [1]. Real time automated fabric defect detection plays a crucial role in the textile manufacturing industry to ensure that the industry meets its high quality standards. Indeed. The production of good quality products is a key issue for increasing profitability and customer satisfaction and thus improving the industry's competitive edge in the global market. If defects in the fabrics are not discovered prior to the garment manufacturing process, significant financial losses can incur.

Typically web textile fabric is 1–3 m wide and is driven with speed ranging from 20 to 200 m/min. At present, the quality inspection process is manually preformed by experts. However they cannot detect more than 60% of the overall defects for the fabric if it is moving faster than 30 m/min. To increase the quality and homogeneity of fabrics, an automated visual inspection system is needed for better productivity [2]. One way to reduce the total manufacturing cost and to provide a more reliable, objective, and consistent quality control process is to use an automated visual inspection system to detect possible defects in textile fabrics. However, automated visual inspection becomes a significant challenge due to some specific features pertaining to textile fabrics, for example:

• Large variety of fabric surfaces has to be examined.

• Defects may take different forms that are usually difficult to classify.

• New classes of defects arising from possible changes or aging of machineries in the production process.

This paper proposes a new GLCM based technique to address the above discussed problem and compares it with the existing Gabor filter based approach. A sample of different kind of texture defects is shown in Fig. 1. The paper is organized as follows: Section 2 discusses few related works while Sections 3 GLCM, 4 Gabor filter describe basics of GLCM and Gabor filter respectively. Section 5 presents implementation of FDDS using both approaches and performance of the system is evaluated in Section 6. Finally, Section 7 draws the conclusion of the research.