A Method of Fabric Defect Detection Using Local Contrast Deviation

Abstract

Defect segmentation has been a focal point in fabric inspection research, and remains challenging because it detects delicate features of defects complicated by variations in weave textures and changes in environmental factors (e.g., illumination, noise). Based on characteristics of fabric structure, an approach of using local contrast deviation (short for LCD) is proposed for fabric defect detection in this paper. LCD is a parameter to describe features of the contras difference in four directions between the analyzed image and a defect-free image of the same fabric, and used a threshold function for defect segmentation. The validation tests on the developed algorithms were performed with images from TILDA's Textile Texture Database, and comparing with an approach of using modified local binary patterns (short for LBP), experimental results show that the proposed method has robustness and simplicity.

Real time fabric defect detection system on an embedded DSP platform

“…Different edge detections methods with connecting broken edges have been discussed in [11]. Shi [12] used a segmentation method based on local contrast deviation for fabric defect detection. Real time region of interst croping for faster processing is discussed in [13].…”

Fabric Defect Detection Using Wavelet-Enhanced Single-Point Photoelectric Sensing System

“…Defective window is found through comparing ALBP features with threshold. Shi et al [3] proposed an approach using local contrast deviation for fabric defect detection based on characteristics of fabric structure. There are two methods for fabric defect detection: non-motif-based approach [4][5] and motif-based approach [6][7].…”