skip to main content


Search for: All records

Award ID contains: 1907838

Note: When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external site maintained by the publisher. Some full text articles may not yet be available without a charge during the embargo (administrative interval).
What is a DOI Number?

Some links on this page may take you to non-federal websites. Their policies may differ from this site.

  1. Free, publicly-accessible full text available April 12, 2024
  2. null (Ed.)
    Fabric density and distortion offer important information on fabric attributes and quality during the manufacturing process. However, most current procedures require human effort, which is often inefficient, time-consuming, and imprecise. In this paper, we propose to use an automatic method using the 2D Fast Fourier Transform (2D-FFT) to count the number of yarns and determine the angle rotation of weft yarns in fabric images. First, we explain the mathematical background of Fourier Transform and 2D-FFT. Then, we use a customized and optimized software package to apply a 2D- FFT to extract image magnitude, phase, and power spectrum. We apply the inverse 2D Fast Fourier Transform (2D-iFFT) on selected frequencies corresponding to periodic structures – basic weave patterns – to reconstruct the original image and extract warp and weft yarns separately. Finally, we use a local adaptive threshold process to convert reconstructed images into binary images for the counting and calculating process. For the weft rotation, we apply a mathematical calculation on the frequency domain to collect the angular distribution and then figure out the major rotation of weft yarns. Our experiments show that the proposed method is highly accurate and capable of inspecting different patterns of fabric. We also observe that the processing time of our proposal method is practical and time-efficient. 
    more » « less
  3. null (Ed.)
    This paper presents a fabric defect detection network (FabricNet) for automatic fabric defect detection. Our proposed FabricNet incorporates several effective techniques, such as Feature Pyramid Network (FPN), Deformable Convolution (DC) network, and Distance IoU Loss function, into vanilla Faster RCNN to improve the accuracy and speed of fabric defect detection. Our experiment shows that, when optimizations are combined, the FabricNet achieves 62.07% mAP and 97.37% AP50 on DAGM 2007 dataset, and an average prediction speed of 17 frames per second. 
    more » « less