Fabric Defect Detection Using Activation Layer Embedded Convolutional Neural Network

Abstract

Loom malfunctions are the main cause of faulty fabric production. A fabric inspection system is a specialized computer vision system used to detect fabric defects for quality assurance. In this paper, a deep-learning algorithm was developed for an on-loom fabric defect inspection system by combining the techniques of image pre-processing, fabric motif determination, candidate defect map generation, and convolutional neural networks (CNNs). A novel pairwise-potential activation layer was introduced to a CNN, leading to high accuracy of defect segmentation on fabrics with intricate features and imbalanced dataset. The average precision and recall of detecting defects in the existing images reached, respectively, over 90% and 80% at the pixel level and the accuracy on counting the number of defects from a publicly available dataset exceeded 98%.

Novel Nonlinear Hypothesis for the Delta Parallel Robot Modeling

“…In this section, the novel nonlinear hypothesis called NNH and second order nonlinear hypothesis called SONH of this document are compared with the nonlinear hypothesis of previous investigations [21], [22] called NH for modeling ; z m ), the parameterized equations of an ELP in three-dimensional space as a function of time t can be written as:…”

“…Figure 12 shows the ECP obtained with equation (27). Before the training with equations (4), (6), and α = 1, the initial parameters for the NH are in equation (21), rand is a random number with values between 0 and 1.…”

“…The goal is to find the modeling parameters Figure 1 shows the structure of the linear hypothesis for the output θ i 1 , it is similar for the outputs θ i 1 , θ i membership maps of [7], [8], or the linear activation maps of [15], [16]. Nonlinear hypothesis are described as the combination of the first order terms and it is detailed as the sum of the multiplication of the first order modeling parameters with the first order inputs, and after of the mentioned combination, one bounded map like the sigmoid [19], [20], hyperbolic tangent [21], [22], Gaussian [13], [14], or other is applied to alter the result.…”

“…The associate editor coordinating the review of this manuscript and approving it for publication was Wei-Chang Yeh . authors use hypothesis containing sigmoid activation maps. In [21], [22], authors use hypothesis containing hyperbolic tangent activation maps. In [23], [24], authors use hypothesis containing relu activation maps.…”

Weakly-Supervised Defect Segmentation on Periodic Textures Using CycleGAN

“…In recent years, convolutional neural networks (CNNs) have shown remarkable outcomes in various computer vision applications. In particular, several networks including FCN [19], SegNet [20], and AdapNet [21] demonstrated a notable performance for semantic segmentation, and thus, deep learning-based approaches have been widely applied to defect inspections [22]- [24].…”