Fabric defect detection based on feature fusion of a convolutional neural network and optimized extreme learning machine

Abstract

Aiming to accurately detect various defects in the fabric production process, we propose a fabric defect detection algorithm based on the feature fusion of a convolutional neural network (CNN) and optimized extreme learning machine (ELM). Firstly, we use transfer learning to transfer the parameters of the first 13 convolutional layers and first two fully connected layers of a VGG16 network model as pre-trained by ImageNet to the initial model and fine-tune the parameters. Subsequently, the fine-tuned model is used as a feature extractor to extract features of RGB images and their corresponding L-component images. A principal component analysis is used to reduce the dimensionality of the features and fuse the reduced features. The moth flame optimization (MFO) algorithm is used to initialize the optimization variables of a parallel chaotic search (PCS) algorithm, and the PCS algorithm (as optimized by the MFO algorithm) is used to optimize the input weight and bias of the ELM (i.e., the PCS-MFO-ELM (PMELM)). Finally, the PMELM is used to replace the softmax classifier of the CNN to classify and detect fabric defect features. The experimental results show that on the amplified TILDA dataset, the precision, recall, F1-score, and accuracy rates of this algorithm for fabric holes, stains, warp breaks, dragging, and folds in fabric can reach 98.57%, 98.52%, 98.52%, and 98.50%, respectively, that is, higher than those of other algorithms. Through a validity experiment, this method is shown to be suitable for defect detection for unpatterned fabrics, regular patterned fabrics, and irregularly patterned fabrics.

Fabric defect image generation method based on the dual-stage W-net generative adversarial network

“…Learning-based defect detection methods, particularly those based on deep learning, have emerged as a new trend in defect detection, and have been applied in fabric defect detection, [5][6][7] circuit board defect detection, [8][9][10] surface defect detection, [11][12][13] and more. Deep learning-based defect detection algorithms [14][15][16][17][18][19][20][21][22][23] offer advantages such as high detection efficiency, accuracy, and real-time capabilities. Zhou and colleagues [21][22][23] improved the typical deep learning network and applied it to the fabric defect detection experiment.…”

“…Deep learning-based defect detection algorithms [14][15][16][17][18][19][20][21][22][23] offer advantages such as high detection efficiency, accuracy, and real-time capabilities. Zhou and colleagues [21][22][23] improved the typical deep learning network and applied it to the fabric defect detection experiment. The accuracy of the improved model in the fabric defect detection experiment was higher than that of traditional detection algorithms and manual detection.…”

Classifying fabric defects with evolving Inception v3 by improved L2,1-norm regularized extreme learning machine

“…The aim of this study was to design a method with lower cost and higher classification accuracy to address the issue of fabric defects. The key step of the machine learningbased classification method 4 is the classification of image features. The highly developed deep learning technology [5][6][7] provides a good platform for image feature extraction, and a trained deep learning network can effectively extract the features of fabric images.…”

Clothing image classification algorithm based on convolutional neural network and optimized regularized extreme learning machine

“…CNNs have been diffusely applied in the area of clothing picture classification, in the process of extracting the features of clothing image. Generally, there will be a lot of unnecessary information in the image, which is not helpful for classification, but will reduce the role of important classification features., and the CNN needs a lot of data sets to train, 25 so that they continuously learn the deep features of the image, However, it takes a lot of time. 26 At this time, selecting a neural network that meets the requirements can not only reduce the time cost, but also make the classification effect of the method better.…”