Search results

Automatic defect detection is a fundamental and vital topic in the research field of industrial intelligence. In this work, the authors develop a more flexible deep learning method for the industrial defect detection.

The authors propose a unified framework for detecting defects in industrial products or planar surfaces based on an end-to-end learning strategy. A lightweight deep learning architecture for blade defect detection is specifically demonstrated. In addition, a blade defect data set is collected with the dual-arm image collection system.

Numerous experiments are conducted on the collected data set, and experimental results demonstrate that the proposed system can achieve satisfactory performance over other methods. Furthermore, the data equalization operation helps for a better defect detection result.

An end-to-end learning framework is established for defect detection. Although the adopted fully convolutional network has been extensively used for semantic segmentation in images, to the best knowledge of the authors, it has not been used for industrial defect detection. To remedy the difficulties of blade defect detection which has been analyzed above, the authors develop a new network architecture which integrates the residue learning to perform the efficient defect detection. A dual-arm data collection platform is constructed and extensive experimental validation are conducted.

The current popular image processing technologies based on convolutional neural network have the characteristics of large computation, high storage cost and low accuracy for tiny defect detection, which is contrary to the high real-time and accuracy, limited computing resources and storage required by industrial applications. Therefore, an improved YOLOv4 named as YOLOv4-Defect is proposed aim to solve the above problems.

On the one hand, this study performs multi-dimensional compression processing on the feature extraction network of YOLOv4 to simplify the model and improve the feature extraction ability of the model through knowledge distillation. On the other hand, a prediction scale with more detailed receptive field is added to optimize the model structure, which can improve the detection performance for tiny defects.

The effectiveness of the method is verified by public data sets NEU-CLS and DAGM 2007, and the steel ingot data set collected in the actual industrial field. The experimental results demonstrated that the proposed YOLOv4-Defect method can greatly improve the recognition efficiency and accuracy and reduce the size and computation consumption of the model.

This paper proposed an improved YOLOv4 named as YOLOv4-Defect for the detection of surface defect, which is conducive to application in various industrial scenarios with limited storage and computing resources, and meets the requirements of high real-time and precision.

This paper aims to propose a deep learning model that can be used to expand the number of samples. In the process of manufacturing and assembling electronic components on the printed circuit board in the surface mount technology production line, it is relatively easy to collect non-defective samples, but it is difficult to collect defective samples within a certain period of time. Therefore, the number of non-defective components is much greater than the number of defective components. In the process of training the defect detection method of electronic components based on deep learning, a large number of defective and non-defective samples need to be input at the same time.

To obtain enough electronic components samples required for training, a method based on the generative adversarial network (GAN) to generate training samples is proposed, and then the generated samples and real samples are used to train the convolutional neural networks (CNN) together to obtain the best detection results.

The experimental results show that the defect recognition method using GAN and CNN can not only expand the sample images of the electronic components required for the training model but also accurately classify the defect types.

To solve the problem of unbalanced sample types in component inspection, a GAN-based method is proposed to generate different types of training component samples and then the generated samples and real samples are used to train the CNN together to obtain the best detection results.

The purpose of this study is to propose a new method for the end-to-end classification of steel surface defects.

This study proposes an AM-AoN-SNN algorithm, which combines an attention mechanism (AM) with an All-optical Neuron-based spiking neural network (AoN-SNN). The AM enhances network learning and extracts defective features, while the AoN-SNN predicts both the labels of the defects and the final labels of the images. Compared to the conventional Leaky-Integrated and Fire SNN, the AoN-SNN has improved the activation of neurons.

The experimental findings on Northeast University (NEU)-CLS demonstrate that the proposed neural network detection approach outperforms other methods. Furthermore, the network’s effectiveness was tested, and the results indicate that the proposed method can achieve high detection accuracy and strong anti-interference capabilities while maintaining a basic structure.

This study introduces a novel approach to classifying steel surface defects using a combination of a shallow AoN-SNN and a hybrid AM with different network architectures. The proposed method is the first study of SNN networks applied to this task.

This paper aims to realize an in-situ quality inspection system rapidly for new injection molding (IM) tasks via transfer learning (TL) approach and automation technology.

The proposed in-situ quality inspection system consists of an injection machine, USB camera, programmable logic controller and personal computer, interconnected via OPC or USB communication interfaces. This configuration enables seamless automation of the IM process, real-time quality inspection and automated decision-making. In addition, a MobileNet-based deep learning (DL) model is proposed for quality inspection of injection parts, fine-tuned using the TL approach.

Using the TL approach, the MobileNet-based DL model demonstrates exceptional performance, achieving validation accuracy of 99.1% with the utilization of merely 50 images per category. Its detection speed and accuracy surpass those of DenseNet121-based, VGG16-based, ResNet50-based and Xception-based convolutional neural networks. Further evaluation using a random data set of 120 images, as assessed through the confusion matrix, attests to an accuracy rate of 96.67%.

The proposed MobileNet-based DL model achieves higher accuracy with less resource consumption using the TL approach. It is integrated with automation technologies to build the in-situ quality inspection system of injection parts, which improves the cost-efficiency by facilitating the acquisition and labeling of task-specific images, enabling automatic defect detection and decision-making online, thus holding profound significance for the IM industry and its pursuit of enhanced quality inspection measures.

This paper aims to provide a novel integrated fault detection method for industrial process monitoring.

A novel integrated fault detection method based on the combination of Mallat (MA) algorithm, weight-elimination (WE) algorithm, conjugate gradient (CG) algorithm and multi-dimensional Taylor network (MTN) dynamic model, namely, MA-WE-CG-MTN, is proposed in this paper. First, MA algorithm is taken as data pre-processing. Second, in virtue of approximation ability and low computation complexity owing to the simple structure of MTN, MTN dynamic models are constructed for each frequency band. Furthermore, the CG algorithm is used to discipline the model parameters and the outputs of MTN model of each frequency band are gained. Third, the authors introduce the WE algorithm to cut down the number of middle layer nodes of MTN, reducing the complexity of the network. Finally, the outputs of MTN model for each frequency band are superimposed to achieve outputs of MTN model, and fault detection is proceeded by the residual error generator based on the difference between the output of MTN model and the actual output.

The novel proposed method is used to perform fault detection for industrial process monitoring effectively, such as the Benchmark Simulation Model 1 wastewater treatment process.

The novel proposed method has generality and provides considerably improved performance and effectiveness, which is used to perform fault detection for industrial process monitoring. The proposed method has good robustness, low complexity and easy implementation.

This paper aims to propose a new method to solve the two problems in fabric defect detection. Current state-of-the-art industrial products defect detectors are deep learning-based, which incurs some additional problems: (1) The model is difficult to train due to too few fabric datasets for the difficulty of collecting pictures; (2) The detection accuracy of existing methods is insufficient to implement in the industrial field. This study intends to propose a new method which can be applied to fabric defect detection in the industrial field.

To cope with exist fabric defect detection problems, the article proposes a novel fabric defect detection method based on multi-source feature fusion. In the training process, both layer features and source model information are fused to enhance robustness and accuracy. Additionally, a novel training model called multi-source feature fusion (MSFF) is proposed to tackle the limited samples and demand to obtain fleet and precise quantification automatically.

The paper provides a novel fabric defect detection method, experimental results demonstrate that the proposed method achieves an AP of 93.9 and 98.8% when applied to the TILDA(a public dataset) and ZYFD datasets (a real-shot dataset), respectively, and outperforms 5.9% than fine-tuned SSD (single shot multi-box detector).

Our proposed algorithm can provide a promising tool for fabric defect detection.

The paper includes implications for the development of a powerful brand image, the development of “brand ambassadors” and for managing the balance between stability and change.

This work provides technical support for real-time detection on industrial sites, advances the process of intelligent manual detection of fabric defects and provides a technical reference for object detection on other industrial

Therefore, our proposed algorithm can provide a promising tool for fabric defect detection.

In the field of steel defect detection, the existing detection algorithms struggle to achieve a satisfactory balance between detection accuracy, computational cost and inference speed due to the interference from complex background information, the variety of defect types and significant variations in defect morphology. To solve this problem, this paper aims to propose an efficient detector based on multi-scale information extraction (MSI-YOLO), which uses YOLOv8s as the baseline model.

First, the authors introduce an efficient multi-scale convolution with different-sized convolution kernels, which enables the feature extraction network to accommodate significant variations in defect morphology. Furthermore, the authors introduce the channel prior convolutional attention mechanism, which allows the network to focus on defect areas and ignore complex background interference. Considering the lightweight design and accuracy improvement, the authors introduce a more lightweight feature fusion network (Slim-neck) to improve the fusion effect of feature maps.

MSI-YOLO achieves 79.9% mean average precision on the public data set Northeastern University (NEU)-DET, with a model size of only 19.0 MB and an frames per second of 62.5. Compared with other state-of-the-art detectors, MSI-YOLO greatly improves the recognition accuracy and has significant advantages in computational cost and inference speed. Additionally, the strong generalization ability of MSI-YOLO is verified on the collected industrial site steel data set.

This paper proposes an efficient steel defect detector with high accuracy, low computational cost, excellent detection speed and strong generalization ability, which is more valuable for practical applications in resource-limited industrial production.

The defect detection problem of color-patterned fabric is still a huge challenge due to the lack of manual defect labeling samples. Recently, many fabric defect detection algorithms based on feature engineering and deep learning have been proposed, but these methods have overdetection or miss-detection problems because they cannot adapt to the complex patterns of color-patterned fabrics. The purpose of this paper is to propose a defect detection framework based on unsupervised adversarial learning for image reconstruction to solve the above problems.

The proposed framework consists of three parts: a generator, a discriminator and an image postprocessing module. The generator is able to extract the features of the image and then reconstruct the image. The discriminator can supervise the generator to repair defects in the samples to improve the quality of image reconstruction. The multidifference image postprocessing module is used to obtain the final detection results of color-patterned fabric defects.

The proposed framework is compared with state-of-the-art methods on the public dataset YDFID-1(Yarn-Dyed Fabric Image Dataset-version1). The proposed framework is also validated on several classes in the MvTec AD dataset. The experimental results of various patterns/classes on YDFID-1 and MvTecAD demonstrate the effectiveness and superiority of this method in fabric defect detection.

It provides an automatic defect detection solution that is convenient for engineering applications for the inspection process of the color-patterned fabric manufacturing industry. A public dataset is provided for academia.