An Automatic Annotation Method for Defective Cigarette Package Appearance Images Based on the Improved YOLOv8n Model

Accurate detection of surface defects for steel is essential to improve surface quality and service life. Deep learning (DL) used in steel surface defect detection can solve the problems of low efficiency and poor accuracy of traditional manual detection. The classic YOLOv5 as a DL method is used to accomplish defect detection tasks without attention mechanisms, resulting in a loss of global information. Besides, it is difficult to complete complex network detection tasks with low-configuration hardware, especially for surface defects with complex defect types and variable defect sizes. To solve these issues, this paper introduces an improved global feature reuse and hardware-aware YOLOv5 by using BoTNet, RepGhost, and EfficientRep model (BGE-YOLOv5). The multi-head self-attention layer is used to obtain global information and only part of the convolutional layers is replaced to avoid the excessive computational cost. The RepGhost model is introduced to extract the remaining feature information for feature reuse. EfficientRep is used to replace the original structure to achieve hardware-aware and to balance the detection veracity and efficiency. The distance IOU is replaced by SCYLLA-IOU to accelerate the iteration and improve stability. The results of the framework on the surface defect database (NEU-DET) show that BGE-YOLOv5 achieves a mean average precision of 79.5%, which is 10.3% greater than the baseline. The proposed BGE-YOLOv5 has a better performance in steel surface defect detection.

In modern industrial production, quality inspection and control play a crucial role in enhancing production efficiency and ensuring product reliability. Traditional inspection methods primarily rely on manual labor, which not only results in low efficiency but also introduces human error, leading to unstable inspection outcomes. With the advancement of science and technology, automated image processing technology has emerged, revolutionizing industrial testing. By leveraging computer vision and image analysis algorithms, automated image processing technology can rapidly and accurately identify and analyze defects and anomalies in images, significantly improving inspection efficiency and accuracy. This technology not only enables real-time monitoring of product quality on production lines but also provides data support for process optimization, driving industrial production toward intelligent and automated development.

In practical industrial inspection scenarios, the challenges of limited samples and severe domain shift render achieving reliable few-shot image classification a pressing challenge. In response to this, we propose a lightweight small-sample industrial quality inspection image classification and confidence calibration framework based on Fast AutoAugment. The framework employs EfficientNetV2 as the basic feature extractor, using an advanced policy search mechanism enabled by Fast AutoAugment, which successfully enhances the model’s generalization ability to defect features. In relation to feature fusion, we design a multi-branch feature fusion module that combines spatial attention, channel recalibration and contrast alignment mechanisms to further promote discriminant expression in small sample conditions. Additionally, we design an uncertainty calibration module based on Dirichlet distribution, which integrates task-adaptive priors and temperature scaling strategies to effectively calibrate the model’s confidence and reduce the risk of "overconfidence errors" in industrial settings. Experiments on the NEU-CLS industrial quality inspection dataset show that the proposed method outperforms the existing Few-shot baseline, demonstrating a classification accuracy that is 4.6%-7.1% higher than that of the existing few-shot baselines.

Industrial automatic fabric inspection system, a critical technology in the industry, enhances both total production quantity and quality compared to conventional inspection techniques. This study aims to create a reliable and effective real-time automated visual inspection system for fabrics, focusing on defect detection. The goals of the study can be stated as; installing a system with advanced technology for capturing and processing images swiftly, the development and deployment of a system capable of autonomously learning and scanning fabrics in use, and the creation of a smart framework for accurate fabric defect detection and classification. We focus on the development of unsupervised fabric defect detection using a convolutional autoencoder model, and defect classification using a convolutional neural network model, which takes input as the feature vector generated by the convolutional autoencoder. The experimental outcomes have displayed significant success rates in both detecting defects and classifying them, confirming the effectiveness of the framework in real-time visual inspection systems.

As industrial production becomes more modern and intelligent today, the inspection of product quality of the workshop is becoming more and more accustomed to replacing the old manual visual inspection methods with automated inspection systems. In the welding field, automated welding robots are not only used in traditional large-scale automobile assembly lines. In more general welding work, welding robots also plays an important role. The inspection of the welding quality of the welding robot is mainly to detect the four main types of weld defects. Compared to traditional defect classification based on support vector machines and defect detection based on template matching, this paper uses a welding surface defect detection system designed based on deep learning methods. By working with workshop welding experts, a large-scale image of nearly 5000 pictures is built. Large-scale weld defect datasets, while using the real-time and accuracy of the YOLO series of deep learning object detection frameworks, the weld defects detection model reaches 75.5% mean average precision(mAP) in constructed weld defect data set. In addition, the construction cost of the detection model and the deployment time of the detection system are greatly reduced. During the field test of the system in the workshop, among a batch of welding workpieces provided by the factory, the detection accuracy of weld defects reached 71%, which initially met the requirements of the workshop for an automated defect detection system.

In industrial quality inspection, large amounts of data on the desired product appearance are available at the time of training, while significantly few defective samples are available. In this study, we proposed new memory-augmented adversarial autoencoders to detect and localize defects in real-time using defect-free samples alone for model training. This research was conducted by reconstructing images using an adversarial autoencoder and detection results from the Fréchet Markov distance (FMD). A threshold was determined based on the statistical characteristics of defect-free samples in the training set. Innovatively, we introduced a memory module and redesigned the reconstruction loss function to avoid the situation where the reconstruction ability is too strong or too poor, which lead to missing detection of defects. Then we proposed FMD, which can accurately measure the distance between the distribution of test samples and positive samples. Moreover, the statistics-based threshold determination method is used to meet different industrial needs. The accuracy, robustness, and computational overhead of the proposed MAA were evaluated using three datasets obtained from the production line and two benchmark datasets. The results indicated the effectiveness and ability of the proposed method to adapt to the real-time nature of industrial production.

To improve the performance of the adhesive layer defect detection (DD) planar array capacitor system, an adhesive layer DD method based on particle swarm optimization and fuzzy C-means (FCM) clustering algorithm is proposed. By classifying erroneous capacitance data, the experiment distinguishes the capacitance values at defect locations from those at defect-free locations. In response to the presence of artifacts in the image reconstruction results, the optimal threshold method is used to post-process the reconstructed images using the mixed regularization method. These experiments confirmed that FCM clustering reduced the mean absolute deviation (MAD) of data by about 6.1%, and the proposed clustering algorithm reduced the MAD of data by about 23.2%. Compared to the Tikhonov method, the mixed regularization method reduced the relative error of reconstructed images by about 13.1% and increased the correlation coefficient of reconstructed images by about 21.6%. Therefore, this research method had significant effects in reducing data fluctuations and errors, effectively improving the image reconstructing quality and the accuracy of DD. This study can effectively improve the accuracy of adhesive layer DD, help reduce production costs, and improve production efficiency.

Automated peanut defect detection using hyperspectral imaging and deep learning: A real-time approach for smart agriculture

As power grid structure becomes increasingly complex, the impact of the operation status of key substation equipment on the safe and stable operation of the power grid is gradually increasing. Aiming at the problems of dense targets and occlusion recognition difficulties in the complex background of substations, an improved you-only-look-once (YOLO) v4 insulator defect detection algorithm is proposed. A large number of insulator and defect datasets were collected from the Internet. We use MobileNet-v2 to replace the backbone network of YOLOv4, and in view of the problem that the accuracy of YOLOv4 target detection model is not high for small targets or dense targets, convolutional block attention module (CBAM) attention mechanism is added to the backbone network to enhance the feature extraction ability, and ultra-lightweight subspace attention module (ULSAM) is added to the detection head to learn the cross-channel information. Experiments based on InDataset show that the insulator defect detection algorithm based on deep learning proposed achieves a high detection accuracy and detection speed; the mean average precision of the improved YOLOv4 model is 87.48%, and the recall rate of insulator defects reaches 79.84%, which meets the requirements of robustness and accuracy of insulator defect detection in real scenarios.

Laser scanning systems are well established in the world of fast industrial in-process quality inspection systems. The materials inspected by laser scanning systems are e.g. "endless" sheets of steel, paper, textile, film or foils. The web width varies from 50 mm up to 5000 mm or more. The web speed depends strongly on the production process and can reach several hundred meters per minute. The continuous data flow in one of different channels of the optical receiving system exceeds ten Megapixels/sec. Therefore it is clear that the electronic evaluation system has to process these data streams in real time and no image storage is possible. But sometimes (e.g. first installation of the system, change of the defect classification) it would be very helpful to have the possibility for a visual look on the original, i.e. not processed sensor data. At first we show the principle set up of a standard laser scanning system. Then we will introduce a large image memory especially designed for the needs of high-speed inspection sensors. This image memory co-operates with the standard on-line evaluation electronics and provides therefore an easy comparison between processed and non-processed data. We will discuss the basic system structure and we will show the first industrial results.

Injection molded parts are increasingly utilized across various industries due to their cost-effectiveness, lightweight nature, and durability. However, traditional defect detection methods for these parts often rely on manual visual inspection, which is inefficient, expensive, and prone to errors. To enhance the accuracy of defect detection in injection molded parts, a new method called MRB-YOLO, based on the YOLOv8 model, has been proposed. This method introduces several key improvements: (1) the MAFHead, a four-detection head based on multiplicative feature fusion, which replaces the original detection head to enhance feature representation; (2) the RepGFPN-SE module, a re-parameterized generalized feature pyramid network that improves detection of small objects by replacing the original C2f. module; (3) and the BiNorma module, employing a bi-level routing attention mechanism to optimize the training process by reducing input distribution changes across layers. The effectiveness of the MRB-YOLO model was validated through ablation and contrast experiments using a specially constructed dataset of injection molded parts defects. The results demonstrated an accuracy of 88.8%, a recall rate of 86.8%, and a mean average precision (mAP) of 91.5%. Compared to the YOLOv8n model, the MRB-YOLO model shows an increase in accuracy by 8.2%, in recall rate by 17.2%, and in mAP by 11.8%. These findings confirm that the MRB-YOLO model meets the requirements for accurate detection of defects in injection molded parts.

Due to the harsh service environment, oil and gas pipelines are susceptible to corrosion defects. Unmanned aerial vehicle (UAV) vision technology is a non-destructive testing method for in-service pipelines. Nevertheless, it is challenging to provide an accurate defect detection under environmental strong interference. Additionally, limited data can reduce the detection accuracy and model generalisation capability. In this paper, a cross-domain defect detection method for oil and gas pipelines is proposed based on UAV vision technology. Firstly, an experimental setup with the UAV inspection is designed to collect the pipeline images. To alleviate domain shift posed by variations in working conditions, an image preprocessing module based on adaptive Retinex algorithm combined with Unet and attention mechanism is developed. Then, a cross-domain detection module that leverages residual autoencoder and pseudo-label strategy is established to achieve accurate defect detection under limited labelled data. Finally, the comparative experiments are conducted to verify the domain generalisation performance for strong interference of low light and background noise. The experimental results demonstrate that the proposed method outperforms state-of-the-art methods in terms of cross-domain transfer tasks. This study makes a significant contribution to enhance the preventive maintenance of oil and gas pipelines.

In this paper, we propose an accurate automatic defect detection method using the quadtree decomposition on scanning electron microscope (SEM) images. The proposed method consists of five steps including alignment between the inspection and reference images, edge relaxation, defect candidate detection, multichannel integration, and defect scoring. Since the periodicity of semiconductor SEM images is usually high, alignment can be failed due to a lot of local minima. To resolve this problem, we propose the enhancement method of non-periodic areas in SEM images. Then, we propose an edge relaxation technique to minimize the falsely remained edges in the aligned and subtracted image due to different focusing and exposure between two images. Then, the defect candidates are detected by analyzing the inhomogeneity based on quadtree decomposition, which is enhanced by our outlier-robust decomposition rule and adaptive decomposition thresholding to prevent over/under decomposition. Whereas the conventional defect detection system focuses only on the intensity of defects, our method considers both the inhomogeneity and intensity of defects. Field test results demonstrated that the proposed method provided more accurate defect detection with faster speed as compared to the conventional system. The proposed method can much reduce the automatic defect review rework compared with the conventional system.

The timely detection of insulator defects in transmission lines is vital for ensuring social production and people’s livelihoods. Aiming to solve the problem of the low accuracy of insulator defect detection in current detection models, this study improves the YOLO11n model and proposes an insulator defect detection model, FPFS-YOLO, that integrates FasterNet and an attention mechanism. In this study, to mitigate parameter redundancy in the backbone of the YOLO11n model, the FasterNet lightweight network was introduced, and some convolution was embedded into the shallow network to enhance its feature extraction ability. To solve problems such as insufficient attention to important features and the low detection ability of small defects in the YOLO11n model network, the ParNet attention mechanism was added, along with a small-defect detection layer, which improved the detection accuracy of the model. Finally, in order to alleviate the computational redundancy caused by these additions, the C3k2_faster module and the PSP-Head detection head were introduced. These amendments further improved the accuracy of the model network in detecting insulator defects while simultaneously reducing its computational redundancy. The experimental results show that the improved FPFS-YOLO model achieved a 91.5% mAP@50 and a 56.6% mAP@0.5-0.95, increases of 3.1% and 1.2%, respectively, while the precision and recall reached 93.2% and 86.4%, increases of 1.5% and 4.2%, respectively. The FPFS-YOLO model achieved a higher detection accuracy than the YOLO11n model and thus could be widely applied in the detection of insulator defects.

Fluorescent Penetrant Inspection (FPI) is a widely used Non-Destructive Testing (NDT) method in the aerospace industry to inspect precision casting components. Currently, FPI inspection relies on visual examination, which can lead to challenges in distinguishing between defects and false indications. Moreover, human factors introduce variability in the results, impacting the consistency and reliability of the inspection process. This highlights the desirability of the automation of FPI to increase consistency, reliability and productivity. The deep learning method is gradually replacing the traditional approaches that involve image processing and machine learning classifiers in automated defect detection system. Deep learning method offers automatic feature extraction and high robustness, which contribute to more accurate and efficient defect detection. The use of various convolutional neural networks (CNN) in defect detection for flat superalloy plates processed with FPI and photographed to create digital images was investigated. Among the CNN models, MobileNetV2 exhibited outstanding performance, with a remarkable recall rate of 99.2% and an accuracy of 99.2%. Additionally, the effect of dataset imbalance on model performance was carefully examined. Moreover, the features extracted by the model are visualized using Guided Grad-CAM to reveal the attention of the CNN model to the fluorescent display features. These results underscore the strong capability of deep learning architectures in detect defects in aerospace precision casting components, paving the way for the automation of the entire FPI process.