Application of Shearlet transform to classification of surface defects for metals

Eddy current testing is a nondestructive testing method, which is used to detect discontinuities and defects in conductive materials. Using this technique, two different types of artificial defects in a railhead were evaluated in order to analyze the relationship between different types of defects and eddy current signals, and to obtain data on the size of the rail surface defects and crack location. The actually used rail sample was also studied. Surface cracks and defects were clearly observed as amplitude and phase changes of detected signals. This study succeeds in quantitatively analyzing and discriminating the damage types.

In the machining of precision components, a range of defects, including scratches, pits and uneven materials, will inevitably manifest on the surface of the components. Such defects not only result in a reduction in the quality and precision of the parts, but also have an impact on the overall quality, performance, and longevity of the equipment. It is therefore of particular importance to implement effective detection and repair procedures for surface defects in parts. The implementation of efficient detection methods enables the identification and rectification of defects in a prompt and effective manner, thereby ensuring the reliability and long-term stability of the equipment. In this work, we propose a deep neural network-based model that utilises deep convolutional neural networks (DCNNs) to extract features on the surface of parts, and gradually extracts the spatial and textural features of images through multiple convolutional layers and pooling layers. Subsequently, the aforementioned features are fed into the fully connected layer, where complex feature representations are further learned. Finally, the accurate classification of defect types is achieved through multilayer neural networks. The model is thus capable of effectively capturing and identifying different types of surface defects, thereby improving the accuracy and efficiency of identification. The experimental analysis demonstrates that the model exhibits high accuracy and robustness in the identification of different types of defects, thereby verifying its effectiveness and reliability in practical applications.

In the actual industrial production environment, the surface defects of products are subtle, and the number of different types of defect data samples is also quite small. Most deep learning models rely on a large number of training samples and parameters to achieve high-precision defect detection. At the same time, the edge computing layer in the actual industrial environment may also encounter transmission delays and insufficient resources. Training a proper model for a specific type of surface defect while simultaneously satisfying the real-time accuracy of defect detection is still a challenging task. To effectively deal with the above challenges, we propose an edge-cloud computing defect detection model based on the intrinsic mean feature detector in the Lie Group space. The modules in the model adopt a symmetrical structure, which can extract related features more effectively. Different from existing models, this model utilizes the Lie Group space intrinsic mean feature as a metric to characterize the essential attributes of different types of surface defects. In addition, we propose an intrinsic mean attention mechanism in the Lie Group manifold space that is easy to implement at the edge service layer without increasing the number of model parameters, thereby enhancing the detection performance of tiny surface defects. Extensive experiments on three publicly available and challenging datasets reveal the superiority of our model in terms of detection accuracy, real-time detection, number of parameters, and computational performance. In addition, our proposed model also shows competitiveness and advantages compared with state-of-the-art models.

Surface defects that affect the quality of metals are an important factor. Machine vision systems commonly perform surface inspection, and feature extraction of defects is essential. The rapidity and universality of the algorithm are two crucial issues in actual application. A new method of feature extraction based on contourlet transform and kernel locality preserving projections is proposed to extract sufficient and effective features from metal surface images. Image information at certain direction is important to recognition of defects, and contourlet transform is introduced for its flexible direction setting. Images of metal surfaces are decomposed into multiple directional subbands with contourlet transform. Then features of all subbands are extracted and combined into a high-dimensional feature vector, which is reduced to a low-dimensional feature vector by kernel locality preserving projections. The method is tested with a Brodatz database and two surface defect databases from industrial surface-inspection systems of continuous casting slabs and aluminum strips. Experimental results show that the proposed method performs better than the other three methods in accuracy and efficiency. The total classification rates of surface defects of continuous casting slabs and aluminum strips are up to 93.55% and 92.5%, respectively.

Sanitary ceramic products, such as toilet and wash basin, are widely used in our daily life. Sanitary ceramics are expected to have some excellent physical properties, such as corrosion resistance, easy cleaning, and low water absorption. However, surface defects in sanitary ceramics are inevitable due to complex production processes and changing production environment. Therefore, surface defect detection must be performed in the manufacturing process of sanitary ceramics. There are many types of surface defects in sanitary ceramics, and different types of defects have large differences in characteristics and scales. Traditional detection methods with artificially designed features and classifiers are difficult to apply in this context. In addition, there are few studies on surface defect detection methods of sanitary ceramics based on deep neural networks. In this article, a lightweight real-time defect detection network based on the lightweight backbone MobileNetV3 is presented. The proposed network achieves multi-scale detection of surface defects in sanitary ceramics with a multi-layer feature pyramid. Combining region proposal network and anchor-free method, real-time defect detection is achieved. Finally, a detection head with channel attention structure and a low-level mixed feature classification strategy is used to perform defect classification with higher accuracy. Experimental results show that the proposed approach achieves at least 22.9% detection speed improvement and 35.0% average precision improvement while reducing memory consumption by at least 8.4% compared with the classic one-stage SSD, YOLO V3 and two-stage Faster R-CNN methods.

Ultrasonic guided waves are sensitive to many different types of defects and have been studied for defect recognition in rail. However, most fault recognition algorithms need to extract features from the time domain, frequency domain, or time-frequency domain based on experience or professional knowledge. This paper proposes a new method for identifying many different types of rail defects. The segment principal components analysis (S-PCA) is developed to extract characteristics from signals collected by sensors located at different positions. Then, the Support Vector Machine (SVM) model is used to identify different defects depending on the features extracted. Combining simulations and experiments of the rails with different kinds of defects are established to verify the effectiveness of the proposed defect identification techniques, such as crack, corrosion, and transverse crack under the shelling. There are nine channels of the excitation-reception to acquire guided wave detection signals. The results show that the defect classification accuracy rates are 96.29% and 96.15% for combining multiple signals, such as the method of single-point excitation and multi-point reception, or the method of multi-point excitation and reception at a single point.

Surface defect detection aims to classify and locate a certain defect that exists in the target surface area. It is an important part of industrial quality inspection. Most of the research on surface defect detection are currently based on convolutional neural networks (CNNs), which are more concerned with local information and lack global perception. Thus, CNNs are unable to effectively extract the defect features. In this paper, a defect detection method based on the Swin transformer is proposed. The structure of the Swin transformer has been fine-tuned so that it has five scales of output, making it more suitable for defect detection tasks with large variations in target size. A bi-directional feature pyramid network is used as the feature fusion part to efficiently fuse to the extracted features. The focal loss is used as a loss function to weight the hard- and easy-to-distinguish samples, potentially making the model fit the surface defect data better. To reduce the number of parameters in the model, a shared detection head was chosen for result prediction. Experiments were conducted on the flange surface defect dataset and the steel surface defect dataset, respectively. Compared with the classical CNNs target detection algorithm, our method improves the mean average precision (mAP) by about 15.4%, while the model volume and detection speed are essentially the same as those of the CNNs-based method. The experimental results show that our proposed method is more competitive compared with CNNs-based methods and has some generality for different types of defects.

Surface defects are an inevitable problem in steel production and processing. To solve the problems of error detection, leakage detection and low accuracy of traditional manual methods for steel surface defect detection, this paper proposes a steel surface defect detection algorithm based on Faster RCNN, a classical model in the field of computer vision object detection. The proposed model uses ResNet50 as its backbone and introduces FPN to fuse the multi-layer feature maps of the backbone to improve the detection capability for defects of different scales. The experimental results on NEU-DET dataset show that the proposed model in this paper has different detection accuracies for six different types of defects in the dataset, and the overall mAP reaches 0.708. In addition, the model execution speed reaches 43.3 frame/s, which meets the applications in industrial scenarios.

The surface defect detection of aluminum sheet is of great significance to ensure the appearance and quality of aluminum sheet. The surface defects of aluminum sheets have the characteristics of different shapes, obvious size differences, and difficult to obtain defect samples, which make defect detection challenging. In order to solve this problem, we make the following improvements to YOLOv3: Adding attention mechanism modules after the three feature layers output by the model backbone and after neck upsampling; Freezing the model backbone and using pretrained for transfer learning. The proposed YOLOv3 + ECA model is compared with the target detection models such as YOLOv3 and Faster-RCNN. It is found that the mAP of our model reaches 96.22%, which is higher than the current conventional algorithm. The AP values for different types of defects have good detection results.

Chromium doped aluminum nitride (AlN: Cr) thin films were grown on silicon, glass and copper substrates by DC and RF magnetron sputtering co-deposition. After growth, thin films on silicon substrates were annealed at 1373 K for 30 min in N2 atmosphere. The AlN: Cr thin films were characterized by x-ray diffraction for structural analysis, by FS5 spectrofluorometer for the study of photoluminescence, absorption, transmission, and chromaticity. As-deposited and annealed silicon substrate and as-deposited glass substrate thin films of AlN: Cr exhibited intense photoluminescence emission in the range of 400 to 679.5 nm. Spectral evidence demonstrated conclusively that the AlN: Cr thin films on as-deposited glass substrate and annealed silicon substrate have excellent photoluminescence emission which is due to both AlN (host) and Cr3+ ions. The reasons of photoluminescence of AlN in the visible region are surface defects and impurities. Impurities become the cause to produce different types of defects and vacancies just like oxygen point defects (O+N), nitrogen vacancies (VN) and various defect complexes (V3-Al – 3 O+N). It may also be due to the recombination of photogenerated hole with the electron occupied by the nitrogen vacancies and due to the transition between deep level of (V3-Al – 3 O+N) defect complexes and shallow level of VN and the reason behind the photoluminescence of Cr3+ ions is due to vibrational energy levels 4T1 and 4T2 and due to 4T1→4A2 and 4T2→4A2 transitions. AlN: Cr thin films can give better results in the applications like light emitting diodes (LEDs), laser diodes (LDs), field emission displays, microelectromechanical system (MEMS), optical MEMS and biomedical applications. Key words: III-V Semiconductor Material, Thin films, Photoluminescence Mechanism

This study was carried out with the aim of detecting the defects on the surfaces of metallic products, which are frequently encountered in our daily lives and widely used in industry, with deep learning. Many metal products in the industry undergo different processes during the production phase. As a result of these processes, the detection of defects such as breakage, cracking, etc. on the surfaces of metal products is carried out by quality control personnel or production personnel. However, this error detection made by manpower both slows down production and causes overlooked errors. In order to easily detect these defects, in our study, we used a dataset consisting of 285 images in total with images of two different surface defects and images containing flawless metal parts and a CNN architecture, ResNet50 architecture, to defects. There are three classes in total in the dataset. Two of these classes consist of images of different types of defects on the surface of the metal piston part used in air conditioners, and one of them consists of images of perfect metal piston parts. Convolutional Neural Network (CNN) method was used to determine the features of the images. Precision, recall, F1-Score and accuracy metrics were used to measure the performance of the model. With the ResNet50 architecture, defects on the surfaces of metal piston parts are detected quickly and with high accuracy. As a result of the study, it was suggested that the proposed model can detect surface defects that occur in the usage areas of metal products in various sectors more quickly and accurately using deep learning. This shows that the quality control problems experienced in the industry can be reduced by using deep learning, saving time and manpower.

With rapidly growing photoconversion efficiencies, hybrid perovskite solar cells have emerged as promising contenders for next generation, low-cost photovoltaic technologies. Yet, the presence of nanoscale defect clusters, that form during the fabrication process, remains critical to overall device operation, including efficiency and long-term stability. To successfully deploy hybrid perovskites, we must understand the nature of the different types of defects, assess their potentially varied roles in device performance, and understand how they respond to passivation strategies. Here, by correlating photoemission and synchrotron-based scanning probe X-ray microscopies, we unveil three different types of defect clusters in state-of-the-art triple cation mixed halide perovskite thin films. Incorporating ultrafast time-resolution into our photoemission measurements, we show that defect clusters originating at grain boundaries are the most detrimental for photocarrier trapping, while lead iodide defect clusters are relatively benign. Hexagonal polytype defect clusters are only mildly detrimental individually, but can have a significant impact overall if abundant in occurrence. We also show that passivating defects with oxygen in the presence of light, a previously used approach to improve efficiency, has a varied impact on the different types of defects. Even with just mild oxygen treatment, the grain boundary defects are completely healed, while the lead iodide defects begin to show signs of chemical alteration. Our findings highlight the need for multi-pronged strategies tailored to selectively address the detrimental impact of the different defect types in hybrid perovskite solar cells.

Automatic surface defect segmentation for hot-rolled steel strip using depth-wise separable U-shape network

Surface Detection of Continuous Casting Slabs Based on Curvelet Transform and Kernel Locality Preserving Projections

Hot-rolled steel bars and other hot-rolled steel shapes are produced from ingots, blooms, or billets converted from ingots or from strand cast blooms or billets and comprise a variety of sizes and cross sections. Most carbon steel and alloy steel hot-rolled bars and shapes contain surface imperfections with varying degrees of severity. Seams, laps, and slivers are probably the most common defects in hot-rolled bars and shapes. Another condition that could be considered a surface defect is decarburization. Hot-rolled steel bars and shapes can be produced to chemical composition ranges or limits, mechanical property requirements, or both. Hot-rolled carbon steel bars are produced to two primary quality levels: merchant quality and special quality. Merchant quality is the least restrictive descriptor for hot-rolled carbon steel bars. Special quality bars are employed when end use, method of fabrication, or subsequent processing treatment requires characteristics not available in merchant quality bars.