False Defects Research Articles

40‐4: Improving Visibility Coherence between Auto Macro Inspection and Auto Visual Inspection Using AI Image Translation

In OLED production, the glass fabrication process is prone to frequent occurrences of persistent mura defects, leading to significant damage in the event of defects, as the subsequent final inspection, AVI (Auto Visual Inspection), is time‐consuming. Therefore, there is a need for a consistent inspection. Currently, manual inspection by operators introduces variations in inspection criteria among individuals, resulting in ongoing challenges related to false judgments and post‐process leakage issues The invested automated inspection system for quantitative determination is utilized as a reference only, as the distinction between pseudo‐defects deemed acceptable in AVI and true defects identified as faulty in AVI poses limitations with the traditional logic‐based inspection methods. We enhanced the mura visibility of automatic inspection system (Auto Macro) images through pre‐processing, followed by translation into an image environment similar to AVI images using Pix2pix GAN. Subsequently, we compared the mura index extracted from the processed images with the AVI mura index for evaluation. The Auto Macro images transformed by Pix2pix GAN exhibited similarities with AVI images in terms of overall luminance, gray distribution, and mura visibility intensity. Consequently, the R‐Squared correlation between the mura index of Auto Macro and AVI improved from 0.00 to 0.68. Additionally, defects unrecognized in AVI during Auto Macro inspection were automatically eliminated. False defects in the Auto Macro test that were not recognized in AVI were automatically removed. In this study, we applied the AI model to the Auto Macro inspection system to overcome the limitations of logic‐based inspections. We proposed a method to proactively detect defect‐induced muras before cell‐level processing, aiming to improve yield and reduce incidents of customer leakage accidents.

Road marking defect detection based on CFG_SI_YOLO network

Existing road marking detection primarily focuses on the direction, position, and color of road markings. However, clear and accurate road markings directly impact issues such as directional guidance, lane selection, speed limits, and parking positions. Therefore, we introduce the CFG_SI_YOLO model for road marking defect detection. This model introduces a multi-channel CoordConv module, which enhances the network's focus on fine details based on the distribution characteristics of road markings defect. This helps prevent the loss of road marking information caused by model compression and pooling operations; Moreover, the model introduces the Focal-EIoU loss function to address the issue of imbalanced samples between easy and difficult cases. Additionally, the GELU activation function is incorporated to prevent gradient explosions, enhance the network's non-linear expressiveness, and improve the detection accuracy of the model. Finally, we add a similarity attention module to enhance the network's focus on the target, reduce interference from other objects, and mitigate false detection defects caused by inter-class similarity. Experiments conducted on a self-made dataset containing various types of road markings have shown that our approach achieved Precision, Recall, F1, IoU, and mAP of 85.7%, 85.8%, 85.7%, 75.1% and 82.8%, respectively. These results are significantly better than other methods, confirming the effectiveness and feasibility of our approach. Our code and results can be found on https://github.com/ly6660/Road_marking_line_code_data.

Effect of Temperature on Penetration of Test Liquid into Boiler Pipe

Limitations of time in conducting superheater inspections during overhaul, allowing inspectors to conduct liquid penetrant testing activities above pre-determined temperatures. This results in results obtained from the superheater inspection tube boiler indicating a false defect. This study aims to determine the optimal temperature and the impact of testing above the temperature that has been determined in ASME boiler and pressure vessel code using liquid penetrant testing method. This study uses 3 different penetrant brands and test specimens in the form of a superhater tube boiler. Spraying of penetrant fluid is carried out at normal temperature of 30–52oC and above normal temperature is 53–65 oC. The best crack and easy to read image results are tested at 40 °C A, B and C. For testing at temperature 53–65 oC crack thickness level is not clear and hard to read.

The Core Industry Manufacturing Process of Electronics Assembly Based on Smart Manufacturing

This research takes a case study approach to show the development of a diverse adoption and product strategy distinct from the core manufacturing industry process. It explains the development status in all aspects of smart manufacturing, via the example of ceramic circuit board manufacturing and electronic assembly, and outlines future smart manufacturing plans and processes. The research proposed two experiments using artificial intelligence and deep learning to demonstrate the problems and solutions regarding methods in manufacturing and factory facilities, respectively. In the first experiment, a Bayesian network inference is used to find the cause of the problem of metal residues between electronic circuits through key process and quality correlations. In the second experiment, a convolutional neural network is used to identify false defects that were overinspected during automatic optical inspection. This improves the manufacturing process by enhancing the yield rate and reducing cost. The contributions of the study built in circuit board production. Smart manufacturing, with the application of a Bayesian network to an Internet of Things setup, has addressed the problem of residue and redundant conductors on the edge of the ceramic circuit board pattern, and has improved and prevented leakage and high-frequency interference. The convolutional neural network and deep learning were used to improve the accuracy of the automatic optical inspection system, reduce the current manual review ratio, save labor costs, and provide defect classification as a reference for preprocess improvement.

Don't overthink it: The paradoxical nature of expertise for the detection of errors in conceptual business process models.

Business process models are widely used artifacts in design activities to facilitate communication about business domains and processes. Despite being an extensively researched topic, some aspects of conceptual business modeling are yet to be fully explored and understood by academicians and practitioners alike. We study the attentional characteristics specific to experts and novices in a semantic and syntactic error detection task across 75 Business Process Model and Notation (BPMN) models. We find several intriguing results. Experts correctly identify more error-free models than novices, but also tend to find more false positive defects. Syntactic errors are diagnosed faster than semantic errors by both groups. Both groups spend more time on error-free models. Our findings regarding the ambiguous differences between experts and novices highlight the paradoxical nature of expertise and the need to further study how best to train business analysts to design and evaluate conceptual models.

Detecting Compiler Warning Defects Via Diversity-Guided Program Mutation

Compiler diagnostic warnings help developers identify potential programming mistakes during program compilation. However, these warnings could be erroneous due to the defects of compiler warning diagnostics. Although the existing technique (i.e., Epiphron) can automatically generate test programs for compiler warning defect detection, the effectiveness of Epiphron on defect-finding is still limited, due to the limitation for generating warning-sensitive test program structures. Therefore, in this paper, we propose a DIversity-guided PROgram Mutation approach, called DIPROM, to construct diverse warning-sensitive programs for effective compiler warning defect detection. Given a seed test program, DIPROM first removes its dead code to reduce false positive warning defects. Then, the abstract syntax tree (AST) of the test program is constructed; DIPROM iteratively mutates the structures of the AST to generate warning-sensitive program variants. To effectively construct diverse warning-sensitive structures, DIPROM applies a novel diversity-guided strategy to generate program variants in each iteration. With the generated program variants, differential testing is conducted to detect warning defects in different compilers. In the experiments, we evaluate DIPROM with two popular C compilers (i.e., GCC and Clang). Experimental results show that DIPROM significantly outperforms three state-of-the-art approaches (i.e., HiCOND, Epiphron, and Hermes) by up to 18.93% <inline-formula><tex-math notation="LaTeX">$\sim$</tex-math></inline-formula> 76.74% in terms of the bug-finding capability on average. Meanwhile, DIPROM is efficient, which spends less time on finding the same average number of warning defects. We at last applied DIPROM to the latest development versions of GCC and Clang. After two months’ running, we reported 8 new warning defects; 5 of them have been confirmed/fixed by developers.

Visual patterns of breast attenuation artifacts in women and men with an upright and supine cadmiun-zinc-telluride camera

BackgroundBreast attenuation artifacts occurring with upright cadmium-zinc-telluride (CZT) cardiac imaging systems have not been well characterized. Methods216 consecutive patients with Single Photon Emission Computerized Tomography myocardial perfusion imaging and no angiographically significant obstructive coronary artery disease were identified. All upright and supine SPECT images as well as coronary angiograms were reviewed and analyzed in blinded fashion. ResultsIn women imaged upright, more visual false positive defects were noted in the inferior wall compared to the anterior wall (26 vs. 10 at rest, p = 0.006, and 33 vs. 13 at stress, p < 0.001). Visual inferior wall defects were more common in the upright than supine position at stress (33 vs. 23, p = 0.018) and rest (26 vs. 14, p = 0.011), and most apparent in non-obese women (13 vs. 8, at stress, p = 0.059 and 11 vs. 5, at rest, p = 0.014). ConclusionsWith upright CZT myocardial perfusion imaging, women often have visible inferior wall attenuation artifact defects, likely from pendant breast tissue. These inferior wall attenuation artifacts may be seen in non-obese female patients.

Особенности оценки сплошности свай методом сейсмоакустической дефектоскопии

Introduction. The authors present the results of pile continuity studies conducted using seismoacoustic defectoscopy. Tests were conducted in the Rostov region and Moscow. Reflectograms of tested piles, as well as the subsequent interpretation and confirmation of the design length and continuity of the pile shaft are provided. Materials and methods. The research results were obtained using Spektr 4.3, whose operating principle is based on the theory of elastic wave propagation. This device is used for the recording of the pile response to external impacts with the preset pulse parameters. The signal is recorded by means of an accelerometer sensor, connected to a special computer with the software for the digitization and interpretation of incoming information. Data processing results are presented in the form of computer graphic images (reflectograms), which show the pile length, continuity, damages (if any) and their location. Results. The dependence between soil homogeneity and signal purity has been identified. At the interface between two types of soils that are different in density, the sound wave was reflected and partially passed further, creating characteristic peaks of false defects. To eliminate false defects, studies of layered soils are necessary. Conclusions. The results confirm the effect of the soil homogeneity on signal purity. The higher the heterogeneity of the soil structure, the bigger the number of false peaks that the reflectogram demonstrates. A prerequisite for the correct interpretation of the results of pile continuity testing is the study of the geological section and preliminary instrumental identification of the ultrasound propagation velocity for each individual pile.

Effectively Detecting Left Bundle Branch Block False Defects in Myocardial Perfusion Imaging (MPI) with a Convolutional Neural Network (CNN).

In this work, we evaluated the ability of a CNN-based solution, using transfer learning, to produce an expert-like judgment in recognizing LBBB false defects. We collected retrospectively, MPI polar maps, of patients having small to large fixed anteroseptal perfusion defect. Images were divided into two groups. The LBBB group included patients where this defect was judged as false defect by two experts. The LAD group included patients where this defect was judged as a true defect by two experts. We used a transfer learning approach on a CNN (ResNet50V2) to classify the images into two groups. After 60 iterations, the reached accuracy plateau was 0.98, and the loss was 0.19 (the validation accuracy and loss were 0.91 and 0.25, respectively). A first test set of 23 images was used (11 LBBB, and 12 LAD). The empiric ROC (Receiver operating characteristic) Area was estimated at 0.98. A second test set (18x2 images) was collected after the final results. The ROC area was estimated again at 0.98. Artificial intelligence, using CNN and transfer learning, could reproduce an expert-like judgment in differentiating between LBBB false defects, and LAD real defects.

Optimization of Character Image Matching based on Artificial Bee Colony Algorithm

When template matching is used to detect printing defects in industrial production, the characters to be tested are not consistent with the main structure of template characters, which leads to detection errors and false defects. In order to solve this problem, this paper proposes a character image matching optimization based on artificial bee colony algorithm, using the improved artificial bee colony algorithm to find the optimal parameter model and optimize the parameter model, so that the character image matching effect is the best. The experimental results show that the improved artificial bee colony algorithm can obtain good image registration results in a relatively short time.

An Improved VGG19 Transfer Learning Strip Steel Surface Defect Recognition Deep Neural Network Based on Few Samples and Imbalanced Datasets

The surface defects’ region of strip steel is small, and has various defect types and, complex gray structures. There tend to be a large number of false defects and edge light interference, which lead traditional machine vision algorithms to be unable to detect defects for various types of strip steel. Image detection techniques based on deep learning require a large number of images to train a network. However, for a dataset with few samples with category imbalanced defects, common deep learning neural network training tasks cannot be carried out. Based on rapid image preprocessing algorithms (improved gray projection algorithm, ROI image augmentation algorithm) and transfer learning theory, this paper proposes a set of processes for complete strip steel defect detection. These methods achieved surface rapid screening, defect feature extraction, sample dataset’s category balance, data augmentation, defect detection, and classification. Through verification of the mixed dataset, composed of the NEU surface dataset and dataset in this paper, the recognition accuracy of the improved VGG19 network in this paper reached 97.8%. The improved VGG19 network performs slightly better than the baseline VGG19 in six types of defects, but the improved VGG19 performs significantly better in the surface seams defects. The convergence speed and accuracy of the improved VGG19 network were taken into account, and the detection rate was greatly improved with few samples and imbalanced datasets. This paper also has practical value in terms of extending its method of strip steel defect detection to other products.

Blockchain in manufacturing quality control: A computer simulation study.

Blockchain has been applied to quality control in manufacturing, but the problems of false defect detections and lack of data transparency remain. This paper proposes a framework, Blockchain Quality Controller (BCQC), to overcome these limitations while fortifying data security. BCQC utilizes blockchain and Internet-of-Things to form a peer-to-peer supervision network. This paper also proposes a consensus algorithm, Quality Defect Tolerance (QDT), to adopt blockchain for during-production quality control. Simulation results show that BCQC enhances data security and improves defect detections. Although the time taken for the quality control process increases with the number of nodes in blockchain, the application of QDT allows multiple inspections on a workpiece to be consolidated at a faster pace, effectively speeding up the entire quality control process. The BCQC and QDT can improve the quality of parts produced for mass personalization manufacturing.

Detection of Surface Defects in Ceramic Tiles With Complex Texture

To solve the problem of false defect detection owing to the interference of the texture attribute of ceramic tiles, a method for detecting surface defects in complex-textured ceramic tiles is proposed. Based on the visual detection principle of ceramic tile surfaces, an image acquisition system is established to obtain the ceramic tile image. After image segmentation and correction, the surface defects are preliminarily detected using a saliency detection method. Then, the image sub-block containing the defect area is cut out for secondary detection. The false defects are eliminated, and the final detection of the ceramic tile surface defects is completed using the defect determination method. The feasibility and effectiveness of the defect detection method are studied via comparative experiments. Experimental results show that the maximum accuracy rate of the proposed defect detection method is 98.75%, which satisfies the actual detection requirements and confirms the practical significance of the proposed method.

Effectiveness of specularity removal from hyperspectral images on the quality of spectral signatures of food products

Specularity or highlight problem exists widely in hyperspectral images, provokes reflectance deviation from its true value, and can hide major defects in food objects or detecting spurious false defects causing failure of inspection and detection processes. In this study, a non-iterative method based on the dichromatic reflection model and principle component analysis (PCA) was proposed to detect and remove specular highlight components from hyperspectral images acquired by various imaging modes and under different configurations for numerous agro-food products. To demonstrate the effectiveness of this approach, the details of the proposed method were described and the experimental results on various spectral images were presented. The results revealed that the method worked well on all hyperspectral and multispectral images examined in this study, effectively reduced the specularity and significantly improves the quality of the extracted spectral data. Besides the spectral images from available databases, the robustness of this approach was further validated with real captured hyperspectral images of different food materials. By using qualitative and quantitative evaluation based on running time and peak signal to noise ratio (PSNR), the experimental results showed that the proposed method outperforms other specularity removal methods over the datasets of hyperspectral and multispectral images.

A coarse-to-fine method for glass fiber fabric surface defect detection

Surface defect detection of glass fiber fabrics has always been one of the research hotspots in the field of textile industry. The main reasons are the complexity of defect types, the interference of false defects and the random arrangement of gray values. In this paper, a surface defect detection method for glass fiber fabric based on coarse-to-fine strategy is presented. Firstly, the grayscale values of the images to be detected are arranged in the ascending order by row and column respectively, and the corresponding grayscale value curves are drawn. According to the coordinate defect position obtained by the curve, the defect candidate area of glass fiber fabric image is roughly segmented. Then, the convolution kernel is constructed to convolute the coarse segmented image. Finally, the OTSU algorithm is used to locate defects. Qualitative and quantitative experiments show that the detection results of the proposed method are better than Gabor method, FT method and Zhang’s method. The average detection accuracy of the proposed method can reach 98.8%, which can meet the actual needs of enterprises.

Detection of defects on apples using hyperspectral reflection visualization combining both vegetation index analysis and neural network

The article shows the possibility of using vegetation indices to build algorithms for recognizing defects in plant tissue of apples. The authors demonstrate that it is advisable to use halogen lamps as light sources for apples hyperspectral control, but LED backlighting to correct the spectrum at the edges of the 400-1000 nm range is preferable to use. The experimentally obtained spectra of the optical radiation reflected from the apples plant tissue in the mentioned wavelength range are presented and show the possibility of identifying defects using vegetation indices. At the same time, the risk of obtaining false defects for some varieties of apples is revealed. To reduce this risk, it is proposed to use artificial Deep Feedforward neural networks with one hidden layer.

Data-Driven Bias Correction and Defect Diagnosis Model for In-Service Vehicle Acceleration Measurements.

Track quality instruments use low-cost accelerometers placed on or attached to the floors of operating trains, and these instruments collect substantial amounts of data over short inspection periods. The measurements collected by the instruments are the main data source for track irregularity evaluation. However, considerable measurement bias exists in the vertical and lateral vibration data obtained from such instruments. False positive track vibration defects detected by track quality instruments occur frequently. This results in considerable time and effort being expended needlessly because maintenance workers have to visit the railway track sites to check and review the track vibration defects. Therefore, we propose a model for data-driven bias correction and defect diagnosis for in-service vehicle acceleration measurements based on track degradation characteristics. Substantial amounts of historical track measurement data from different inspection methods were mined extensively to eliminate the false positive detection of track vibration defects and diagnose the causes of track vibration defects. Actual measurement data from the Lanxin Railway were used to validate our proposed model. The success rate achieved in identifying false positive track vibration defects was 84.1%, and that in track vibration defect diagnosis was 75.8%. These high success rates suggest that the proposed model can be of practical use in improving railway track maintenance management.

3D Defect detection using weighted principal component thermography

Quantitive infrared thermography like active thermography is a non-destructive testing technique that is used to inspect surfaces of components for defects. A problem with infrared-based defect detection is that misclassifications based on geometrically dependent measurement characteristics can occur. This problem becomes more problematic with the inspection of complex 3D shapes. In this paper, a new infrared quality control procedure using a quality map is proposed that models the geometrically dependent measurement characteristics based on available CAD data and CAD matching techniques. Misclassifications are reduced by using this quality map in combination with a modified version of principal component thermography post-processing. We applied our proposed methodology on a prototype bicycle part and a plaster cast angel figurine. In these experiments, the procedure using quality maps is able to prevent false defect detection.

Graphical User Interface for Ultrasound Guided Wave Imaging

In order to more accurately grasp the exact shape, contour size and distribution of defects in the plate-like structures material and other detailed feature information, this paper designed the graphical user interface GUIGWT0.1, which is specifically used for ultrasonic guided wave imaging. This software integrates filtering back projection algorithm, damage probability reconstruction algorithm and diffraction tomography algorithm, which can image the plate-like structures structure as required, and develop the defect detection results to the visual direction. To investigate the validity of the interface, the projection data of lamb wave in the plate-like with a partial-thickness notch are numerically simulated by three-dimensional elastodynamic finite integration technique (EFIT). The resulting projected data is imported into the graphical user interface for imaging. For each existing reconstruction imaging, the result can be represented in terms of grayscale image of projected data, defect imaging, image color inversion and false color defect recognition. Through simple operation, reconstruction of image with defect of plate-like structure are obtained in the user interface.

Checking the integrity of piles by seismoa cousticdefectos copy

The article presents the results of studies of pile continuity by seismic acoustic flaw detection. The test results revealed the dependence of soil homogeneity on the signal frequency. At the interface of two different density soils, the sound wave is reflected and partially passes on, creating characteristic peaks of false defects. In order to eliminate false defects, geological studies of soils with layer-by-layer occurrence are necessary.