Bearing defect inspection based on machine vision

Composite materials are widely used in aerospace industries due to their light weight, strength, and various other desired properties. However, they are susceptible to various defects occurring during the manufacturing process or in service. Typical defects include porosity, wrinkles, and delamination. Nondestructive means of detection of the defects at any stage are of great importance to ensure quality and safety of composite structures. A nonintrusive removable Lamb wave system and accompanied methodology that is not material dependent are presented in this paper to detect various types of typical defects in laminated composite plates, flat or curved. Through multidimensional data acquisition and processing, abnormality in waves caused by defects is captured and presented in inspection images. The methodologies are demonstrated in two cases: delamination in a curved plate and wrinkles in a flat plate. Overall, the results show that Lamb waves using the piezoelectric transducer and laser vibrometer system can be used for various types of defects inspection in flat or curved composite plates.

Automatic inspection of gas pipeline welding defects using an expert vision system

Solar cells defects inspection plays an important role to ensure the efficiency and lifespan of photovoltaic modules. However, it is still an arduous task because of the diverse attributes of electroluminescence images, such as indiscriminative complex background with extremely unbalanced defects and various types of defects. In order to deal with these problems, this paper proposes a new precise and accurate defect inspection method for photovoltaic electroluminescence (EL) images. The proposed algorithm leverages the advantage of multi attention network to efficiently extract the most important features and neglect the nonessential features during training. Firstly, we designed a channel attention to exploit contextual representations and spatial attention to effectively suppress background noise. Secondly, we incorporate both attention networks into modified U-net architecture and named it multi attention U-net (MAU-net) to extract effective multiscale features for defects inspection. Finally, we propose a hybrid loss which combines focal loss and dice loss aiming to solve two problems: a) overcome the class imbalance problem, and b) allowing the network to train with irregular image labels for some complex defects. The proposed multi attention U-net is evaluated on real photovoltaic EL images datasets using 5-fold cross validation technique. Experimental results demonstrate that the proposed network can segment and detect various complex defects correctly. The proposed method achieved the mean intersection over-union (m-IOU) of 0.699 and F-measure of 0.799 which outperforms the previous methods.

Deep learning breakthrough stimulates the new research trend in infrastructure defects inspection. The lack of quality-controlled, human-annotated, free of charge, and publicly available defect datasets with sufficient amounts of data hinders the progress of deep learning in defects inspection. To boost research in deep learning-based defects inspection, we first summarize 37 publicly available defect datasets in this two-part survey. These defect datasets cover common defects in various types of infrastructure, and the taxonomy of the datasets is based on specific deep learning objectives. Besides, taking crack as the research target, we have combined the existing datasets with self-labeled crack images to establish a benchmark dataset for crack classification and segmentation. Moreover, based on the established crack dataset, we make a comprehensive comparison between state-of-the-art algorithms for classification, segmentation, and detection. In this paper, we concentrate on datasets and algorithms for defect classification. Altogether 11 classification-oriented defect datasets are summarized and demonstrated in detail. Based on the established crack dataset, we comprehensively compare existing state-of-the-art algorithms for object classification, which provides a baseline for future research in defects inspection. The companion paper of this work surveys datasets and algorithms for defect segmentation and detection, where 26 defect datasets are elaborated, and systematic comparison between state-of-the-art algorithms has been conducted. The classification algorithms illustrated in this paper are conducive to the coarse localization of the defects' position in images.

This paper reports a simple and systematic experiment to distinguish various types of defects in bonded wafer pairs with the dynamic blade insertion (DBI) method and real-time observation. In addition, a type of defect was observed in bonded wafer pairs after they were heated to 1273 K and unevenly cooled. It exhibits a fairly large size with coarse Newton-ring contours in IR observation. During DBI examination, these defects exhibit quite a different behavior from other types of defects (void, dust particle, etc.) with the significant intention of resisting the movement of the approaching blade. However, this defect demonstrates a very strong bonding strength in tensile testing.

The perfect Printed Circuit Board (PCB) plays a very important role in every electronic device as well as in automation systems. So, it is very important to find defects in the PCB before installing it to any system or any device. However, PCB Manufacturers use various inspection systems in the process of manufacturing PCBs for detecting various types of defects in the PCB. In this article, we present the Automated assembled PCB Inspection System. This system finds defects such as missing components and improper position of its components by using the Pattern matching Technique where a good known score of template image is matched with the score of the test image. This system gives results at each inspection within 10 Seconds and the result given by this system are passed or fail in the form of an array sheet. This automated inspection system is created by using NI Vision Builder AI and NI LabVIEW technology. Ni Vision Builder AI has been used to create the algorithm. And NI LabVIEW has been used to create the application.

The article analyzes the fields of application of machine vision. Special attention is focused on the application of Machine Vision in intelligent technological systems for product quality control. An important aspect is a quick and effective analysis of product quality directly at the stage of the technological process with high accuracy in determining product defects. The appropriateness and perspective of using the mathematical apparatus of artificial neural networks for the development of an intelligent technological system for monitoring the geometric state of products have been demonstrated. The purpose of this study is focused on the identification and classification of reed tuber quality parameters. For this purpose, new methods of identification and classification of quality control of various types of defects using computer vision and machine learning algorithms were proposed.

Visualization of X-ray Fourier images of deformation fields for various types of defects in monocrystals

This work assessed individual widegap semiconductor nanoparticles using ultraviolet photoluminescence (PL) microscopy. ZnO nanoparticles having a bandgap energy of 3.4 eV were first synthesized with a variety of reaction times and then evaluated, employing a unique spectroscopic system. The resulting spectra were processed to show components related to an exciton and three types of defects. The nanoparticles formed in the early stage of the reaction process all exhibited the smallest exciton and largest defect PL intensities, whereas the fully grown nanoparticles provided an intense exciton peak but the lowest defect-related emissions. The particles sampled in the central reaction stage showed good exciton intensity but the greatest variation of the component peaks. The exciton PL peaks obtained from individual nanoparticles exhibited a narrow distribution of energy levels around the average peak position, in agreement with ensemble data acquired by using ultraviolet–visible spectroscopy. PL characteristics related to defects in the specimens at each reaction stage were explained based on various types of defects and the defect repair process. Individual nanoparticle assessments suggested variations in the local environment during the formation of these materials. Detailed examinations in this manner are expected to assist in the fabrication of identical nanomaterials for future semiconductor technologies.

Theoretical studies on the mechanical and electronic properties of 2D and 3D structures of Beryllium-Oxide graphene and graphene nanobud

An accurate fuzzy measure-based detection method for various types of defects on strip steel surfaces

Although natural diamonds are found on the surface of the earth, they did not originate there. They grew in the depths of the earth under conditions of high temperature and high pressure and were subsequently extruded to the surface. This complicated temperature and pressure history results in particular types of defect being present in natural diamonds. This article describes the various types of defect as well as a number of physical properties of diamond. The synthesis of diamond is also described. It is hoped that information about natural diamonds will give some insight into their conditions of growth and their subsequent extrusion to the surface.

The inspection of steam-generator (SG) tubes in a nuclear power plant (NPP) is a time-consuming, laborious, and hazardous task because of several hard constraints such as a highly radiated working environment, a tight task schedule, and the need for many experienced human inspectors. This paper presents a new distributed intelligent system architecture for automating traditional inspection methods. The proposed architecture adopts three basic technical strategies in order to reduce the complexity of system implementation. The first is the distributed task allocation into four stages: inspection planning (IF), signal acquisition (SA), signal evaluation (SE), and inspection data management (IDM). Consequently, dedicated subsystems for automation of each stage can be designed and implemented separately. The second strategy is the inclusion of several useful artificial intelligence techniques for implementing the subsystems of each stage, such as an expert system for IP and SE and machine vision and remote robot control techniques for SA. The third strategy is the integration of the subsystems using client/server-based distributed computing architecture and a centralized database management concept. Through the use of the proposed architecture, human errors, which can occur during inspection, can be minimized because the element of human intervention has been almost eliminated; however, the productivity of the human inspector can be increased equally. A prototype of the proposed system has been developed and successfully tested over the last six years in domestic NPP's.

Abstract— In industrial manufacturing processes, ensuring the quality of products is crucial. This paper proposes a system for detecting defects in products using image template matching techniques implemented in MATLAB. The system's primary function is to compare captured images of products with predefined templates to identify potential defects accurately. The method employed in this system is template matching, a well-established approach in image analysis that allows for efficient defect detection. MATLAB, a widely used software tool for image processing, provides the necessary functionality and a robust set of algorithms to implement the proposed system. The experimental results demonstrate the effectiveness of the approach in detecting various types of defects, such as scratches, cracks, and misalignments. This defect detection system offers a reliable and automated solution for improving the efficiency and productivity of manufacturing industries. By enabling early detection and intervention, it contributes to enhancing product quality control and minimizing defective outputs, ultimately leading to cost savings and customer satisfaction. Keywords— Defect Detection; MATLAB; Template Matching; Image Processing

As the service life of sewer pipelines increases, various types of defects inevitably occur, posing significant risks to urban infrastructure and public safety. Timely detection and assessment of these defects are crucial to ensuring the structural integrity and functionality of sewer systems. Traditional manual inspections are time-consuming, laborintensive, and prone to human errors. Object detection technology based on deep learning provides an efficient and automated alternative by training models to accurately identify the type and location of pipeline defects. However, the complex internal environment of pipelines, including low-light conditions, noise interference, and varying defect appearances, presents significant challenges for detection accuracy. To address these issues, we conducted an in-depth study on common sewer pipeline defects and applied image preprocessing techniques such as grayscale conversion and denoising to enhance dataset quality. Furthermore, we improved the YOLOv8 model by integrating the CPA Enhancer module into its Backbone structure, optimizing feature extraction and defect recognition. Based on this enhancement, we developed a deep active learning framework, YOLOv8-CPA, which leverages a chain-thinking prompt mechanism to refine detection performance iteratively. Experimental results demonstrate that the YOLOv8-CPA pipeline inspection system achieves high accuracy in detecting and classifying pipeline defects. By improving detection efficiency, ensuring consistency, and accelerating validation processes, our system significantly enhances the sewer inspection workflow. The proposed method contributes to effective defect management, aiding in the timely implementation of appropriate maintenance and rehabilitation strategies for sewer infrastructure.