Computer-aided Inspection Research Articles

Enhancing Photovoltaic Module Fault Diagnosis with Unmanned Aerial Vehicles and Deep Learning-Based Image Analysis

Artificial intelligence (AI) has evolved into a powerful tool that has wide-spread application in computer vision such as computer-aided inspection, industrial control systems, and navigation of robots. Monitoring the condition of machineries and mechanical components for the presence of faults with the aid of image-based automated analysis is one major application of computer vision. Diagnosing machinery faults from images can be made feasible with the adoption of deep learning and machine learning techniques. The primary objective of this study is to detect malfunctions in photovoltaic (PV) modules by utilizing a combination of deep learning and machine learning methodologies, with the assistance of RGB images captured via unmanned aerial vehicles. Six test conditions of PV modules such as good panel, snail trail, delamination, glass breakage, discoloration, and burn marks were considered in the study. The overall experimentation was carried out in two phases: (i) deep learning phase and (ii) machine learning phase. In the initial deep learning phase, the final fully connected layer of six pretrained networks, namely, DenseNet-201, VGG19, ResNet-50, GoogLeNet, VGG16, and AlexNet, was utilized to extract PVM image features. During the machine learning phase, feature selection from the extracted features was carried out using the J48 decision tree algorithm. Post selection of features, three families of classifiers such as tree, Bayes, and lazy were applied to determine the best feature extractor-classifier pair. The combination of DenseNet-201 features with k-nearest neighbour (IBK) classifier produced the overall classification accuracy of 100.00% among all other pretrained network features and classifiers considered.

“Republicanism”: a grounding concept for the American Revolution?

ABSTRACT This essay revisits the long-standing debate concerning the sources for the underlying political beliefs and commitments held by the “founding generation,” those colonists who came to the conclusion that separation from the mother country was necessary and inevitable. It uses a mixed mode of enquiry and analysis, blending standard close reading of texts with computer-aided inspection of the archive at scale. It seeks to clarify the extent to which a set of political assumptions and theories widely assumed to be gathered under the rubric of “republicanism” could have shaped the thinking of this founding generation. Given that the term “republicanism” was very infrequent until the last decade or so of the eighteenth century, the essay proposes ways of interrogating the historical formation of concepts for which no label can be easily discerned. Following the consideration of some candidate terms or phrases that might, at the time of the American Revolution, have effectively been hosts for the political idea of “republicanism,” the essay arrives at the conclusion that the founding generation reimagined the role that “republicanism” had played in their thinking at the time.

A Survey of Process Monitoring Using Computer-Aided Inspection in Laser-Welded Blanks of Light Metals Based on the Digital Twins Concept

The benefits of laser welding include higher production values, deeper penetration, higher welding speeds, adaptability, and higher power density. These characteristics make laser welding a superior process. Many industries are aware of the benefits of switching to lasers. For example, metal-joining is migrating to modern industrial laser technology due to improved yields, design flexibility, and energy efficiency. However, for an industrial process to be optimized for intelligent manufacturing in the era of Industry 4.0, it must be captured online using high-quality data. Laser welding of aluminum alloys presents a daunting challenge, mainly because aluminum is a less reliable material for welding than other commercial metals such as steel, primarily because of its physical properties: high thermal conductivity, high reflectivity, and low viscosity. The welding plates were fixed by a special welding fixture, to validate alignments and improve measurement accuracy, and a Computer-Aided Inspection (CAI) using 3D scanning was adopted. Certain literature has suggested real-time monitoring of intelligent techniques as a solution to the critical problems associated with aluminum laser welding. Real-time monitoring technologies are essential to improving welding efficiency and guaranteeing product quality. This paper critically reviews the research findings and advances for real-time monitoring of laser welding during the last 10 years. In the present work, a specific methodology originating from process monitoring using Computer-Aided Inspection in laser-welded blanks is reviewed as a candidate technology for a digital twin. Moreover, a novel digital model based on CAI and cloud manufacturing is proposed.

Using 3D scanning in assessing the dimensional accuracy of mechanically machined parts

In recent years, productivity and competitiveness have played a major role in manufacturing. To improve these aspects, many companies use computer-aided inspection (CAI) technologies. Inspection of parts, using 3D scanning technologies, is one the most modern and competitive ways to ensure the quality control of products. This paper presents a study on the assessment of the machined parts’ dimensional accuracy. An Atos Core device for 3D scanning of the part was used. The Atos device allows a non-contact inspection of the part’s entire surface, generating a very dense point cloud, which stores, with great accuracy, all the details of the inspected object. The machined part is compared with its numerical model, obtained by computer-aided design (CAD). The accuracy of the machining was proven to be very good, and the inspection’s result proves that the dimensions of the piece fall within the tolerance limits. A complete inspection of the piece, using 3D scanning, guarantees a fast and accurate quality control, thus reducing the production times.

A Computer-Aided Inspection System to Predict Quality Characteristics in Food Technology

Physicochemical and sensory analyses are commonly used to determine the quality characteristics of food samples in Food Industries. These methods are tedious, laborious, produce chemical residues, and involve the destruction of the samples. For the meat industries, this work proposes a non-invasive and non-destructive computer-aided inspection system, based on computer vision and ensemble machine learning techniques. The paper presents all the possibilities for the development of the system, making an exhaustive comparison of different algorithms used to extract features from the images of the samples, and various machine learning approaches, studying up to 6160 different models, and selecting the top 110 for the ensemble proposal. The system determines all the physicochemical, textural, and sensory quality characteristics of pork and beef loins in four meat states (fresh, thawed, cooked, and cured) with good precision, being a real alternative to the usual methods for the Food Industry.

Discipline-specific cognitive factors that influence grade 9 students’ performance in chemistry

Students’ academic performance in chemistry can be the result of a number of cognitive and affective factors. This study explored the influence of the discipline-specific cognitive factors of knowledge structure, cognitive perspectives, and cognitive patterns on grade 9 students' chemistry performance. One instrument measured chemistry academic performance based on concept knowing, application and problem solving. Six tasks with marking keys measured the discipline-specific cognitive factors of knowledge structure, cognitive perspectives, and cognitive patterns. Different groups of grade 9 students participated in pilot tests and the field tests. The quality of the chemistry academic performance instrument and the six tasks was inspected by both expert assessment with six raters and computer-aided inspection including Rasch analysis and Kendall rater-consistency reliability tests. Correlation analysis and multiple regression analysis explored the relationship among academic performance and knowledge structure, cognitive perspectives, and cognitive patterns. According to the results of this research, knowledge structure, cognitive perspective and cognitive pattern all influenced grade 9 students’ chemistry performances; cognitive perspective was the most important factor. Based on these findings, we discuss individual student performance relative to their discipline-specific cognitive factors. We recommend that instruction of discipline-specific learning in chemistry pay attention to each of the three discipline-specific cognitive factors and that tasks be designed to promote the progress of each of these three discipline-specific cognitive factors, especially cognitive perspectives.

Interlaboratory Empirical Reproducibility Study Based on a GD&T Benchmark

The ASME Y14.5 geometric dimensioning and tolerancing (GD&T) and ISO-GPS (geometrical product specifications) standards define tolerances that can be added to components to achieve the necessary functionality and performance. The zone that each feature must lie within is defined in each tolerance. Measurement processes, including planning, programming, data collection (with contact or without contact), and data processing, check the compliance of the part with these specifications (tolerances). Over the last two decades, many works have been realized by the metrology community to investigate the accuracy, the measuring methods, and, specifically, the measurement errors of fixed and portable coordinate measuring machines (CMMs). A review of the literature showed the progression of CMMs in terms of accuracy and repeatability. However, discrepancies were observed between measurements using different CMMs or operators. This paper proposed a GD&T-based benchmark for the evaluation of the performance of different CMM operators in computer-aided inspection (CAI), considering different criteria related to the dimensional and geometrical features. An artifact was designed using basic geometries (cylinder and plane) and free-form surfaces. The results obtained from the interlaboratory comparison study showed significant performance variability for complex GD&T, such as in the composite profile and localization. This, in turn, emphasized the importance of GD&T training and certification in order to ensure a uniform understanding among different operators, combined with a fully automated inspection code generator for GD&T purposes.

Systematic review of STEP-NC-based inspection

Closed-loop inspection manufacturing (CLIM) and computer-aided inspection planning (CAIP) have increasingly been evolved into a size method that can be used to look at fashions in the industrial area with a number of methods. CAIP is hastily carried out in the gadget and will become an indispensable phase of the production system. Numerous scientific findings that tackle the theme from exclusive perspectives are accompanied by CLIM and CAIP in the measuring process. A systematic overview of these heterogeneous methods can be considered vital for CLIM to become aware of the ways in which CAIP can be used in your personal context. Thus, this systematic review of literature aims to analyze current inspection article and use of CLIM in CNC machines, assess heritage research to create a steady taxonomy, and provide a framework for forthcoming CLIM lookup in accordance with CAIP in the desktop gadget to enhance product quality. The consequences of this study provide a comprehensive taxonomy and a define in terms of the advanced features of this emerging technology. Articles on CAIP and CLIM, system-based technology, implementation of STEP/STEP-NC, and feature-based technology were discussed. As a result, various CLIM lookup findings regarding the machining machine tended to the center of attention rather on the VISION/aspects of the ISO14649-based computing device and further on inspection process planning perspectives, as well as factors indicating CLIM in technologies. This review paper will strengthen this study by supplying a deep analysis of the present facts to shut down the hole and lengthen the current literature.

Improved Bi-Criterion flexible registration for fixtureless inspection of compliant parts

Because of the effects of gravity and/or residual stress, some manufactured mechanical parts, such as sheet metals and skins, often have a significantly different shape in a free-state position as compared to their state-of-use position. These parts are described as compliant, flexible or nonrigid. Expensive specialized fixtures are currently used prior to performing geometrical inspection operations in order to maintain compliant parts in the state-of-use position. This paper introduces an automatic bi-criterion flexible registration method for the dimensional and geometric inspection of such parts. The proposed method deforms the data acquired via a non-contact scanner of a compliant part in a free-state position until it reaches the nominal CAD shape for inspection with conventional Computer-Aided Inspection (CAI) tools. In other words, the method neutralizes the deviations induced in a compliant part by the effects of gravity and residual stress, allowing the acquired data to be treated as if it were obtained from a rigid part, using already available conventional (rigid) CAI tools. A proposed algorithm based on the BOFR-2 (the 2nd version of a Bi-Objective Flexible Registration algorithm) method is validated against both virtual simulated and experimental real industrial case studies from the aerospace sector. The resulting cost reduction and agility increasing make this fixtureless method well adapted to the requirements of unit-batch production in the context of Industry 4.0.

Investigation of radiation level and assessment of dimensional accuracy of acquired CBCT images

The aim of this study is to investigate whether the radiation dose level, during 3D cone beam computed tomography scanning of the patient, influences the dimensional accuracy of the reconstructed surface 3D models, and if it does, to analyse the accuracy improvement in relations to radiation dose levels. In this study is proposed a workflow where two datasets of images acquired with different radiation dose levels of the lower jaw were used for analysis, with the focus on region where the tooth was extracted on the patient’s mandible. Statistical evaluators used for analysis are Jaccard index, Dice coefficient and Tanimotto coefficient, while accuracy and precision were also used as a statistical measure. The results of evaluation of six sample images for two datasets by Jaccard index are ranging from 0.684 to 0.873 and 0.579 to 0.849, respectively, by Dice are 0.812 to 0.932, and 0.733 to 0.918, while by Tanimoto are ranging from 0.763 to 0.895, and 0.626 to 0.867, respectively. Regarding the accuracy and precision, the average difference for those six images is a little over 1% and 2%, respectively. Reconstructed tooth cavities were also obtained where volume analysis showed insignificant difference between two reconstructed surface 3D models of cavities. This was also confirmed with performed dimensional measurements obtained for the two tooth cavities, as well as bone density evaluation inside those cavities. Computer-aided inspection was also performed which shows similar results of two datasets. It can be concluded, on bases of the obtained results, that the reduced level of radiation dose during image acquisition doesn’t have significant impact on dimensional accuracy compared to the higher dosage.

Integrated system for automation of process, fixture and inspection planning

A novel planning system which can automatically accomplish machining plans with the appropriate fixture setup and inspection strategies have been presented. In this work, the primary steps that are required to achieve an automated process scheme, modular fixture design, and inspection planning have been described. The object-oriented approach has been employed to attain automation and integration of the design process, including fixtures and inspection modules. This kind of system is unique as it provides an integrated platform for three important manufacturing activities, namely process planning, fixture design, and inspection. The STEP has been employed as an input to the proposed framework. The computer-aided process planning module consists of geometric and dimensional data, machining feature recognition information and process planning files. The modular fixture layout design has been developed using the set of defined rules, search strategies and the graphical database of various fixturing components. The computer-aided inspection planning module automatically generates an inspection plan and Dimensional Measuring Interface Standard code for coordinate measuring machines. It is noticeable from the reported case study and its outcomes that the developed integrated environment can effectively be employed in different manufacturing industries for automated process planning, fixture design, and inspection.

Reproducibility Experimentation among Computer-Aided Inspection Software from a Single Point Cloud

The ISO GPS and ASME Y14.5 standards have defined dimensional and geometrical tolerance as a way to express the limits of surface part variations with respect to nominal model surfaces. A quality-control process using a measuring device verifies the conformity of the parts to these tolerances. To convert the control measurement points as captured by a device such as a coordinate measurement machine (CMM) or noncontact scan, it is necessary to select the appropriate algorithm (e.g., least square size and maximum inscribed size) and to include the working hypotheses (e.g., treatment of outliers, noise filtering, and missing data). This means that the operator conducting the analysis must decide on which algorithm to use. Through a literature review of current software programs and algorithms, many inaccuracies were found. A benchmark was therefore developed to compare the algorithm performance of three computer-aided inspection (CAI) software programs. From the same point cloud and on the same specifications (requirements and tolerances), three CAI options have been tested with several dimensional and geometrical features.

Evaluation of synthetically generated patterns for image-based 3D reconstruction of texture-less objects

Every material and every surface has its own visual texture. Surfaces with monotone, repetitive or uniform visual texture represent a challenge for the image-based Structure From Motion Multi-View Stereo three-dimensional (3D) reconstruction method. It is possible to overcome the lack of visual texture on the surfaces by projecting synthetically generated images (patterns) using a video projector. This research proposes the generation of the synthetic images that are based on digits of irrational numbers pi, phi, e, sqrt2, sqrt3 and digits produced by the random number generator. Images were divided into three classes based on the number of hues. The aim is to evaluate synthetically generated images and determine the characteristics of the most suitable synthetic image(s) that increase the accuracy of the final reconstructed polygonal 3D model. The synthetically generated images were evaluated using results of the multi-criteria analysis as well as real and virtual planar surface 3D digitization methods, where images with uniform distribution histogram have the most suitable characteristics. To verify evaluation results, 3D reconstruction of aluminium test model was carried out. Four polygonal 3D models of the aluminium test model were reconstructed. Three polygonal 3D models were reconstructed using projected patterns of each class, and one was reconstructed in daylight condition. The resulting accuracies of reconstructed polygonal 3D models were evaluated using the Computer-Aided Inspection. The polygonal 3D model obtained by projecting class I pi I pattern with the strong random stochastic visual texture achieved std. distance of 0.173 mm and mean distance of 0.016 mm compared to the polygonal 3D model obtained in daylight condition with std. and mean distance of 1.188 mm and −0.139 mm respectively.

Computer-Aided Inspection Planning: A Multisensor High-Level Inspection Planning Strategy

Computer-aided inspection planning (CAIP) has gained significant research attention in the last years. So far, most CAIP systems have focused on the use of a touch probe mounted on a coordinate measuring machine (CMM). This article investigates multisensor measurement aiming to perform automatic and efficient inspection plans. High-level inspection planning, which deals with sequencing of measuring operations, is the main concern of inspection planning. This paper presents an automatic approach to generate inspection sequences by combining laser sensor and touch probe, and by giving preference to the measurement using the laser sensor if quality requirements are satisfied. The proposed approach consists of three steps. In the first step, recognition of inspection data from the computer-aided design (CAD) part model is carried out based on the concept of inspection feature (IF), and the extracted information is stored in a database. In the second step, a list of privileged scanner orientations is proposed by analyzing the accessibility of both sensors. In the third step, a sequence of operations is generated iteratively. For a given scanner orientation, the ability of the laser sensor is assessed according to an original process based on fuzzy logic model. If the laser sensor does not meet the ability requirements, touch probe ability is assessed. The proposed approach is implemented and tested on a part defined by its CAD model and specifications.

“What-if” scenarios towards virtual assembly-state mounting for non-rigid parts inspection using permissible loads

Recent developments in the fixtureless inspection of non-rigid parts based on computer-aided inspection (CAI) methods significantly contribute to diminishing the time and cost of geometrical dimensioning and inspection. Generally, CAI methods aim to compare scan meshes, which are acquired using scanners as point clouds from non-rigid manufactured parts in a free-state, with associated nominal computer-aided design (CAD) models. However, non-rigid parts are deformed in a free-state due to their compliance behavior. Industrial inspection approaches apply costly and complex physical inspection fixtures to retrieve the functional shape of non-rigid parts in assembly-state. Therefore, fixtureless inspection methods are developed to eliminate the need for these complex fixtures and to replace them with simple inspection supports. Fixtureless inspection methods intend to virtually (numerically) compensate for flexible deformation of non-rigid parts in a free-state. Inspired by industrial inspection techniques wherein weights (e.g., sandbags) are applied as restraining loads on non-rigid parts, we present a new fixtureless inspection method in this article. Our proposed virtual mounting assembly-state inspection (VMASI) method aims at predicting the functional shape (in assembly-state) of a deviated non-rigid part (including defects such as plastic deformation). This method is capable of virtually mounting the scan mesh of a deviated non-rigid part (acquired in a free-state) into the designed assembly-state. This is fulfilled by applying permissible restraining forces (loads) that are introduced as pressures on surfaces of a deviated part. The functional shape is then predicted via a linear FE-based transformation where the value and position of required restraining pressures are assessed by our developed restraining pressures optimization (RPO) approach. In fact, RPO minimizes the orientation difference and distance between assembly mounting holes on the predicted shape of a non-rigid part with respect to nominal ones on the CAD model. Eventually, the inspection is accomplished by examining the mounting holes offset on the predicted shape of the scan model concerning the nominal CAD model. This ensures that the mounting holes on the predicted shape of a scan model in assembly-state remain in the dedicated tolerance range. This method is evaluated on two non-rigid parts to predict the required restraining pressures limited to the permissible forces during the inspection process and to predict the eventual functional shape of the scan model. We applied numerical validations for each part, for which different types of synthetic (numerically simulated) defects are included into scan meshes, to determine whether the functional shape of a geometrically deviated part can be virtually retrieved under assembly constrains.

Towards a standard protocol for antimony intralesional infiltration technique for cutaneous leishmaniasis treatment.

BACKGROUNDDespite its recognised toxicity, antimonial therapy continues to be the first-line drug for cutaneous leishmaniasis (CL) treatment. Intralesional administration of meglumine antimoniate (MA) represents an alternative that could reduce the systemic absorption of the drug and its side effects.OBJECTIVESThis study aims to validate the standard operational procedure (SOP) for the intralesional infiltration of MA for CL therapy as the first step before the assessment of efficacy and safety related to the procedure.METHODSThe SOP was created based on 21 trials retrieved from the literature, direct monitoring of the procedure and consultation with experts. This script was submitted to a formal computer-aided inspection to identify readability, clarity, omission, redundancy and unnecessary information (content validation). For criterion and construct validations, the influence of critical condition changes (compliance with the instructions and professional experience) on outcome conformity (saturation status achievement), tolerability (pain referred) and safety (bleeding) were assessed.FINDINGSThe median procedure length was 12 minutes and in 72% of them, patients classified the pain as mild. The bleeding was also classified as mild in 96.6% of the procedures. Full compliance with the SOP was observed in 66% of infiltrations. Despite this, in 100% of the inspected procedures, lesion saturation was observed at the end of infiltration, which means that it tolerates some degree of modification in its execution (robustness) without prejudice to the result.CONCLUSIONSThe procedure is reproducible and can be used by professionals without previous training with high success and safety rates.

Assessment of the Robustness of a Fixtureless Inspection Method for Nonrigid Parts Based on a Verification and Validation Approach

The increasing practical use of computer-aided inspection (CAI) methods requires assessment of their robustness in different contexts. This can be done by quantitatively comparing estimated CAI results with actual measurements. The objective is comparing the magnitude and dimensions of defects as estimated by CAI with those of the nominal defects. This assessment is referred to as setting up a validation metric. In this work, a new validation metric is proposed in the case of a fixtureless inspection method for nonrigid parts. It is based on using a nonparametric statistical hypothesis test, namely the Kolmogorov–Smirnov (K–S) test. This metric is applied to an automatic fixtureless CAI method for nonrigid parts developed by our team. This fixtureless CAI method is based on calculating and filtering sample points that are used in a finite element nonrigid registration (FENR). Robustness of our CAI method is validated for the assessment of maximum amplitude, area, and distance distribution of defects. Typical parts from the aerospace industry are used for this validation and various levels of synthetic measurement noise are added to the scanned point cloud of these parts to assess the effect of noise on inspection results.

STEP File Analyzer Software.

The STEP File Analyzer is a software tool that generates a spreadsheet or a set of CSV (comma-separated value) files from a STEP (ISO 10303 –STandard for Exchange of Product model data) Part 21 file. STEP files are used to represent product and manufacturing information (PMI) and for data exchange and interoperability between Computer-Aided Design (CAD), Manufacturing (CAM), Analysis (CAE), and Inspection (CMM) software related to the smart manufacturing digital thread. STEP is also used for the long-term archiving and retrieval of product data. A spreadsheet simplifies inspecting information from the STEP file at an entity and attribute level. Typical STEP file viewers show a 3D visualization of the part or model represented by the STEP file. The viewers usually have a high-level hierarchical display of the information in the STEP file where the user can drill down to individual attributes of parts. However, there is no way to view all of the actual STEP entities and their attributes at once. The STEP File Analyzer provides this capability by creating a spreadsheet from the STEP file. The STEP File Analyzer also generates reports for PMI Representation, PMI Presentation, and Validation Properties based on Recommended Practices defined by the CAx Implementor Forum (CAx-IF) [5]. The objective of the CAx-IF is to advance CAx (mainly Computer-Aided Design and Engineering) software system STEP translator development and to ensure that user requirements for interoperability are satisfied.

Imperfection sensitivity of externally-pressurized, thin-walled, torispherical-head buckling

The bottom head in the residential electric water heater tank is an externally-pressurized, thin-walled, torispherical structure. In this paper, the buckling behavior of the bottom head was studied by experimental and finite element analysis (FEA) methods. The shapes of four randomly selected water heater tanks were measured by Computer Aided Inspection (CAI), and hydrostatic tests and strain measurements were performed. The buckling simulations of the tested tanks were carried out by finite element method, and the obtained results were in good agreement with the experimental data. Two parameters, contact imperfection ratio, Rimc, and geometry imperfection ratio, Rimg, were used to quantitatively characterize the magnitudes of the main imperfections in the water heater tank manufacturing process. FEA models, including both contact and geometry imperfections, were used to investigate the buckling pressure of the bottom head. The results show that the geometry imperfection, within the range of ASME Code Ⅷ-2 requirements, has a greater effect on the buckling pressure than the contact imperfection. The obtained curves of buckling pressure versus imperfections can provide guidance for water heater tank design and manufacture.

A touch-probe path generation method through similarity analysis between the feature vectors in new and old models

The On-machine measurement (OMM), which measures a workpiece during or after the machining process in the machining center, has the advantage of measuring the workpiece directly within the work space without moving it. However, the path generation procedure used to determine the measuring sequence and variables for the complex features of a target workpiece has the limitation of requiring time-consuming tasks to generate the measuring points and mostly relies on the proficiency of the on-site engineer. In this study, we propose a touch-probe path generation method using similarity analysis between the feature vectors of three-dimensional (3-D) shapes for the OMM. For the similarity analysis between a new 3-D model and existing 3-D models, we extracted the feature vectors from models that can describe the characteristics of a geometric shape model; then, we applied those feature vectors to a geometric histogram that displays a probability distribution obtained by the similarity analysis algorithm. In addition, we developed a computer-aided inspection planning system that corrects non-applied measuring points that are caused by minute geometry differences between the two models and generates the final touch-probe path.