Manual Visual Inspection Research Articles

A migration strategy for automating a vision-based concentricity inspection station

This paper presents the development of an automation migration strategy in transforming a manual visual inspection work cell into a semi-automated one for a medical device manufacturer in China. The object under study is a washer/magnet subassembly used in an air release valve. These two circular components must be bonded concentrically and then inspected with bare eyes by a human inspector. Such inspection process was prone to error as the inspector struggled to keep up with the production cycle time. The methodology employed in this research consists of four steps. First, we examined the cost of rework through the Pareto analysis . The results indicated that the washer/magnet misalignment accounted for more than 40% of valve defects and thus deserved immediate attention. Next, we conducted two Kappa analyses to evaluate repeatability and reproducibility of the human inspectors assigned to perform the inspection tasks. The results showed that the human inspectors failed to pass these tests and a suitable automation solution must be sought. Afterwards, efforts were made to develop a vision based semi-automated concentricity inspection station to eliminate human inspection errors. Hardware setup, software algorithms, lighting and other supporting devices are presented in this paper as well as potential savings for such an installment. Finally, we conducted an economical analysis to compare the semi-automated solution with a fully automated one to identify the best automation migration strategy. The analysis results showed that the semi-automated solution was a favorable choice due to a shorter payback period and its ease of reinstallation if the factory is to be relocated. ► We develop an automated inspection migration strategy for a medical device manufacturer in China. ► The gage R&R analysis shows human performance is unacceptable for inspection of small parts. ► By applying machine vision, the parts misalignment problem can virtually be eliminated. ► We show that semi-automation is economically justified over full-automation.

High End-Expiratory Airway Pressures Caused by Internal Obstruction of the Draeger Apollo® Scavenger System That Is Not Detected by the Workstation Self-test and Visual Inspection

High End-Expiratory Airway Pressures Caused by Internal Obstruction of the Draeger Apollo® Scavenger System That Is Not Detected by the Workstation Self-test and Visual Inspection

Combination of Anomaly Algorithms and Image Features for Explosive Hazard Detection in Forward Looking Infrared Imagery

A novel approach is proposed for combining multiple anomaly algorithm decisions with image space cell-structured features in a long wave infrared (LWIR) system in the context of forward looking (FL) buried explosive hazard detection along a road. A pre-screener is applied first, which is an ensemble of trainable size-contrast filters with mean shift clustering in Universal Transverse Mercator (UTM) space. Next, features from image chips representing anomaly decisions from different algorithms are extracted from UTM confidence maps based on maximally stable extremal regions (MSERs) and Gaussian mixture models (GMMs). Pre-screener hits in UTM space are back-projected into the video at multiple standoff distances and cell-structured local binary patterns (LBPs), histogram of gradients (HOGs) and mean-variance descriptors are extracted. Experiments are conducted using buried materials with varying metal contents and depths at a U.S. Army test site. Results are extremely encouraging for FL imaging and show a significant decrease in the number of false alarms (FAs). Targets not currently detected by our system are also not detected by a human under manual visual inspection.

Automated Visual Inspection System for Mass Production of Hard Disk Drive Media

Manual visual inspection is currently used in the modern Hard Disk assembly process. This research shows a feasible design to automate the visual inspection process based on wavelength dependant detection. Two defect detection algorithms, one - computationally simple and another - complex, are explored in this paper. The developed system meets the performance of the current manual inspection method while providing high accuracy in targeted defects.

Monitoring of HTS Compound Library Quality via a High-Resolution Image Acquisition and Processing Instrument

This report presents the high-resolution image acquisition and processing instrument for compound management applications (HIAPI-CM). The HIAPI-CM combines imaging spectroscopy and machine-vision analysis to perform rapid assessment of high-throughput screening (HTS) compound library quality. It has been customized to detect and classify typical artifacts found in HTS compound library microtiter plates (MTPs). These artifacts include (1) insufficient volume of liquid compound sample, (2) compound precipitation, and (3) colored compounds that interfere with HTS assay detection format readout. The HIAPI-CM is also configured to automatically query and compare its analysis results to data stored in a LIMS or corporate database, aiding in the detection of compound registration errors. To demonstrate its capabilities, several compound plates (n=5760 wells total) containing different artifacts were measured via automated HIAPI-CM analysis, and results compared with those obtained by manual (visual) inspection. In all cases, the instrument demonstrated high fidelity (99.8% empty wells; 100.1% filled wells; 94.4% for partially filled wells; 94.0% for wells containing colored compounds), and in the case of precipitate detection, the HIAPI-CM results significantly exceeded the fidelity of visual observations (220.0%). As described, the HIAPI-CM allows for noninvasive, nondestructive MTP assessment with a diagnostic throughput of about 1min per plate, reducing analytical expenses and improving the quality and stewardship of HTS compound libraries.

Artificial neural network-based approach for detection and classification of defects in polymeric composites using machine vision in SEM study

Miscibility and adhesion of polymeric composites are adversely affected by some microstructural defects commonly found in polymeric composites. Conventionally, the above-mentioned defects are detected and classified through manual visual analysis of the scanning electron micrographs obtained via scanning electron microscopy (SEM). However, in manual visual inspection of the micrographs, inspection personnel have their own standard of detecting and classifying defects. So, it is quite possible that two different inspection personnel may classify the same defect into two different predefined classes of defects. In order to overcome this problem of misclassification, a machine vision-based approach, which utilises an artificial neural network (ANN) model, is proposed in this study for automated detection and classification of the defects commonly found in polymeric composites. Results obtained reveal that quite a reasonable level of accuracy in detection and classification of microstructural defects of polymeric composites is achieved with the proposed approach.

A Semiautomated Approach for Quality Controlling Large Historical Ocean Temperature Archives

Abstract This paper describes a method consisting of both automated statistical screening and manual quality control through expert visual inspection, which produces a historical ocean temperature archive of high quality—that is, nearly all profiles are unique (duplicate elimination) and 95% of bad data is eliminated. The complete process involves comprehensive duplicate elimination, an unreasonable gradient check, and statistical screening to distill out suspect profiles, which are then only eliminated (or partially so) during an expert manual visual inspection step. Statistical screening was optimized using an archive of known quality. Two iterations of statistical screening were required to identify the bulk of the bad data. Of an archive of about 121 000 profiles, the authors found they had to manually inspect 35% of profiles to remove 95% of the bad data. While costly, they argue such an effort is worthwhile so that the historical ocean temperature archives, which have cost the global community millions of dollars to obtain, are made more immediately useful for climate and ocean sciences. An archive of upper ocean temperature profiles from the Indian Ocean is near completion and extensions into the Pacific Ocean have begun.

Portable Weld Inspection Management System

Manual visual inspection is by far the most widely used weld inspection method. A given weld may be visually examined multiple times as parts are joined and made into assemblies. Because visual inspection is somewhat subjective, and prone to error (a typical inspector only identifies about 85% of the visible defects); welds that pass one inspection may fail subsequent inspections, resulting in multiple inspection and repair cycles. This paper discusses a project that tested the benefits and limitations of a semiautomated weld inspection system. The goal of this semiautomated inspection approach is to provide quantitative, nonsubjective quality measurements of welded structures in order to: eliminate redundant inspections, reduce unnecessary multiple repair cycles, avoid repair of welds that meet minimum size requirements, and enable recording of weld size so that overwelding can be identified and reduced. The technology would also allow tracking of weld quality and statistical analysis of welding process capability to support lean/six-sigma continuous improvement initiatives. A prototype system was assembled and field tested by inspecting actual ship structures. The equipment evaluated has potential, but needs both hardware and software modifications before it can be used as a tool on a regular basis in a shipbuilding environment. As is, it will be useful as an audit tool to gauge the health of the visual inspection process and to further document the inherent variability of the visual inspection process. Recommendations were made for improvements to refine the prototype tool prior to broader deployment.

A Quantitative Zebrafish Phenotyping Tool for Developmental Biology and Disease Modeling [Life Sciences

This paper discusses the challenges of automated quantitative phenotyping of zebrafish images. These challenges can be addressed by developing automated image analysis tools and integrating them in user-friendly software packages. As part of this strategic direction, we discussed the ZFIQ (zebrafish image quantitator) image processing pipeline and two examples, which show significant advantages over traditional methods based on manual labeling or visual inspection. By building ZFIQ on top of the public domain OpenCV platform, we aim to encourage the computer vision and signal/image processing communities to contribute their image quantification pipelines to the ZFIQ, and, in turn, expect the zebrafish research community to contribute quantification problems and associated image datasets to the development

Irregular shapes classification by back-propagation neural networks

This paper proposes a back-propagation neural network approach to classify irregular shapes by their convex hulls and brightness in automated production lines. An image-based defect detection and classification system is established, which would examine the quality of rolls of aluminum foil and determine the type of defects, such as bolt, fracture, scratch, or spot on the aluminum foil. The developed approach is capable of performing image acquisition, image processing, defect detection and recognition, and, subsequently, the classification of the aluminum foil sheets by a back-propagation neural network. In order to verify the effectiveness of the developed approach, ten-fold cross-validation is used. The experimental results show that, using a small number of training iterations, the average accuracy rate of classification reaches 96.4%. Thus, the developed approach can be used to replace manual visual inspection for process control and process improvement, which significantly reduces the cost of labor and increases the consistency of product quality.

Making Microtubules and Mitotic Spindles in Cells without Functional Centrosomes

Centrosomes are considered to be the major sites of microtubule nucleation in mitotic cells (reviewed in ), yet mitotic spindles can still form after laser ablation or disruption of centrosome function . Although kinetochores have been shown to nucleate microtubules, mechanisms for acentrosomal spindle formation remain unclear. Here, we performed live-cell microscopy of GFP-tubulin to examine spindle formation in Drosophila S2 cells after RNAi depletion of either gamma-tubulin, a microtubule nucleating protein, or centrosomin, a protein that recruits gamma-tubulin to the centrosome. In these RNAi-treated cells, we show that poorly focused bipolar spindles form through the self-organization of microtubules nucleated from chromosomes (a process involving gamma-tubulin), as well as from other potential sites, and through the incorporation of microtubules from the preceding interphase network. By tracking EB1-GFP (a microtubule-plus-end binding protein) in acentrosomal spindles, we also demonstrate that the spindle itself represents a source of new microtubule formation, as suggested by observations of numerous microtubule plus ends growing from acentrosomal poles toward the metaphase plate. We propose that the bipolar spindle propagates its own architecture by stimulating microtubule growth, thereby augmenting the well-described microtubule nucleation pathways that take place at centrosomes and chromosomes.

Comparing human “local” bone versus human cancellous bone for spinal fusion in an athymic rat model

Yu-Po Lee, MD, Los Angeles, CA, USA; Linda E.A. Kanim, MA, Santa Monica, CA, USA; Jeffrey C. Wang, MD, Los Angeles, CA, USAIntroduction: “Local” bone graft harvested from spinous processes and lamina is sometimes used as a source of graft for accompanying spinal fusion procedures. Local bone graft is periodically used as a substitute for iliac crest bone in order to avoid the morbidity associated with harvesting iliac crest bone graft. Local bone has also been used to augment the available iliac crest bone or mixed with other grafting materials. Local bone is characteristically more cortical compared with iliac crest bone, which is more cancellous. The ability of this cortical bone to induce de novo bone growth and eventual fusion has not been thoroughly tested in spite of its increasing use. Further, the relative effect between local versus iliac crest on the rapidity and/or the rate of fusion remains unknown. The athymic rat is a useful animal model to study the relative differences of these bone grafts, because it provides a nonimmunogenic environment for in vivo evaluation of human tissues for bone induction and spinal fusion. This is a preliminary study to compare both the rapidity and the rate of fusion between local bone and iliac crest bone.Methods: Mature athymic nude female rats were used in this study (175 to 240 g, Harlan Sprague Dawley, IN). The rats each were implanted with either human bone (0.6 cc per side in one group) obtained from the spinous processes of patients undergoing laminectomy (cortical bone) or human bone (0.6 cc per side in one group) obtained from the iliac crests of patients (cancellous bone). All procedures were performed through a midline incision, exposing the transverse processes. Decortication was performed on the L4 and L5 transverse processes (lamina and facet joints were left intact without decortication). An aliquot equal to 0.6 cc was measured in a 1.0 cc syringe and placed into each posterolateral gutter spanning the transverse processes for a total of 1.2 cc of graft per animal. The wound was irrigated and closed. High-resolution radiographs were obtained monthly. The presence or absence of remodeling was evaluated on the radiographs. Explanted lumbar spines were manually tested for intersegmental motion after sacrifice. Any motion detected by manual testing was considered a failure of fusion. Absence of any motion was considered a successful fusion. Histological analysis consisted of nondecalcified sections.Results: All of the animals implanted with human iliac crest bone showed radiographic evidence of new osseous bone formation and incorporation up to 8 weeks time. Conversely, none of the animals implanted with human local bone showed significant new osseous bone formation or incorporation up to the 8-week time point. Comparing the radiograph, the animals implanted with human iliac crest bone appeared more mature than the other group. Manual palpation and visual inspection of the explanted spines confirmed these results. At each time point, the human iliac crest bone demonstrated greater stiffness with more restriction of motion. However, none of the animals showed complete fusion by manual palpation at 8 weeks.Conclusion: This is the first study to directly test the ability of cortical versus cancellous human bone to induce spinal fusion in an athymic rat model. The athymic nude rat model allows for human tissues to be tested, because the rat does not reject the human tissue. This study demonstrated differences in the osteoinductive ability of human local bone versus human cancellous bone. These data suggest that human local bone is inferior to human iliac crest bone. Hence, more research is needed before clinical substitution of human iliac crest bone with human local bone.

Automated X-ray inspection of chip-to-chip interconnections of Si-on-Si MCM's

The silicon-on-silicon (Si-on-Si) multichip module (MCM) is one of the newest system integration technologies that has the potential for high functional integration, enhanced performance, and cost effectiveness. Aside from functional tests, X-ray inspection is required to ensure good quality chip-to-chip interconnections during the fabrication process. This paper presents an automated inspection system that is capable of detecting defects such as swollen solders, misaligned solders, missing solders, solder robbing, and solder bridging in a semi-finished MCM. The semi-finished MCM is in wafer form and has completed the following operations: stencil printing device placing, and solder reflowing. The defect detection methodology is detailed. Over a test set of 54 sample images of wafer tiles, 100% inspection accuracy was obtained. This system has the potential to automate the manual visual inspection operation which is tedious, slow, and error-prone. >