Industrial Quality Control Research Articles

Efficient fabric anomaly detection: A transfer learning framework with expedited training times

In industrial quality control, anomaly detection plays a critical role in identifying defective products. However, because of the rarity and time-consuming nature of defect collection, training models often rely solely on defect-free samples. This necessitates the use of unsupervised anomaly-detection techniques trained exclusively on defect-free data. Alternatively, defect data can be synthesized to augment the dataset with defective samples. In the textile industry, expeditious model training is crucial to ensure a smooth production flow. Unfortunately, most unsupervised methods require extensive training time. This paper proposes a novel transfer learning approach designed to achieve training times in seconds while effectively adapting the model to the target domain of fabric anomaly detection. The key contributions of our method include significantly reduced training times, up to 10 times faster than current state-of-the-art methods, and comparable performance in anomaly detection, achieving results on par with state-of-the-art approaches on benchmark datasets (MVTEC Anomaly Detection, TILDA, AITEX and DAGM). Additionally, our approach improves inference times, ensuring expedited and efficient anomaly detection during production. The proposed method offers a practical and efficient solution for real-time industrial quality control.

Optimized near infrared reflectance model based on preprocessing selection strategy for rapid analysis of 2-acetyl-1-pyrroline content in rice flour

The chemical compound 2-acetyl-1-pyrroline (2-AP) is a major contributor to the fragrance of rice. This report presents the first efforts towards determining the quantities of 2-AP in rice by grating-based visible–near-infrared reflectance spectroscopy using a diverse set of rice varieties. Calibration models generated by preprocessing treatments using Design of Experiment (DoE) and ‘one-variable-at-atime’ (OVAT) methods were compared and optimized. A modified partial least squares model with “2, 3, 1, 1”/SNV and 174 variables yielded a coefficient of determination for calibration values of 0.978. The results showed that a DoE-based preprocessing selection could efficiently and accurately optimize 2-AP calibration models, with advantages over OVAT. The findings of this study provide high-throughput screening technology for high-precision fragrance quality breeding, monitoring and control in the rice industry. The method demonstrated here also has potential for application in many other disciplines, including identification of plant fragrances and quality control of cereal products.

Intelligent Framework Design for Quality Control in Industry 4.0

This research aims to develop an intelligent framework for quality control and fault detection in pre-production and post-production systems in Industry 4.0. In the pre-production system, the health of the manufacturing machine is monitored. In this study, we examine the gear system of induction motors used in industries. In post-production, the product is tested for quality using a machine vision system. Gears are fundamental components in countless mechanical systems, ranging from automotive transmissions to industrial machinery, where their reliable operation is vital for overall system efficiency. A faulty gear system in the induction motor directly affects the quality of the manufactured product. Vibration data, collected from the gear system of the induction motor using vibration sensors, are used to predict the motor’s health condition. The gear system is monitored for six different fault conditions. In the second part, the quality of the final product is inspected with the machine vision system. Faults on the surface of manufactured products are detected, and the product is classified as a good or bad product. The quality control system is developed with different deep learning models. Finally, the quality control framework is validated and tested with the evaluation metrics.

Microwave-assisted sample preparation for screening of heavy metal elements in food additives by ICP-MS

The screening of heavy metals in food additives is important for public health due to their probable critical effects, induced by excessive intakes, and difficulties in tracing in food matrix. This study aimed to assess the applicability of microwave (MW)-based digestion and inductively coupled plasma-mass spectrometry (ICP-MS) for quantification of heavy metals in food additives. MW-assisted diluted acid digestion (1550 W at 220 °C, 0.79 mol/L of HNO3 solution) was applied as a pretreatment and ICP-MS was used for 19 different heavy metals at different concentrations, up to 20 ng/g. The optimized protocol was validated by spiking heavy metals to glycerin and applied to 18 food additives. The method was linear for 14 heavy metals (R2 > 0.990). For 11 heavy metals, both intra- and inter-day accuracies in glycerin were estimated to 88.5%–123.7%, with intra- and inter-day precisions of less than 5.5% and 4.9%, respectively, and their contents in food additives were discussed. The findings in this work provide an alternative analytical method for regulatory compliance and quality control in food industries.

Efficiency of natural and forced convection systems in cooking ham: Industrial scale validation

AbstractThis study evaluated the efficacy of cooking processes in natural and forced convection systems, for ensuring microbiological safety in industrial‐scale ham production. The cold point of the tanks was determined by measuring the water temperature every 2 min, with a maximum variation of 1.4°C between different regions of the tanks. The calculation of the lethal value (Fcal) was carried out using Enterococcus faecalis as the target microorganism due to its high resistance to heat treatment. The cooking temperature was measured at the geometric center of the ham and, in both convection systems, it reached the Fref value (23.6 min), necessary to achieve a reduction of 8 log (8 D), with Fcal of 45.7 and 45 min of cooking in the natural and forced convection systems. Both convection systems achieved microbial reductions exceeding 15‐log cycles (15 D), surpassing regulatory requirements despite occasional core temperatures below standard thresholds. In both systems (natural and forced), the cold spot was found on the left side, at a lower height, and toward the front of the tank. After heat treatment, no microorganisms were detected in the ham cooked in natural or forced convection systems. Both systems (natural and forced convection) proved viable at an industrial scale, provided that appropriate cooking protocols are followed.Practical ApplicationsThe practical applications of this study are significant for the food industry and regulatory bodies involved in food safety. These include quality assurance in ham production, food safety compliance, process optimization, risk mitigation, and guidance for industrial practices. Overall, this study has practical implications for improving food safety, quality control, and production efficiency in the ham industry. It fills a gap in validation studies on industrial‐scale ham cooking, providing valuable information for industry stakeholders and regulatory bodies alike.

Modern methods for controlling and regulating impurities in pharmaceuticals

The article is dedicated to modern principles of regulation and control of impurities in pharmaceutical products and active pharmaceutical substances, which is a crucial task for the pharmaceutical industry. The relevance of this research stems from the fact that the presence of impurities in pharmaceutical products can significantly affect their safety and efficacy, making the development of stringent control methods throughout the entire life cycle of a pharmaceutical product essential. Even small amounts of impurities can cause adverse effects or reduce the therapeutic value of drugs, as evidenced by historical examples. The aim of the study is to analyze and systematize contemporary approaches to the detection and regulation of impurities in pharmaceutical products. The paper reviews current regulatory requirements, including guidelines from the The International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) and regulatory documents of the Eurasian Economic Community, as well as national pharmacopeias of the Russian Federation, the United States of America, and the European Union. Conclusion. Additionally, the article emphasizes the need for harmonization of standards and practices for impurity control, which is especially relevant in the context of the globalization of the pharmaceutical market. Harmonization not only enhances the safety and efficacy of pharmaceutical products but also improves patients' quality of life by reducing the risk of adverse effects associated with impurities. The article also explores prospects for further improvement of regulatory acts and the implementation of advanced technologies aimed at enhancing standardization and quality control in the pharmaceutical industry.

Dielectric properties of Pouteria sapota compounds in terahertz frequency range

The analysis and a mathematical approximation of the dielectric properties of Pouteria sapota (Mamey) pulp compounds in the Terahertz (THz) range are presented. Since THz electromagnetic waves show a high sensitivity to detect water content when interacting with organic materials, this technique was used to determine the effective dielectric function (EDF) of a wet and a dry sample from the THz spectrum data. By fitting the EDF to the Landau-Lifshitz-Looyenga (LLL) dielectric mixture equation, the theoretical dielectric functions of each pulp compounds were approximated. The approach proposed here not only considered dry matter and water content in the Pouteria sapota pulp, as has been reported in the literature for organic samples, but also included the contribution of the volatile components (VC) that allowed a better adjustment to the experimental results. The water and VC content increase the value of the real and imaginary part of the calculated EDF of the wet sample. The use of the non-destructive THz spectroscopy technique allows determining the dielectric properties of fruits that have been dehydrated. Therefore, the methodology proposed here can be used as a method of in situ and ex situ quality control in the agri-food industry.

Experimental review and accuracy of etchants used for phase analysis of SAF2507 Super Duplex Stainless Steel

The mechanical and corrosion properties of duplex stainless steels (DSS) depend on the distribution of ferrite and austenite as well as the presence of secondary phases. It is therefore necessary to accurately determine both the distribution of the phases and their morphology. The aim of this research is to define a standard method for the determination of the volume fraction of the phases present in a SAF2507 SDSS by analysing images acquired by light microscope. To do this, the material is heat-treated at 850 °C for 5 min, 15 min, 30 min, 1h, 5h to stimulate the formation of the σ-phase. The metallographic etchant capable of maximising phase contrast is then identified. Micrographic analyses are performed using different reagents at different holding times and temperatures, in order to find the best combination. A standard procedure is then defined for the image analysis. The data obtained were are processed to calculate the phase distribution. The results are compared with the quantitative analysis performed using X-ray diffraction as reference.The research evidence that Kalling's reagent is best etchant for fully solubilised materials, as it accurately estimates ferrite and austenite content. On the other hand, Marble's reagent should be preferred for the detection of σ-phase. For comprehensive analysis of all phases, Murakami's reagent is recommended.The approach defined in this paper not only improves the understanding of the microstructure of DSS, but also provides an important tool for industrial quality control and material characterisation.

Assessment of Lead and Iron Concentration in Cosmetics Traded in the AL-Ajilat City Market

This study examined the concentrations of heavy metals, namely lead (Pb) and iron (Fe), in lipsticks, foundations, and face powders to ensure compliance with international safety standards and assess potential health risks. Using atomic absorption spectroscopy (AAS), samples were analyzed for heavy metal content and compared to WHO, FDA, and Health Canada guidelines. Results included that lead levels in lipsticks ranged from 0.03 to 6.15 ppm, with one sample exceeding the WHO limit of 0.01 ppm, raising safety concerns. Iron levels were alarmingly high, with a maximum of 2,566 ppm, well above the WHO guideline of 0.3 ppm. Baseline samples consistently showed low levels of lead at 0.03 ppm, but iron concentrations reached 5,735 ppm, indicating significant safety deviations. Lead concentrations in the face powders ranged from 0.03 to 2.14 ppm, with some exceeding the WHO limit, while iron levels averaged 2,613 ppm, well above acceptable limits. The study concluded that high iron levels in these cosmetic products pose health risks, stressing the need for stricter regulatory oversight and quality control in the cosmetics industry.

Elemental and stable isotopic signatures for dynamic traceability of genuine and "bathing" cultured Yangcheng Eriocheir sinensis crabs

This study examines whether the "bathing" culture of Chinese mitten crabs Eriocheir sinensis outside the Yangcheng Lake area converges with the origin characteristics of crabs from Yangcheng Lake and Kunshan Ponds. Inductively coupled plasma mass spectrometry (ICP-MS) and stable isotope ratio mass spectrometry (IRMS) were used to analyze mineral elements and stable isotopes in crabs at various stages of "bathing". The results indicated a dynamic transformation in the origin characteristics of "bathed" crabs; however, full convergence with genuine crabs was not achieved. Notably, disparities persisted between genuine crabs and "bathed" crabs. The combination of mineral elements and stable isotopes exhibited a discrimination accuracy of 95 % and 81.3 % between genuine crabs and "bathed crabs" from Kunshan Ponds and the Yangcheng Lake during each period of “bathing", respectively, which was higher than that using either method alone. This suggests limitations in altering original characteristics through "bathing". The combination of mineral elements and stable isotopes provides promising avenues for origin identification and understanding changes in crab characteristics during the "bathing" process. These findings highlight the complexities of crab culture practices and have implications for quality control and origin authentication in the crab industry.

A review of organometallic compounds as versatile sensors in environmental, medical, and industrial applications

Abstract Sensing technology is gaining attention and continuously advancing, making it a recommended element of individualized healthcare management. This is due to the powers exhibited by organometallic compounds, which are further enhanced by the field of bioengineering. Organometallic compounds have a wide range of biological activity and find uses in industrial and material science fields. Their unique ability to specifically target and overcome constraints faced by traditional counterparts makes them potential contenders for sensor technology. These compounds are highly sensitive to changes in their environment, allowing them to be utilized as sensors for detecting various chemicals or conditions. Additionally, the versatility of organometallic compounds enables their integration into different sensor platforms, making them suitable for environmental monitoring, medical diagnostics, and industrial quality control. This article provides a comprehensive summary of recent advancements in the design and synthesis of organometallic compounds, with a specific emphasis on their potential use as sensors. It also discusses the changes made to the structure, the processes used for functionalization, the incorporation of microfluidics, and the resulting impact on the materials’ sensing capabilities. These biologically derived methods align with sustainability goals and enhance the affordability, applicability, and effectiveness of sensing.

Analysis of the Volatile and Enantiomeric Compounds Emitted by Plumeria rubra L. Flowers Using HS-SPME-GC.

The volatile components emitted by fresh aromatic flowers of Plumeria rubra L., harvested in southern Ecuador during three different months were determined to evaluate the fluctuation of secondary metabolites. The volatile compounds were analyzed using headspace solid-phase microextraction (HS-SPME) followed by gas chromatography coupled to mass spectrometry (GC-MS) and a flame ionization detector (GC-FID) using two types of columns: a non-polar (DB-5ms) and polar column (HP-INNOWax). The principal chemical groups were hydrocarbon sesquiterpenes (43.5%; 40.0%), oxygenated sesquiterpenes (23.4%; 26.4%), oxygenated monoterpenes (14.0%; 11.2%), and hydrocarbon monoterpenes (12.7%; 9.3%). The most representative constituents were (E,E)-α-Farnesene (40.9-41.2%; 38.5-50.6%), (E)-nerolidol (21.4-32.6%; 23.2-33.0%), (E)-β-ocimene (4.2-12.5%; 4.5-9.1%), (Z)-dihydro-apofarnesol (6.5-9.9%; 7.6-8.6%), linalool (5.6-8.3%; 3.3-7.8%), and perillene (3.1-5.9%; 3.0-3.2%) in DB-5ms and HP-INNOWax, respectively. Finally, we reported for the first time the enantiomeric distribution of P. rubra flowers, where the enantiomers (1R,5R)-(+)-α-pinene, (S)-(-)-limonene, (S)-(+)-Linalool, and (1S,2R,6R,7R,8R)-(+)-α-copaene were present as enantiomerically pure substances, whereas (S)-(+)-(E)-Nerolidol and (R)-(+)-(E)-Nerolidol were observed as scalemic mixtures. This study provides the first comprehensive and comparative aroma profile of Plumeria rubra cultivated in southern Ecuador and gave us a clue to the variability of P. rubra chemotypes depending on the harvesting time, which could be used for future quality control or applications in phytopharmaceutical and food industries.

Task-Adaptive Angle Selection for Computed Tomography-Based Defect Detection.

Sparse-angle X-ray Computed Tomography (CT) plays a vital role in industrial quality control but leads to an inherent trade-off between scan time and reconstruction quality. Adaptive angle selection strategies try to improve upon this based on the idea that the geometry of the object under investigation leads to an uneven distribution of the information content over the projection angles. Deep Reinforcement Learning (DRL) has emerged as an effective approach for adaptive angle selection in X-ray CT. While previous studies focused on optimizing generic image quality measures using a fixed number of angles, our work extends them by considering a specific downstream task, namely image-based defect detection, and introducing flexibility in the number of angles used. By leveraging prior knowledge about typical defect characteristics, our task-adaptive angle selection method, adaptable in terms of angle count, enables easy detection of defects in the reconstructed images.

Innovative strategies for protein content determination in dried laver (Porphyra spp.): Evaluation of preprocessing methods and machine learning algorithms through short-wave infrared imaging

In this study, we explored the application of Short-Wave Infrared (SWIR) hyperspectral imaging combined with Competitive Adaptive Reweighted Sampling (CARS) and advanced regression models for the non-destructive assessment of protein content in dried laver. Utilizing a spectral range of 900–1700 nm, we aimed to refine the quality control process by selecting informative wavelengths through CARS and applying various preprocessing techniques (standard normal variate [SNV], Savitzky-Golay filtering [SG], Orthogonal Signal Correction [OSC], and StandardScaler [SS]) to enhance the model's accuracy. The SNV-OSC-StandardScaler- Support vector regression (SVR) model trained on CARS-selected wavelengths significantly outperformed the other configurations, achieving a prediction determination coefficient (Rp2) of 0.9673, root mean square error of prediction of 0.4043, and residual predictive deviation of 5.533. These results highlight SWIR hyperspectral imaging's potential as a rapid and precise tool for assessing dried laver quality, aiding food industry quality control and dried laver market growth.

AIoT-enabled defect detection with minimal data: A few-shot learning approach combining prototypical and relational networks for smart manufacturing

Defect detection is crucial in manufacturing processes but traditional AI-based algorithms require large datasets for accurate results. For new or customized products, the number of images with detected defects is limited. Therefore, we developed a few-shot learning approach integrating a prototypical and relation network (PRN), algorithms with meta-learning, and the Artificial Internet of Things (AIoT). For rapid defect detection with IoT sensors, such minimal data are used for a smart manufacturing ecosystem., making it ideal for dynamic production environments. We tested the AIOT-enhanced PRN on two datasets using the following data augmentation methods: random rotation and horizontal translation (RH), random rotation and vertical translation (RV), and horizontal and vertical translation (HV). The developed PRN efficiently learned from minimal data to reduce the occurrence of overfitting issues in the MVTec 3D-AD dataset which are caused by a limited number of defect sample images. When testing the AIOT-enhanced PRN with the NEU-DET dataset, accuracies in 5-way 5-shot settings using RV, RH, 15° rotation, and HV were 100 %. Under Gaussian noise, the AIOT-enhanced PRN showed an accuracy of 100 % in 5-way 5-shot and 5-way 1-shot scenarios using HV. For salt-and-pepper noise, the accuracy of the AIOT-enhanced PRN ranged from 98.49 to 99.04 %. The developed AIOT-enhanced PRN improved defect detection accuracy and real-time monitoring capability with minimal data. The developed AIOT-enhanced PRN can be used for efficient and flexible product quality control in Industry 4.0.

AEDN-YOLO: an efficient one-stage detection network for strip steel surface defects

Steel surface defect detection is one of the key tasks in industrial production and quality control. Research on defect detection using deep learning algorithms has shown promising results. However, due to the complex backgrounds, large differences in defect sizes, and diverse defect types present in steel strip surface defect images, existing deep learning algorithms struggle to achieve precise detection. To address these challenges, this paper proposes an efficient detection model named AEDN-YOLO. Firstly, an adaptive feature extraction (AFE) module is designed, embedded into C2f to better capture irregularly shaped objects. Secondly, the Triplet Attention module is incorporated into the bottom layer of the backbone network to enhance the model’s ability to locate defect features accurately. Additionally, replace the standard convolution in the neck network with GSConv, which not only accelerates feature fusion to improve detection speed but also enlarges the model’s receptive field to enhance detection accuracy. Finally, add a small target detection layer to enhance the detection capability for tiny defects. The model achieves mAP of 81.7% and 72.7% on the NEU-DET and GC10-DET datasets, respectively, with a detection speed of 72.1 FPS. Compared to mainstream defect detection algorithms, the proposed algorithm enables accurate and efficient detection of steel surface defects.

DNA metabarcoding unveils the hidden species composition in fish surimi: Implications for the management of unlabeled and mixed seafood products

Fish surimi products are traditional foods primarily made from fish meat and may contain a complex species composition. In Taiwan, the abundant fishery resources and diverse fish species lead to local catches being widely used as ingredients in fish surimi products. However, due to growing market demand and increasingly scarce resources, some surimi products contain sensitive species, such as sharks, posing potential threats to the ecological environment and biodiversity. In this study, by applying metabarcoding techniques, we analyzed 120 fish surimi product samples from different brands and types throughout the four seasons in Taiwan's market. The main fish species identified included milkfish (Chanos chanos), dolphinfish (Coryphaena hippurus), Pomfret (Taractes rubescens), swordfish (Istiophorus spp.) and cartilaginous. Moreover, at least 37 species of cartilaginous fish, including 26 endangered species, were found. Through comprehensive and accurate species identification of surimi product ingredients, we unveiled the usage of sensitive species in products on the market. This finding is important for the surimi industry's quality control and market supervision. Furthermore, it can promote the sustainable use of Taiwan's fishery resources and protect biodiversity.

A multi-scale attention mechanism for detecting defects in leather fabrics

Defect detection is critical to industrial quality control in leather production engineering. The various sizes and locations of defects in leather, as well as significant differences within the same class and indistinctive variations between different classes of defects, contribute to the complexity of the problem. To address this challenge, we propose a Multi-Layer Residual Convolutional Attention (MLRCA) approach. MLRCA enhances its ability to capture both intra-class and inter-class differences by enhancing the semantic feature representation in the backbone network. To improve multiscale fusion effects, we also incorporate the MLRCA module into the feature pyramid network (FPN) and propose a new multi-layer residual convolution attention feature pyramid network (ML-FPN). This approach enables more accurate identification of leather defects at a more detailed level by selectively capturing contextual information from different domains. We then implement the Side-Aware Boundary Localization (SABL) detection head, which accurately locates defects and helps the network distinguish between similar defect categories for more precise positioning. To validate the effectiveness of our approach, we conducted ablation experiments on the created leather dataset. Comparative experiments demonstrate the excellent capability of our model to detect minor defects. The model achieved 83.4, 89.7, and 85.6 for the AP, AP50, and AP75 evaluation metrics. In addition, the model achieves 71.3, 89.9, and 88.9 for APS, APM, and APL. Our approach has been confirmed feasible through experimentation and provides new insights for automated leather defect detection methods.

Lipidomic Profiling of Flammulina velutipes (Curtis) Singer (Agaricomycetes) through Ultra-Performance Liquid Chromatography Coupled with Quadrupole Time-of-Flight Mass Spectrometry: Examining Lipid Dynamics Changes during Fruiting Body Formation and Development.

Flammulina velutipes (enokitake) is widely recognized for its nutritional and medicinal properties. Understanding the biochemical processes, such as lipid metabolism during fruiting body formation, is essential for enhancing mushroom cultivation and utilization. This study aimed at elucidating the dynamic lipidomic changes during seven growth stages of F. velutipes using ultra-performance liquid chromatography coupled with quadrupole time-of-flight mass spectrometry. Our results revealed significant increases in ceramides along with the growth and a sharp decline in phosphatidylinositols from mycelial to primordial stages. Fatty acid esters of hydroxy fatty acids, recently discovered for their bioactivities, showed high intensities in the mycelial and primordial stages but decreased rapidly thereafter. These findings provide profound insights into the lipid profiles associated with mushroom morphology and development. This lipidomics study establishes a foundational understanding for future research in agricultural and food chemistry applications, potentially improving industrial production and quality control of F. velutipes.

Fabric defect detection based on feature enhancement and complementary neighboring information

Abstract Fabric defect detection is a crucial aspect of quality control in the textile industry. Given the complexities of fabric backgrounds, the high similarity between patterned backgrounds and defects, and the variety of defect scales, we propose a fabric defect detection method based on feature enhancement and complementary neighboring information . The core of this method lies in two main components: the feature enhancement module and the neighboring information complementation strategy. The feature enhancement module includes two sub-modules: similarity feature enhancement(SFE) and edge detail feature enhancement(EDFE). The SFE aims to capture the similarities between features to strengthen the distinction between defects and complex backgrounds, thereby highlighting the correlations among defects and the differences between defects and the background. The EDFE focuses on improving the network's ability to capture the edge details of fabrics, preventing edge information from becoming blurred or lost due to deeper network layers. The neighboring information complementation strategy consists of shallow-level information complementation(SLIC) and top-down information fusion complementation(TDIFC). The SLIC integrates newly introduced shallow features with neighboring features that have a smaller semantic gap, injecting richer detail information into the network. The TDIFC adaptively guides the interaction of information between adjacent feature maps, effectively aggregating multi-scale features to ensure information complementarity between features of different scales. Additionally, to further optimize model performance, we introduced partial convolution(PConv) in the backbone of the feature extraction network. PConv reduces redundant computations and decreases the model's parameter count. Experimental results show that our proposed method achieved an mAP@50 of 82.4%, which is a 6.6% improvement over the baseline model YOLOv8s. The average inference frame rate reached 61.8 FPS, meeting the real-time detection requirements for fabric defects. Moreover, the model demonstrated good generalization capabilities, effectively adapting to detecting defects in different types and colors of fabrics.