Search Results (211)

Weeds continue to pose a serious challenge to agriculture, reducing both the productivity and quality of crops. In this paper, we explore how modern deep learning, specifically Vision Transformers (ViTs), can help address this issue through fast and accurate weed classification. We developed a transformer-based model trained on the DeepWeeds dataset, which contains images of nine different weed species collected under various environmental conditions, such as changes in lighting and weather. By leveraging the ViT architecture, the model is able to capture complex patterns and spatial details in high-resolution images, leading to improved prediction accuracy. We also examined the effects of model optimization techniques, including fine-tuning and the use of pre-trained weights, along with different strategies for handling class imbalance. While traditional oversampling actually hurt performance, dropping accuracy to 94%, using class weights alongside strong data augmentation boosted accuracy to 96.9%. Overall, our ViT model outperformed standard Convolutional Neural Networks, achieving 96.9% accuracy on the held-out test set. Attention-based saliency maps were inspected to confirm that predictions were driven by weed regions, and model consistency under location shift and capture perturbations was assessed using the diverse acquisition sites in DeepWeeds. These findings show that with the right combination of model architecture and training strategies, Vision Transformers can offer a powerful solution for smarter weed detection and more efficient farming practices. Full article

The quality of input data is critical to the performance of time-series classification models, particularly in the domain for industrial sensor data where noise and anomalies are frequent. This study investigates how various filtering-based preprocessing techniques impact the accuracy and robustness of a Transformer model that predicts power efficiency states (Normal, Caution, Warning) from minute-level IIoT sensor data. We evaluated five techniques: a baseline, Simple Moving Average, Median filter, Hampel filter, and Kalman filter. For each technique, we conducted systematic experiments across time windows (360 and 720 min) that reflect real-world industrial inspection cycles, along with five prediction offsets (up to 2880 min). To ensure statistical robustness, we repeated each experiment 20 times with different random seeds. The results show that the Simple Moving Average filter, combined with a 360 min window and a short-term prediction offset, yielded the best overall performance and stability. While other techniques such as the Kalman and Median filters showed situational strengths, methods focused on outlier removal, like the Hampel filter, adversely affected performance. This study provides empirical evidence that a simple and efficient filtering strategy such as Simple Moving Average, can significantly and reliably enhance model performance for power efficiency prediction tasks. Full article

The efficient and non-destructive inspection of the chlorophyll content of grey jujube leaf is of great significance for its growth surveillance and nutritional diagnosis. Near-infrared spectroscopy combined with chemometric methods provides an effective approach to achieve this goal. This study took grey jujube leaves as the research object, systematically collected near-infrared spectral data in the range of 4000–10,000 cm⁻¹, and simultaneously measured their soil and plant analyzer development (SPAD) value as a reference index for chlorophyll content. Through various pretreatment and their combination methods on the original spectrum—smooth, standard normal variable transformation (SNV), first derivative (FD), second derivative (SD), smooth + first derivative (Smooth + FD), smooth + second derivative (Smooth + SD), standard normal variable transformation + first derivative (SNV + FD), standard normal variable transformation + second derivative (SNV + SD)—the effects of different methods on the quality of the spectrum and its correlation with SPAD value were compared. The competitive adaptive reweighted sampling algorithm (CARS) was adopted to extract the characteristic wavelength, aiming to reduce data dimensionality and optimize model input. Both BP neural network and RBF neural network prediction models were established, and the model performance under different training functions was compared. The results indicate that after Smooth + FD pretreatment, followed by CARS screening of the characteristic wavelength, the BP neural network model trained using the LBFGS algorithm demonstrated the best performance, with its coefficient of determination (R²) of 0.87 (training set) and 0.85 (validation set), root mean square error (RMSE) of 1.36 (training set) and 1.35 (validation set), and residual prediction deviation (RPD) of 2.81 (training set) and 2.56 (validation set) showing good prediction accuracy and robustness. Research indicates that by combining near-infrared spectroscopy with feature extraction and machine learning methods, the rapid and non-destructive inspection of the grey jujube leaf SPAD value can be achieved, providing reliable technical support for the real-time monitoring of the nutritional status of jujube trees. Full article

The integration of machine vision systems with programmable logic controllers (PLCs) is increasingly crucial for automated quality assurance in Industry 4.0 environments. This paper presents an applied case study of vision–PLC integration, focusing on real-time synchronization, deterministic communication, and practical industrial deployment. The proposed platform combines a Cognex In-Sight 2802C smart camera (Cognex Corporation, Natick, MA, USA) with an Allen-Bradley Compact GuardLogix PLC through Ethernet/IP implicit cyclic exchange. Three representative case studies were investigated: 3D-printed prototypes with controlled defects, automotive electrical connectors inspected using Cognex ViDi supervised learning tools, and fiber optic tubes evaluated via a custom fixture-based heuristic method. Across all scenarios, detection accuracy exceeded 95%, while PLC-level triple verification reduced false classifications by 28% compared to camera-only operation. The work highlights the benefits of PLC-driven inspection, including robustness, real-time performance, and dynamic tolerance adjustment via HMI interfaces. At the same time, several limitations were identified, including sensitivity to lighting variations, limited dataset size, and challenges in scaling to full production environments. These findings demonstrate a replicable integration framework that supports intelligent manufacturing. Future research will focus on hybrid AI–PLC architectures, extended validation on industrial production lines, and predictive maintenance enabled by edge computing. Full article

In ocean engineering, path following serves as a fundamental capability for autonomous underwater vehicles (AUVs), enabling essential operations such as environmental exploration and inspection. However, for robotic dolphins employing dorsoventral undulatory propulsion, the periodic pitching induces strong coupling between propulsion and attitude, posing significant challenges for precise path following in disturbed environments. In this paper, a real-time robust path-following control framework is proposed for robotic dolphins to address these challenges. First, a novel robotic dolphin platform is presented by integrating a dorsoventral propulsion mechanism with a passive peduncle joint, followed by the systematic formulation of a full-state dynamic model. Then, a minimum-snap-based path optimizer is constructed to generate smooth and dynamically feasible trajectories, improving path quality and motion safety. Subsequently, a robust model predictive controller is developed, which incorporates control surface dynamics, a nonlinear disturbance observer, and a Sigmoid-based disturbance-grading mechanism to ensure fast attitude response and precise tracking performance. Finally, extensive simulations under various environmental disturbances validate the effectiveness of the proposed approach in both trajectory optimization and robust path following. The proposed framework not only demonstrates strong robustness in path following and disturbance rejection, but also provides practical guidance for future underwater missions such as long-term environmental monitoring, inspection, and rescue. Full article

Visual inspection remains a cornerstone of quality control in welded structures, yet manual evaluations are inherently constrained by subjectivity, inconsistency, and limited scalability. This study presents WeldVGG, a deep learning-based visual inspection model designed to automate weld defect classification using radiographic imagery. The proposed model is trained on the RIAWELC dataset, a publicly available collection of X-ray weld images acquired in real manufacturing environments and annotated across four defect conditions: cracking, porosity, lack of penetration, and no defect. RIAWELC offers high-resolution imagery and standardized class labels, making it a valuable benchmark for defect classification under realistic conditions. To improve trust and explainability, Grad-CAM++ is employed to generate class-discriminative saliency maps, enabling visual validation of predictions. The model is rigorously evaluated through stratified cross-validation and benchmarked against traditional machine learning baselines, including SVC, Random Forest, and a state-of-the-art architecture, MobileNetV3. The proposed model achieves high classification accuracy and interpretability, offering a practical and scalable solution for intelligent weld inspection. Furthermore, to prove the model’s ability to generalize, a test on the GDXray was performed, yielding positive results. Additionally, a Wilcoxon signed-rank test was conducted separately to assess statistical significance between model performances. Full article

The increasing complexity of modern food production demands advanced solutions for quality control (QC), safety monitoring, and process optimization. This review systematically explores recent advancements in machine learning (ML) for QC across six domains: Food Quality Applications; Defect Detection and Visual Inspection Systems; Ingredient Optimization and Nutritional Assessment; Packaging—Sensors and Predictive QC; Supply Chain—Traceability and Transparency and Food Industry Efficiency; and Industry 4.0 Models. Following a PRISMA-based methodology, a structured search of the Scopus database using thematic Boolean keywords identified 124 peer-reviewed publications (2005–2025), from which 25 studies were selected based on predefined inclusion and exclusion criteria, methodological rigor, and innovation. Neural networks dominated the reviewed approaches, with ensemble learning as a secondary method, and supervised learning prevailing across tasks. Emerging trends include hyperspectral imaging, sensor fusion, explainable AI, and blockchain-enabled traceability. Limitations in current research include domain coverage biases, data scarcity, and underexplored unsupervised and hybrid methods. Real-world implementation challenges involve integration with legacy systems, regulatory compliance, scalability, and cost–benefit trade-offs. The novelty of this review lies in combining a transparent PRISMA approach, a six-domain thematic framework, and Industry 4.0/5.0 integration, providing cross-domain insights and a roadmap for robust, transparent, and adaptive QC systems in the food industry. Full article

Steel products play an important role in daily production and life as a common production material. Currently, the quality of steel products is judged by manual experience. However, various inspection criteria employed by human operators and complex factors and mechanisms in the steelmaking process may lead to inaccuracies. To address these issues, we propose a learning-based method for steel quality prediction, which is named DRBoost,based on multiple machine learning techniques, including Decision tree, Random forest, and the LSBoost algorithm. In our method, the decision tree clearly captures the nonlinear relationships between features and serves as a solid baseline for making preliminary predictions. Random forest enhances the model’s robustness and avoids overfitting by aggregating multiple decision trees. LSBoost uses gradient descent training to assign contribution coefficients to different kinds of raw materials to obtain more accurate predictions. Five key chemical elements, including carbon, silicon, manganese, phosphorus, and sulfur, which significantly influence the major performance characteristics of steel products, are selected. Steel quality prediction is conducted by predicting the contents of these chemical elements. Multiple models are constructed to predict the contents of five key chemical elements in steel products. These models are symmetrically complementary, meeting the requirements of different production scenarios and forming a more accurate and universal method for predicting the steel product’s quality. In addition, the prediction method provides a symmetric quality control system for steel product production. Experimental evaluations are conducted based on a dataset of 2012 samples from a steel plant in Liaoning Province, China. The input variables include various raw material usages, while the outputs are the content of five key chemical elements that influence the quality of steel products. The experimental results show that the models demonstrate their advantages in different performance metrics and are applicable to practical steelmaking scenarios. Full article

To maintain product reliability and stabilize performance, it is essential to prioritize the identification and resolution of latent defects. Advanced products such as high-precision electronic devices and semiconductors are susceptible to performance degradation over time due to environmental factors and electrical stress. However, conventional performance testing methods typically evaluate products based solely on predefined acceptable ranges, making it difficult to predict long-term degradation, even for products that pass initial testing. In particular, products exhibiting borderline values close to the threshold during initial inspections are at a higher risk of exceeding permissible limits as time progresses. Therefore, to ensure long-term product stability and quality, a novel approach is required that enables the early prediction of potential defects based on test data. In this context, the present study proposes a machine learning-based framework for predicting latent defects in products that are initially classified as normal. Specifically, we introduce the Sigma Deviation Count Labeling (SDCL) method, which utilizes a Gaussian distribution-based approach. This method involves preprocessing the dataset consisting of initially passed test samples by removing redundant features and handling missing values, thereby constructing a more robust input for defect prediction models. Subsequently, outlier counting and labeling are performed based on statistical thresholds defined by 2σ and 3σ, which represent potential anomalies outside the critical boundaries. This process enables the identification of statistically significant outliers, which are then used for training machine learning models. The experiments were conducted using two distinct datasets. Although both datasets share fundamental information such as time, user data, and temperature, they differ in the specific characteristics of the test parameters. By utilizing these two distinct test datasets, the proposed method aims to validate its general applicability as a Predictive Anomaly Testing (PAT) approach. Experimental results demonstrate that most models achieved high accuracy and geometric mean (GM) at the 3σ level, with maximum values of 1.0 for both metrics. Among the tested models, the Support Vector Machine (SVM) exhibited the most stable classification performance. Moreover, the consistency of results across different models further supports the robustness of the proposed method. These findings suggest that the SDCL-based PAT approach is not only stable but also highly adaptable across various datasets and testing environments. Ultimately, the proposed framework offers a promising solution for enhancing product quality and reliability by enabling the early detection and prevention of latent defects. Full article

The use of machine learning methods for the development of robust and flexible visual inspection systems has shown promising results. However, their performance is highly dependent on the large amount and diversity of training data, which is difficult to obtain in practice. Recent developments in synthetic dataset generation have seen increasing success in overcoming these problems. However, the prevailing work revolves around the usage of generative models, which suffer from data shortages, hallucinations, and provide limited support for unobserved edge-cases. In this work, we present the first synthetic data generation pipeline that is capable of generating large datasets of physically realistic textures exhibiting sophisticated structured patterns. Our framework is based on procedural texture modelling with interpretable parameters, uniquely allowing us to guarantee precise control over the texture parameters as we generate a high variety of observed and unobserved texture instances. We publish the dual dataset used in this paper, presenting models of sandblasting, parallel, and spiral milling textures, which are commonly present on manufactured metal products. To evaluate the dataset quality, we go beyond final model performance comparison by measuring different image similarities between the real and synthetic domains. This uncovered a trend, indicating these metrics could be used to predict downstream detection performance, which can strongly impact future developments of synthetic data. Full article

The structural integrity of poultry eggs is essential for food safety, economic value, and hatchability. External eggshell quality—measured by thickness, strength, cracks, color, and cleanliness—is a key criterion for grading and sorting. Traditional assessment methods, although simple, suffer from subjectivity, low efficiency, and destructive nature. In contrast, recent developments in nondestructive testing (NDT) technologies have enabled precise, automated, and real-time evaluation of eggshell characteristics. This review systematically summarizes state-of-the-art NDT techniques including acoustic resonance, ultrasonic imaging, terahertz spectroscopy, machine vision, and electrical property sensing. Deep learning and sensor fusion methods are highlighted for their superior accuracy in microcrack detection (up to 99.4%) and shell strength prediction. We further discuss emerging challenges such as noise interference, signal variability, and scalability for industrial deployment. The integration of explainable AI, multimodal data acquisition, and edge computing is proposed as a future direction to develop intelligent, scalable, and cost-effective eggshell inspection systems. This comprehensive analysis provides a valuable reference for advancing nondestructive quality control in poultry product supply chains. Full article

Fault identification in Printed Circuit Board Assembly (PCBA) testing is essential for assuring product quality; nevertheless, conventional methods still have difficulties due to the lack of labeled faulty data and the “black box” nature of advanced models. This study introduces a label-free, interpretable self-supervised framework that uses two pretext tasks: (i) an autoencoder (reconstruction error and two latent features) and (ii) isolation forest (faulty score) to form a four-dimensional representation of each test sequence. A two-component Gaussian Mixture Model is used, and the samples are clustered into normal and fault groups. The decision is explained with cluster mean differences, SHAP (LinearSHAP or LinearExplainer on a logistic-regression surrogate), and a shallow decision tree that generated if–then rules. On real PCBA data, internal indices showed compact and well-separated clusters (Silhouette 0.85, Calinski–Harabasz 50,344.19, Davies–Bouldin 0.39), external metrics were high (ARI 0.72; NMI 0.59; Fowlkes–Mallows 0.98), and the clustered result used as a fault predictor reached 0.98 accuracy, 0.98 precision, and 0.99 recall. Explanations show that the IForest score and reconstruction error drive most decisions, causing simple thresholds that can guide inspection. An ablation without the self-supervised tasks results in degraded clustering quality. The proposed approach offers accurate, label-free fault prediction with transparent reasoning and is suitable for deployment in industrial test lines. Full article

High sugar intake from commercial beverages is a public health concern, motivating rapid, user-friendly tools for sugar quantification. We present a compact planar microwave metamaterial sensor that estimates sugar concentration by monitoring resonant frequency shifts induced by dielectric loading. Tests with aqueous glucose solutions demonstrated a wide dynamic range (0 to 12,000 mg/dL), perfect linearity (R² = 1), and high repeatability. Validation on two commercial beverages showed sensor-predicted sugar contents consistent with their nutrition labels. The method is reagent-free, tolerates opaque samples, and operates under ambient conditions, making it suitable for on-site consumer use as well as regulatory inspection and quality-control applications. Full article

This literature review examines the relationship between the number and type of deficiencies identified during Port State Control (PSC) inspections and a ship’s overall risk. The main objective is to synthesise the current academic evidence, detailing the analytical methodologies employed and highlighting key research contributions. The selection of literature has focused on peer-reviewed articles and relevant doctoral theses addressing detention risk prediction, accident risk and ship risk profiling. The findings indicate a consistent correlation between PSC deficiencies and ship risk, although the nature and strength of this correlation may vary depending on the type of risk considered and the specific deficiencies. A methodological evolution is observed in the field, from descriptive statistical analyses and regressions towards more complex predictive models, such as Machine Learning (ML) and Bayesian Networks (BNs). This transition reflects a search for greater accuracy in risk assessment, going beyond simple numerical correlation to improve the selection of ships for inspection. Multivariate statistical techniques, on the other hand, focus on the identification of risk patterns and the evaluation of the PSC system. The conclusions underline the importance of deficiencies as indicators of risk, the need for differentiated inspection approaches and the persistent challenges related to data quality and model interpretability. Full article

Thermogravimetric analysis (TGA) has been extensively applied in polymeric characterization and quality inspection, facilitating in-depth investigations of the microstructural thermal response characteristics of polymers, including thermal stability, composition analysis, and thermal decomposition mechanisms. Here, the impacts of six factors on the TG thermal analysis curves obtained during operation are systematically examined while analyzing their causes and recommending solutions. Furthermore, the thermal degradation kinetics of an ionomer formed by neutralizing an ethylene–methacrylic acid copolymer with metal ions (SGP membrane) used in laminated tempered glass is analyzed using the Arrhenius equation, Ozawa–Flynn–Wall hypothesis and Kissinger method. Kinetic parameters at 5% degradation are fitted and used to predict the service lifetime of the SGP membrane. The results indicate that the SGP membrane sample exhibits activation energy Ea = 136.90 kJ/mol, reaction order n = 1.65 and pre-factor A = e^25.93. It can be seen that the service lifetime of the SGP membrane sample is 16 years at 80 °C and 1.65 years at 100 °C. Full article