Search Results (385)

This review analyzes the Electronic Digital Twin (EDT) tools characterizing the industrial transformation phase from Industry 4.0 to Industry 5.0. The goal is to provide innovative research EDT solutions to integrate in manufacturing production processes. Specifically, this research is focused on the possibility of combining the advanced technologies and electronics and mechatronics of industrial machines with Artificial Intelligence (AI) algorithms. Furthermore, this review provides important elements about possible future implementations of AI-EDTs and some circuital examples to support the understanding of the concept of circuit simulation in EDT models. EDTs are useful to comprehend the modeling concepts functional to the AI application using the output of the circuit simulations. The output of the circuit is used to train the AI model, thus strengthening the capability to classify and predict the real behavior of production machines with a good accuracy. This review discusses perspectives, limits, and advantages of EDTs and is useful to define new research patterns integrating structured EDTs in advanced industrial environments. The focus of this paper is the definition of possible perspectives of EDT implementations, including AI, in data-driven processes in specific strategic areas of industrial research by classifying the scientific topics in six main pillars. This paper is also suitable for the researcher to develop innovative topics for projects scaled into different work packages based on EDT facilities. Full article

Anthropomorphic CT phantoms are essential training tools for interventional radiology. Given the high technical demands and stringent safety requirements in this field, realistic CT phantoms are vital simulation tools that support effective hands-on training, procedural planning, and risk mitigation. However, commercially available phantom geometries are limited in their scope. This study investigates the use of polyvinyl alcohol (PVA) to fabricate customizable training phantoms. PVA, a non-toxic material, can be processed into PVA cryogels (PVA-C) with tissue-like mechanical properties. We modified PVA-C (10 wt.% PVA) by incorporating various additives to adjust X-ray attenuation and achieve Hounsfield units (HUs) similar to different soft tissues. HU values were measured at X-ray tube voltages of 70, 120, and 150 kV. The inclusion of barium sulfate (e.g., U = 120 kV; 0.1–2 wt.%: 33.29 ± 5.45–323.72 ± 12.64 HU) and iohexol (e.g., U = 120 kV; 0.1–2 wt.%: 26.05 ± 4.74–161.99 ± 5.69 HU) significantly increased HU values. Iohexol produced more homogeneous HU distributions than barium sulfate and cellulose derivatives, with the latter having air gaps and inconsistencies. The tested formulations encompassed a wide range of soft tissue densities, with HU values varying significantly across the energy range (p < 0.001). While cellulose derivatives showed variable HU modulation, their primary role appears to be in modifying phantom texture and morphology rather than precise attenuation control. In conclusion, PVA-C demonstrates strong potential for use in interventional radiology training phantoms. Further studies may enhance phantom realism by replicating tissue textures, for example, through the incorporation of cellulose-based substances. Full article

The You Only Look Once (YOLO) architecture is crucial for real-time object detection. However, deploying it in resource-constrained environments such as unmanned aerial vehicles (UAVs) requires efficient transfer learning. Although layer freezing is a common technique, the specific impact of various freezing configurations on contemporary YOLOv8 and YOLOv10 architectures remains unexplored, particularly with regard to the interplay between freezing depth, dataset characteristics, and training dynamics. This research addresses this gap by presenting a detailed analysis of layer-freezing strategies. We systematically investigate multiple freezing configurations across YOLOv8 and YOLOv10 variants using four challenging datasets that represent critical infrastructure monitoring. Our methodology integrates a gradient behavior analysis (L2 norm) and visual explanations (Grad-CAM) to provide deeper insights into training dynamics under different freezing strategies. Our results reveal that there is no universal optimal freezing strategy but, rather, one that depends on the properties of the data. For example, freezing the backbone is effective for preserving general-purpose features, while a shallower freeze is better suited to handling extreme class imbalance. These configurations reduce graphics processing unit (GPU) memory consumption by up to 28% compared to full fine-tuning and, in some cases, achieve mean average precision (mAP@50) scores that surpass those of full fine-tuning. Gradient analysis corroborates these findings, showing distinct convergence patterns for moderately frozen models. Ultimately, this work provides empirical findings and practical guidelines for selecting freezing strategies. It offers a practical, evidence-based approach to balanced transfer learning for object detection in scenarios with limited resources. Full article

Influence-based data selection methods, such as TracIn, aim to estimate the impact of individual training samples on model predictions and are increasingly used for dataset curation and reduction. This study investigates whether selecting the most positively influential training examples can be used to create compressed yet effective training datasets for transfer learning in plastic waste classification. Using a ResNet-18 model trained on a custom dataset of plastic waste images, TracIn was applied to compute influence scores across multiple training checkpoints. The top 50 influential samples per class were extracted and used to train a new model. Contrary to expectations, models trained on these highly influential subsets significantly underperformed compared to models trained on either the full dataset or an equally sized random sample. Further analysis revealed that many top-ranked influential images originated from different classes, indicating model biases and potential label confusion. These findings highlight the limitations of using influence scores for dataset compression. However, TracIn proved valuable for identifying problematic or ambiguous samples, class imbalance issues, and issues with fuzzy class boundaries. Based on the results, the utilized TracIn approach is recommended as a diagnostic instrument rather than for dataset curation. Full article

Automated quality control plays a critical role in modern industries, particularly in environments that handle large volumes of packaged products requiring fast, accurate, and consistent inspections. This work presents an anomaly detection system for candle jars commonly used in industrial and commercial applications, where obtaining labeled defective samples is challenging. Two anomaly detection strategies are explored: (1) a baseline model using convolutional neural networks (CNNs) as an end-to-end classifier and (2) a hybrid approach where features extracted by CNNs are fed into One-Class classification (OCC) algorithms, including One-Class SVM (OCSVM), One-Class Isolation Forest (OCIF), One-Class Local Outlier Factor (OCLOF), One-Class Elliptic Envelope (OCEE), One-Class Autoencoder (OCAutoencoder), and Support Vector Data Description (SVDD). Both strategies are trained primarily on non-defective samples, with only a limited number of anomalous examples used for evaluation. Experimental results show that both the pure CNN model and the hybrid methods achieve excellent classification performance. The end-to-end CNN reached 100% accuracy, precision, recall, F1-score, and AUC. The best-performing hybrid model CNN-based feature extraction followed by OCIF also achieved 100% across all evaluation metrics, confirming the effectiveness and robustness of the proposed approach. Other OCC algorithms consistently delivered strong results, with all metrics above 95%, indicating solid generalization from predominantly normal data. This approach demonstrates strong potential for quality inspection tasks in scenarios with scarce defective data. Its ability to generalize effectively from mostly normal samples makes it a practical and valuable solution for real-world industrial inspection systems. Future work will focus on optimizing real-time inference and exploring advanced feature extraction techniques to further enhance detection performance. Full article

Environmental noise is inevitable during seismic data acquisition, with major sources including heavy machinery, rivers, wind, and other environmental factors. During field data acquisition, it is important to assess the impact of environmental noise and evaluate data quality. In subsequent seismic data processing, these noise components also need to be eliminated. Accurate identification of noise traces facilitates rapid quality control (QC) during fieldwork and provides a reliable basis for targeted noise attenuation. Conventional environmental noise identification primarily relies on amplitude differences. However, in seismic data, high-amplitude signals are not necessarily caused by environmental noise. For example, surface waves or traces near the shot point may also exhibit high amplitudes. Therefore, relying solely on amplitude-based criteria has certain limitations. To improve noise identification accuracy, we use the Mel-spectrogram to extract features from seismic data and construct the dataset. Compared to raw time-series signals, the Mel-spectrogram more clearly reveals energy variations and frequency differences, helping to identify noise traces more accurately. We then employ a Vision Transformer (ViT) network to train a model for identifying noise in seismic data. Tests on synthetic and field data show that the proposed method performs well in identifying noise. Moreover, a denoising case based on synthetic data further confirms its general applicability, making it a promising tool in seismic data QC and processing workflows. Full article

In order to increase the effectiveness of model training, data reduction is essential to data-centric Artificial Intelligence (AI). It achieves this by locating the most instructive examples in massive datasets. To increase data quality and training efficiency, the main difficulty is choosing the best examples rather than the complete datasets. In this paper, we propose an effective data reduction strategy based on Pointwise $\mathcal{V}$-Information (PVI). To enable a static method, we first use PVI to quantify instance difficulty and remove instances with low difficulty. Experiments show that classifier performance is maintained with only a 0.0001% to 0.76% decline in accuracy when 10–30% of the data is removed. Second, we train the classifiers using a progressive learning strategy on examples sorted by increasing PVI, accelerating convergence and achieving a 0.8% accuracy gain over conventional training. Our findings imply that training a classifier on the chosen optimal subset may improve model performance and increase training efficiency when combined with an efficient data reduction strategy. Furthermore, we have adapted the PVI framework, which was previously limited to English datasets, to a variety of Chinese Natural Language Processing (NLP) tasks and base models, yielding insightful results for faster training and cross-lingual data reduction. Full article

Medical imaging is vital for diagnosing and treating colorectal cancer (CRC), a leading cause of mortality. Classifying colorectal polyps and CRC precursors remains challenging due to operator variability and expertise dependence. Deep learning (DL) models show promise in polyp classification but face adoption barriers due to their ‘black box’ nature, limiting interpretability. This study presents an example-based explainable artificial intehlligence (XAI) approach using Pix2Pix to generate synthetic polyp images with controlled size variations and LIME to explain classifier predictions visually. EfficientNet and Vision Transformer (ViT) were trained on datasets of real and synthetic images, achieving strong baseline accuracies of 94% and 96%, respectively. Image quality was assessed using PSNR (18.04), SSIM (0.64), and FID (123.32), while classifier robustness was evaluated across polyp sizes. Results show that Pix2Pix effectively controls image attributes like polyp size despite limitations in visual fidelity. LIME integration revealed classifier vulnerabilities, underscoring the value of complementary XAI techniques. This enhances DL model interpretability and deepens understanding of their behaviour. The findings contribute to developing explainable AI tools for polyp classification and CRC diagnosis. Future work will improve synthetic image quality and refine XAI methodologies for broader clinical use. Full article

Vision-based end-to-end navigation systems have shown impressive capabilities, especially when combined with Imitation Learning (IL) and advanced Deep Learning architectures, such as Transformers. One such example is CIL++, a Transformer-based architecture that learns to map navigation states to vehicle controls based on expert demonstrations only. Nevertheless, reliance on experts’ datasets limits generalization and can lead to failures in unknown circumstances. Deep Reinforcement Learning (DRL) can address this issue by fine-tuning the pretrained policy, using a reward function that aims to improve its weaknesses through interaction with the environment. However, fine-tuning with DRL can lead to the Catastrophic Forgetting (CF) problem, where a policy forgets the expert behaviors learned from the demonstrations as it learns to optimize the new reward function. In this paper, we present CILRLv3, a DRL-based training method that is immune to CF, enabling pretrained navigation agents to improve their driving skills across new scenarios. Full article

Unmanned aerial vehicles (UAVs) are commonly used in various fields and industries, but their limited battery life has become a key constraint for their development. Wireless Power Transmission (WPT) technology, with its convenience, durability, intelligence, and unmanned features, significantly enhances UAVs’ battery life and operational range. However, the variety of UAV models and different sizes pose challenges for designing couplers in the WPT system. This paper presents a design method for an array-type coupler in a UAV WPT system that uses a deep neural network. By establishing an electromagnetic 3D structure of the array-type coupler using electromagnetic simulation software, the dimensions of the transmitting and receiving coils are modified to assess how changes in the aperture of the transmitting coil and the length of the receiving coil affect the mutual inductance of the coupler. Furthermore, deep learning methods are utilized to train a high-precision model using the calculated data as the training and testing sets. Finally, taking the FAIRSER-X model UAV as an example, the transmitting and receiving coils are wound, and the feasibility and accuracy of the proposed method are verified through an LCR meter, which notably enhances the design efficiency of UAV WPT systems. Full article

Text-to-SQL for spatial databases enables the translation of natural language questions into corresponding SQL queries, allowing non-experts to easily access spatial data, which has gained increasing attention from researchers. Previous research has primarily focused on rule-based methods. However, these methods have limitations when dealing with complicated or unknown natural language questions. While advanced machine learning models can be trained, they typically require large labeled training datasets, which are severely lacking for spatial databases. Recently, Generative Pre-Trained Transformer (GPT) models have emerged as a promising paradigm for Text-to-SQL tasks in relational databases, driven by carefully designed prompts. In response to the severe lack of datasets for spatial databases, we have created a publicly available dataset that supports both English and Chinese. Furthermore, we propose a GPT-based method to construct prompts for spatial databases, which incorporates geographic and spatial database knowledge into the prompts and requires only a small number of training samples, such as 1, 3, or 5 examples. Extensive experiments demonstrate that incorporating geographic and spatial database knowledge into prompts improves the accuracy of Text-to-SQL tasks for spatial databases. Our proposed method can help non-experts access spatial databases more easily and conveniently. Full article

In novels, readers encounter a variety of characters with distinct personalities, and their satisfaction tends to increase when each character’s utterances consistently reflect their unique traits. Recent advances in large language model (LLM) technology have made it possible to perform complex tasks such as generating long-form narratives and adapting writing styles. However, research on generating character utterances that reflect individual personalities remains limited. In this paper, we identify a key challenge in this task, namely the unconscious influence of the author’s writing style, and propose a novel clustering-based method to mitigate this problem by tuning large language models. We manually annotated Big Five personality trait scores for characters appearing in selected novels and designed prompts to generate examples for instruction-tuning. Experimental results demonstrate that language models trained using our proposed method produce utterances that more consistently reflect character personalities compared to untuned models. Full article

Automated waste classification is an essential step toward efficient recycling and waste management. Traditional deep learning models, such as convolutional neural networks, rely on extensive labeled datasets to achieve high accuracy. However, the annotation process is labor-intensive and time-consuming, limiting the scalability of these approaches in real-world applications. Zero-shot learning is a machine learning paradigm that enables a model to recognize and classify objects it has never seen during training by leveraging semantic relationships and external knowledge sources. In this study, we investigate the potential of zero-shot learning for waste classification using two vision-language models: OWL-ViT and OpenCLIP. These models can classify waste without direct exposure to labeled examples by leveraging textual prompts. We apply this approach to the TrashNet dataset, which consists of images of municipal solid waste organized into six distinct categories: cardboard, glass, metal, paper, plastic, and trash. Our experimental results yield an average classification accuracy of 76.30% with Open Clip ViT-L/14-336 model, demonstrating the feasibility of zero-shot learning for waste classification while highlighting challenges in prompt sensitivity and class imbalance. Despite lower accuracy than CNN- and ViT-based classification models, zero-shot learning offers scalability and adaptability by enabling the classification of novel waste categories without retraining. This study underscores the potential of zero-shot learning in automated recycling systems, paving the way for more efficient, scalable, and annotation-free waste classification methodologies. Full article

In this work, a model-based approach to inferentially obtaining information about the 3D fractal dimension of agglomerates produced in spray fluidized beds is presented. The method utilizes high-detail but scarce offline information from X-ray microcomputed tomography for establishing and training an inferential relationship with online information that is easy and fast to obtain. The online measurement information is the geometric roundness of the single agglomerate. To investigate the interpolation capability of the inferential approach, three different strategies are evaluated: correlation with individual process conditions; correlation with parameters adjusted to process parameters; and correlation with respect to a range of process conditions. It is shown that the approach incorporating process conditions provides sufficient accuracy over a wide range of conditions. The inferential evaluation of single agglomerate 3D fractal dimension is achieved in 5 ms on average. This enables the measurement of the distribution of 3D fractal dimension in an online setting for product quality monitoring and control. Several examples illustrate the capabilities of the approach, as well as current limitations. Full article

In recent years, extreme weather events have occurred more frequently. The resulting equipment failure, renewable energy extreme output, and other extreme operation scenarios affect the smooth operation of power grids. The occurrence probability of extreme operation scenarios is small, and the occurrence frequency in historical operation data is low, which affects the modeling accuracy for scenario generation. Meanwhile, extreme operation scenarios in the form of discrete temporal data lack corresponding modeling methods. Therefore, this paper proposes a definition and generation framework for extreme power grid operation scenarios triggered by extreme weather events. Extreme operation scenario expansion is realized based on the sequential Monte Carlo sampling method and the distribution shifting algorithm. To generate equipment failure scenarios in discrete temporal data form and extreme output scenarios in continuous temporal data form for renewable energy, a Gumbel-Softmax variational autoencoder and an extreme conditional generative adversarial network are respectively proposed. Numerical examples show that the proposed models can effectively overcome limitations related to insufficient historical extreme data and discrete extreme scenario training. Additionally, they can generate improved-quality equipment failure scenarios and renewable energy extreme output scenarios and provide scenario support for power grid planning and operation. Full article