Search Results (3,796)

The “Four Illegal Activities”, which involve occupation, extraction, and construction along shorelines, have become significant challenges in river and lake management in recent years. Due to the diverse and scattered nature of monitoring targets, coupled with the large volumes of data involved, traditional manual inspection methods are no longer sufficient to meet regulatory demands. Late fusion change detection methods in deep learning are particularly effective for monitoring river and lake occupation due to their straightforward principles and processes. However, research on this topic remains limited. To fill this gap, we selected eight popular deep learning networks—VGGNet, ResNet, MobileNet, EfficientNet, DenseNet, Inception-ResNet, SeNet, and DPN—and used the Jinshui River Basin in Wuhan as a case study to explore the application of Siamese networks to monitor river and lake occupation. Our results indicate that the Siamese network based on EfficientNet outperforms all other models. It can be reasonably concluded that the combination of the SE module and residual connections provides an effective approach for improving the performance of deep learning models in monitoring river and lake occupation. Our findings contribute to improving the efficiency of monitoring river and lake occupation, thereby enhancing the effectiveness of water resource and ecological environment protection. In addition, they aid in the development and implementation of efficient strategies for promoting sustainable development. Full article

Background: Breast cancer screening through mammography interpretation is crucial for early detection and improved patient outcomes. However, the manual classification of mammograms using the BIRADS (Breast Imaging-Reporting and Data System) remains challenging due to subtle imaging features, inter-reader variability, and increasing radiologist workload. Traditional computer-aided detection systems often struggle with complex feature extraction and contextual understanding of mammographic abnormalities. To address these limitations, this study proposes MammoViT, a novel hybrid deep learning framework that leverages both ResNet50’s hierarchical feature extraction capabilities and Vision Transformer’s ability to capture long-range dependencies in images. Methods: We implemented a multi-stage approach utilizing a pre-trained ResNet50 model for initial feature extraction from mammogram images. To address the significant class imbalance in our four-class BIRADS dataset, we applied SMOTE (Synthetic Minority Over-sampling Technique) to generate synthetic samples for minority classes. The extracted feature arrays were transformed into non-overlapping patches with positional encodings for Vision Transformer processing. The Vision Transformer employs multi-head self-attention mechanisms to capture both local and global relationships between image patches, with each attention head learning different aspects of spatial dependencies. The model was optimized using Keras Tuner and trained using 5-fold cross-validation with early stopping to prevent overfitting. Results: MammoViT achieved 97.4% accuracy in classifying mammogram images across different BIRADS categories. The model’s effectiveness was validated through comprehensive evaluation metrics, including a classification report, confusion matrix, probability distribution, and comparison with existing studies. Conclusions: MammoViT effectively combines ResNet50 and Vision Transformer architectures while addressing the challenge of imbalanced medical imaging datasets. The high accuracy and robust performance demonstrate its potential as a reliable tool for supporting clinical decision-making in breast cancer screening. Full article

The rise of deepfakes—synthetic media generated using artificial intelligence—threatens digital content authenticity, facilitating misinformation and manipulation. However, deepfakes can also depict real or entirely fictitious individuals, leveraging state-of-the-art techniques such as generative adversarial networks (GANs) and emerging diffusion-based models. Existing detection methods face challenges with generalization across datasets and vulnerability to adversarial attacks. This study focuses on subsets of frames extracted from the DeepFake Detection Challenge (DFDC) and FaceForensics++ videos to evaluate three convolutional neural network architectures—XCeption, ResNet, and VGG16—for deepfake detection. Performance metrics include accuracy, precision, F1-score, AUC-ROC, and Matthews Correlation Coefficient (MCC), combined with an assessment of resilience to adversarial perturbations via the Fast Gradient Sign Method (FGSM). Among the tested models, XCeption achieves the highest accuracy (89.2% on DFDC), strong generalization, and real-time suitability, while VGG16 excels in precision and ResNet provides faster inference. However, all models exhibit reduced performance under adversarial conditions, underscoring the need for enhanced resilience. These findings indicate that robust detection systems must consider advanced generative approaches, adversarial defenses, and cross-dataset adaptation to effectively counter evolving deepfake threats. Full article

Human activity recognition (HAR) using radar technology is becoming increasingly valuable for applications in areas such as smart security systems, healthcare monitoring, and interactive computing. This study investigates the integration of convolutional neural networks (CNNs) with conventional radar signal processing methods to improve the accuracy and efficiency of HAR. Three distinct, two-dimensional radar processing techniques, specifically range-fast Fourier transform (FFT)-based time-range maps, time-Doppler-based short-time Fourier transform (STFT) maps, and smoothed pseudo-Wigner–Ville distribution (SPWVD) maps, are evaluated in combination with four state-of-the-art CNN architectures: VGG-16, VGG-19, ResNet-50, and MobileNetV2. This study positions radar-generated maps as a form of visual data, bridging radar signal processing and image representation domains while ensuring privacy in sensitive applications. In total, twelve CNN and preprocessing configurations are analyzed, focusing on the trade-offs between preprocessing complexity and recognition accuracy, all of which are essential for real-time applications. Among these results, MobileNetV2, combined with STFT preprocessing, showed an ideal balance, achieving high computational efficiency and an accuracy rate of 96.30%, with a spectrogram generation time of 220 ms and an inference time of 2.57 ms per sample. The comprehensive evaluation underscores the importance of interpretable visual features for resource-constrained environments, expanding the applicability of radar-based HAR systems to domains such as augmented reality, autonomous systems, and edge computing. Full article

In nuclide identification, traditional methods based on nuclide library comparisons rely on the identification of characteristic peaks, often overlooking the full spectrum information, which leads to cumbersome operations and low efficiency. In recent years, machine learning and deep learning techniques have been introduced into the field of nuclide recognition to improve identification efficiency; however, most existing methods fail to effectively extract deep features from the data. To address this issue, this paper proposes a method that integrates the Gram Angular Summation Field (GASF) algorithm with a Deep Residual Network (ResNet) for processing nuclide energy spectrum data. First, the GASF algorithm is used to transform one-dimensional spectral data into two-dimensional images, thereby fully extracting spatial features from the data. Then, these two-dimensional images are input into the ResNet model, where features are automatically extracted through multiple convolutional layers. Finally, the Softmax layer is used for nuclide classification. Experimental results demonstrate that the proposed method can effectively improve both the accuracy and efficiency of nuclide identification; the recognition accuracy on the simulated data reaches 99.5%, and, when tested with actual measurement data containing unknown radionuclides, the model still achieves a high accuracy of 92.6%. This study shows that the combination of deep learning and signal processing techniques can significantly improve the accuracy and application scope of nuclide identification, offering substantial practical value. Full article

This study explores the use of machine learning for the automated classification of the ten most abundant groups of marine organisms (in the size range of 5–12 cm) plus marine snow found in the ecosystem of the U.S. east coast. Images used in this process were collected using a shadowgraph imaging system on a towed, undulating platform capable of collecting continuous imagery over large spatiotemporal scales. As a large quantity (29,818,917) of images was collected, the task of locating and identifying all imaged organisms could not be efficiently achieved by human analysis alone. Several tows of data were collected off the coast of Delaware Bay. The resulting images were then cleaned, segmented into regions of interest (ROIs), and fed through three convolutional neural networks (CNNs): VGG-16, ResNet-50, and a custom model created to find more high-level features in this dataset. These three models were used in a Random Forest Classifier-based ensemble approach to reach the best identification fidelity. The networks were trained on a training set of 187,000 ROIs augmented with random rotations and pixel intensity thresholding to increase data variability and evaluated against two datasets. While the performance of each individual model is examined, the best approach is to use the ensemble, which performed with an F1-score of 98% and an area under the curve (AUC) of 99% on both test datasets while its accuracy, precision, and recall fluctuated between 97% and 98%. Full article

This research study utilized artificial intelligence (AI) to detect natural disasters from aerial images. Flooding and desertification were two natural disasters taken into consideration. The Climate Change Dataset was created by compiling various open-access data sources. This dataset contains 6334 aerial images from UAV (unmanned aerial vehicles) images and satellite images. The Climate Change Dataset was then used to train Deep Learning (DL) models to identify natural disasters. Four different Machine Learning (ML) models were used: convolutional neural network (CNN), DenseNet201, VGG16, and ResNet50. These ML models were trained on our Climate Change Dataset so that their performance could be compared. DenseNet201 was chosen for optimization. All four ML models performed well. DenseNet201 and ResNet50 achieved the highest testing accuracies of 99.37% and 99.21%, respectively. This research project demonstrates the potential of AI to address environmental challenges, such as climate change-related natural disasters. This study’s approach is novel by creating a new dataset, optimizing an ML model, cross-validating, and presenting desertification as one of our natural disasters for DL detection. Three categories were used (Flooded, Desert, Neither). Our study relates to AI for Climate Change and Environmental Sustainability. Drone emergency response would be a practical application for our research project. Full article

Background/Objectives: Sports-related bone fractures are a common challenge in sports medicine, requiring accurate and timely diagnosis to prevent long-term complications and enable effective treatment. Conventional diagnostic methods often rely on manual interpretation, which is prone to errors and inefficiencies, particularly for subtle and localized fractures. This study aims to develop a lightweight and efficient deep learning-based framework to improve the accuracy and computational efficiency of fracture detection, tailored to the needs of sports medicine. Methods: We proposed a novel fracture detection framework based on the DenseNet121 architecture, incorporating modifications to the initial convolutional block and final layers for optimized feature extraction. Additionally, a Canny edge detector was integrated to enhance the model ability to detect localized structural discontinuities. A custom-curated dataset of radiographic images focused on common sports-related fractures was used, with preprocessing techniques such as contrast enhancement, normalization, and data augmentation applied to ensure robust model performance. The model was evaluated against state-of-the-art methods using metrics such as accuracy, recall, precision, and computational complexity. Results: The proposed model achieved a state-of-the-art accuracy of 90.3%, surpassing benchmarks like ResNet-50, VGG-16, and EfficientNet-B0. It demonstrated superior sensitivity (recall: 0.89) and specificity (precision: 0.875) while maintaining the lowest computational complexity (FLOPs: 0.54 G, Params: 14.78 M). These results highlight its suitability for real-time clinical deployment. Conclusions: The proposed lightweight framework offers a scalable, accurate, and efficient solution for fracture detection, addressing critical challenges in sports medicine. By enabling rapid and reliable diagnostics, it has the potential to improve clinical workflows and outcomes for athletes. Future work will focus on expanding the model applications to other imaging modalities and fracture types. Full article

The accurate identification of natural disasters is crucial in ensuring effective post-disaster relief efforts. However, the existing models for disaster classification often incur high costs. To address this, we propose leveraging the most advanced pre-trained large language models, which offer superior generative and multimodal understanding capabilities. Using a question-answering approach, we extract textual descriptions and category prediction probabilities for disaster scenarios, which are then used as input to our proposed Mamba Multimodal Disaster Recognition Network (Mamba-MDRNet). This model integrates a large pre-trained model with the Mamba mechanism, enabling the selection of the most reliable modality information as a robust basis for scene classification. Extensive experiments demonstrate consistent performance improvements across various visual models with heterogeneous architectures. Notably, integrating EfficientNet within Mamba-MDRNet yielded 97.82% accuracy for natural scene classification, surpassing the performance of the CNN (91.75%), ViT (94.50%), and ResNet18 (97.25%). These results highlight the potential of multimodal models combining large models and the Mamba mechanism for disaster type prediction. Full article

Lithology classification stands as a pivotal research domain within geological Remote Sensing (RS). In recent years, extracting lithology information from multi-source RS data has become an inevitable trend. Various classification image primitives yield distinct outcomes in lithology classification. The current research on lithology classification utilizing RS data has predominantly concentrated on pixel-level classification, which suffers from a long classification time and high sensitivity to noise. In order to explore the application potential of superpixel segmentation in lithology classification, this study proposed the Multi-scale superpixel Segmentation Integrating Multi-source RS data (MSIMRS), and conducted a lithology classification study in Duolun County, Inner Mongolia Autonomous Region, China combining MSIMRS and the Support Vector Machine (MSIMRS-SVM). In addition, pixel-level K-Nearest Neighbor (KNN), Random Forest (RF) and SVM classification algorithms, as well as deep-learning models including Resnet50 (Res50), Efficientnet_B8 (Effi_B8), and Vision Transformer (ViT) were chosen for a comparative analysis. Among these methods, our proposed MSIMRS-SVM achieved the highest accuracy in lithology classification in a typical semi-arid area, Duolun County, with an overall accuracy and Kappa coefficient of 92.9% and 0.92. Moreover, the findings indicate that incorporating superpixel segmentation into lithology classification resulted in notably fewer fragmented patches and significantly improved the visualization effect. The results showcase the application potential of superpixel primitives in lithology information extraction within semi-arid areas. Full article

Face verification and identification traditionally follow a symmetric matching approach, where the same model (e.g., ResNet-50 vs. ResNet-50) generates embeddings for both gallery and query images, ensuring compatibility. However, real-world scenarios often demand asymmetric matching, especially when query devices have limited computational resources or employ heterogeneous models (e.g., ResNet-50 vs. SwinTransformer). This asymmetry can degrade face recognition performance due to incompatibility between embeddings from different models. To tackle this asymmetric face recognition problem, we introduce the Learnable Anchor Embedding (LAE) model, which features two key innovations: the Shared Learnable Anchor and a Light Cross-Attention Mechanism. The Shared Learnable Anchor is a dynamic attractor, aligning heterogeneous gallery and query embeddings within a unified embedding space. The Light Cross-Attention Mechanism complements this alignment process by reweighting embeddings relative to the anchor, efficiently refining their alignment within the unified space. Extensive evaluations of several facial benchmark datasets demonstrate LAE’s superior performance, particularly in asymmetric settings. Its robustness and scalability make it an effective solution for real-world applications such as edge-device authentication, cross-platform verification, and environments with resource constraints. Full article

Cardiovascular disease (CVD) poses a significant challenge to global health, with cardiac arrhythmia representing one of its most prevalent manifestations. The timely and precise classification of arrhythmias is critical for the effective management of CVD. This study introduces an innovative approach to enhancing arrhythmia classification accuracy through advanced Electrocardiogram (ECG) signal processing. We propose a dual-channel feature fusion strategy designed to enhance the precision and objectivity of ECG analysis. Initially, we apply an Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN) and enhanced wavelet thresholding for robust noise reduction. Subsequently, in the primary channel, region of interest features are emphasized using a ResNet-ICBAM network model for feature extraction. In parallel, the secondary channel transforms 1D ECG signals into Gram angular difference field (GADF), Markov transition field (MTF), and recurrence plot (RP) representations, which are then subjected to two-dimensional convolutional neural network (2D-CNN) feature extraction. Post-extraction, the features from both channels are fused and classified. When evaluated on the MIT-BIH database, our method achieves a classification accuracy of 97.80%. Compared to other methods, our approach of two-channel feature fusion has a significant improvement in overall performance by adding a 2D convolutional network. This methodology represents a substantial advancement in ECG signal processing, offering significant potential for clinical applications and improving patient care efficiency and accuracy. Full article

The objective of this study is to accurately, expeditiously, and efficiently identify the wear state of milling cutters. To this end, a state identification method is proposed that combines continuous wavelet transform and an improved MobileViT lightweight network. The methodology involves the transformation of the cutting force signal during the milling cutter cutting process into a time–frequency image by continuous wavelet transform. This is followed by the introduction of a Contextual Transformer module after layer 1 and the embedding of a Global Attention Mechanism module after layer 2 of the MobileViT network structure. These modifications are intended to enhance visual representation capability, reduce information loss, and improve the interaction between global features. The result is an improvement in the overall performance of the model. The improved MobileViT network model was shown to enhance accuracy, precision, recall, and F1 score by 1.58%, 1.23%, 1.92%, and 1.57%, respectively, in comparison with the original MobileViT. The experimental results demonstrate that the proposed model in this study exhibits a substantial advantage in terms of memory occupation and prediction accuracy in comparison to models such as VGG16, ResNet18, and Pool Former. This study proposes an efficient identification method for milling cutter wear state identification, which can identify the tool wear state in near real-time. The proposed method has potential applications in the field of industrial production. Full article

The natural occurrence of harmful algal blooms (HABs) adversely affects the quality of clean and fresh water. They pose increased risks to human health, aquatic ecosystems, and water bodies. Continuous monitoring and appropriate measures must be taken to combat HABs. Deep learning models that utilize computer vision play a vital role in identifying and classifying harmful algal blooms in aquatic environments and water storage facilities. Inspecting algal blooms using conventional methods, such as algae detection under microscopes, is difficult, expensive, and time-consuming. Deep learning algorithms have shown a notable and remarkable performance in the image classification domain and its applications, including microscopic algae species classification and detection. In this study, we propose a deep learning-based approach for classifying microscopic images of algae using computer vision. This approach employs a convolutional neural network (CNN) model integrated with two additional blocks—squeeze and dense blocks—to determine the presence of algae, followed by adding Grad-CAM to the proposed model to ensure interpretability and transparency. We performed several experiments on our custom dataset of microscopic algae images. Data augmentation techniques were employed to increase the number of images in the dataset, whereas pre-processing techniques were implemented to elevate the overall data quality. Our proposed model was trained on 3200 images consisting of four classes. We also compared our proposed model with the other transfer learning models, i.e., ResNet50 and Vgg16. Our proposed model outperformed the other two deep learning models. The proposed model demonstrated 96.7% accuracy, while Resnet50, EfficientNetB0, and VGG16 showed accuracy of 85.0%, 92.96%, and 93.5%, respectively. The results of this research demonstrate the potential of deep learning-based approaches for algae classification. This deep learning-based algorithm can be deployed in real-time applications to classify and identify algae to ensure the quality of water reservoirs. Computer-assisted solutions are advantageous for tracking freshwater algal blooms. Using deep learning-based models to identify and classify algae species from microscopic images is a novel application in the AI community. Full article

In the design of fluid systems, rapid iteration and simulation verification are essential, and reduced-order modeling techniques can significantly improve computational efficiency and accuracy. However, traditional Physics-Informed Neural Networks (PINNs) often face challenges such as vanishing or exploding gradients when learning flow field characteristics, limiting their ability to capture complex fluid dynamics. This study presents an enhanced reduced-order model (ROM): Physics-Informed Neural Networks based on Residual Networks (Res-PINNs). By integrating a Residual Network (ResNet) module into the PINN architecture, the proposed model improves training stability while preserving physical constraints. Additionally, the model’s ability to capture and learn flow field states is further enhanced by the design of a symmetric parallel neural network structure. To evaluate the effectiveness of the Res-PINNs model, two classic fluid dynamics problems—flow around a cylinder and Vortex-Induced Vibration (VIV)—were selected for comparative testing. The results demonstrate that the Res-PINNs model not only reconstructs flow field states with higher accuracy but also effectively addresses limitations of traditional PINN methods, such as vanishing gradients, exploding gradients, and insufficient learning capacity. Compared to existing approaches, the proposed Res-PINNs provide a more stable and efficient solution for deep learning-based reduced-order modeling in fluid system design. Full article