Improved YOLOv11 with hybrid encoder boosts rotating target detection in remote sensing images
A new SR-YOLO-DETR framework combining an improved YOLOv11 backbone with an RT-DETR hybrid encoder raises rotating target detection accuracy on ...
Tag: deep learning
LDNP-enhanced CNN improves COVID-19 lung CT segmentation in small dataset
A lightweight CNN combining directional texture information with original CT images improved COVID-19 infection segmentation in a computational study of ...
Hybrid model classifies 32 insect species in controlled audio dataset
A hybrid deep-learning model classified 32 insect species from 4,800 recordings with 99.90% accuracy on the InsectSet32 dataset, but independent ...
Deep learning outperforms classical methods for early weed detection
A comparative study of sugar beet and maize seedling imagery found that a custom CNN outperformed classical classifiers, while U-Net ...
Transformer Models Show Promise for Lung Cancer Imaging
A review of 2019–2025 research finds that transformer models show promise for lung cancer detection and classification, but variable validation ...
Conditional diffusion model improves bridge monitoring anomaly detection
A conditional diffusion model reached an F1-score of 0.975 when detecting abnormal data in a 90-day, 18-channel bridge monitoring dataset ...
Hybrid AI model reports strong macro-financial policy classification performance
A hybrid model combining bidirectional deep learning, attention and fuzzy reasoning reported 98.8% accuracy on a 2,500-observation macro-financial policy dataset ...
SERS Fluctuations Become Concentration Fingerprints for Ultrasensitive Analysis
A statistical SERS strategy turned intensity-distribution shapes into concentration fingerprints, enabling a neural network to classify five rhodamine 6G levels ...
Self-attention improves EEG workload classification in one multitasking study
A four-head self-attention module raised cross-validated EEG workload classification accuracy from 80.73% to 91.33% during SIMKAP multitasking in one study, ...
Stacking deep learning reaches 97.2% accuracy for olive disease images
A stacking ensemble combining EfficientNet, ResNet and MobileNet classified healthy and diseased olive leaf images with 97.21% accuracy on a ...
DeiT framework pairs plant disease detection with interpretable severity estimates
A hybrid DeiT system reached 99.13% accuracy on a 38-class plant disease benchmark while combining local and global AI explanations ...
AI predicts mycorrhizal colonization, but field validation lags
A review of 34 studies finds that artificial intelligence performs well at analyzing mycorrhizal root images, but limited external and ...
Multimodal user behavior analysis supports closed loop optimization of smart home interaction flows
A controlled study of 40 smart home app users found that multimodal behavioral analysis guided prototype redesigns associated with lower ...
Multi-scale AI model improves rice blast severity assessment
A CNN that fuses shallow and deep lesion features classified rice blast severity with about 93% accuracy and estimated the ...
MCGE-Net joins image, text and labels to enhance imaging
A controlled study reports that MCGE-Net enhanced images using reference images, text and pixel labels, outperforming comparison methods across medical, ...
Specialized deep learning improves UAV-based solar fault detection
A UAV-assisted system using specialized YOLOv8s detectors achieved 85% accuracy for three photovoltaic fault categories in a held-out test set ...
XPAR-Net: cross-paradigm alignment and reconstruction network for high-fidelity lens-free on-chip microscopy
XPAR-Net improved average reconstruction quality over HoloFormer on an 11-category in-distribution test set while using 8.8% of its trainable parameters ...
Machine Learning Pushes Flexible Pressure Sensors Toward Intelligent Perception
A review in the Journal of Materials Science maps how microstructured flexible pressure sensors made from materials such as PDMS ...
AI Video Analysis Flags Newborns at Risk of Impaired Breast Crawl
A proof-of-concept study reports that a two-stage deep learning pipeline analyzing video of breast crawl sessions can grade newborn crawl ...
From Pixels to Compressed Streams: A Review Maps the Evolving Field of Anomaly Detection
A new review in Knowledge and Information Systems synthesizes a decade of anomaly detection research across images, videos, and compressed ...
POPULAR NEWS
-
Multitask Cambrian Model Improves Protein Melting Temperature Prediction
-
Vasopressors Exert Divergent Immunomodulatory Effects Across a Spectrum in Septic Shock
-
Study Identifies Key Drivers Shaping Salmonid Gill Microbiomes in Aquaculture
-
Chinese Cancer Association issues unified guideline for tumor-related lung injury care