Defensa y seguridad

From Droplet to Diagnosis: Spatio‐Temporal Pattern Recognition in Drying Biofluids

16/11/2025 - This article integrates machine learning (ML) with the spatio‐temporal evolution of biofluid droplets to reveal how drying and self‐assembly encode distinctive compositional fingerprints. By...

Gaussian Mixture Model‐Based Data Association Incorporating a Deep Learning Network for Multivehicle Tracking and Detection in Autonomous Driving Systems

17/09/2025 - This study introduces a real‐time light detection and ranging‐camera fusion framework for vehicle detection and tracking. Using a Gaussian mixture model‐based association and improved affinity...

Multitarget Recognition of Flower Images Based on Lightweight Deep Neural Network and Transfer Learning

07/08/2025 - This article proposes a lightweight YOLOv4‐based detection model using MobileNetV3 or CSPDarknet53_tiny, achieving 30+ FPS and higher mAP. It also presents a ShuffleNet‐based classification model...

Visión artificial en tiempo real para dispositivos portátiles inteligentes

03/08/2025 - This work proposes an attention‐enhanced transformer for image captioning, improving multiregion feature representation and future‐context decoding, achieving 2.1 CIDEr gain and real‐time...

Optimized DeepLabV3+ for Clinical Data Analysis through Advanced Particle Swarm Optimization‐Based Channel Selection

13/07/2025 - A novel medical image segmentation framework that integrates an enhanced Particle Swarm Optimization (EE‐PSO) into DeepLabV3+ to optimize feature selection. By dynamically identifying key channels...

Hi‐LabSpermTracking: A Novel and High‐Quality Sperm Tracking Dataset with an Advanced Ensemble Detection and Tracking Approach for Real‐World Clinical Scenarios

10/07/2025 - This study presents Hi‐LabSpermTracking, a long‐duration, expert‐annotated sperm motility dataset for detection and tracking. It evaluates You Only Look Once version 8, Real‐Time Detection...

Bioinspired Fully On‐Chip Learning Implemented on Memristive Neural Networks

01/07/2025 - This work proposes a memristive neural network based on van der Waals ferroelectric memristors and contrastive Hebbian learning, enabling fully on‐chip learning. The system achieves over 98%...

Highly Efficient Classification of Time‐Series Based on Resistive Switching Cluster‐Assembled Materials

25/06/2025 - Cluster‐assembled nanocomposite devices are employed for the classification of neuronal traces, without any preprocessing of the input time‐series or prior training of the nonlinear devices. The...

Adapting Image‐Based Models for 1D Data via Spider Plot Transformation and Transfer Learning

25/06/2025 - A novel method enables the use of pretrained image‐based neural networks for complex 1D data, including Raman and mid‐infrared spectra, electrocardiograms, and mass spectrometry. 2D spider plots...

Geometry‐Guided Transformer for Monocular 3D Object Detection

11/05/2025 - This work proposes a geometry‐guided transformer‐based framework for monocular 3D object detection. By incorporating spatial features and geometry guidance through the advanced transformer...

Multi‐Disease Detection in Retinal Imaging Using VNet with Image Processing Methods for Data Generation

06/05/2025 - This study introduces a data augmentation method that expands an ophthalmology dataset by 12x, enhancing robustness and reducing overfitting. A novel VNet architecture improves accuracy by 10% over...

Assessing Large Multimodal Models for One‐Shot Learning and Interpretability in Biomedical Image Classification

06/04/2025 - Image classification plays a pivotal role in biomedical image analysis. Herein, it is shown that large multimodal models, such as GPT‐4, achieve superior performance in one‐shot learning,...

Flight of the Future: An Experimental Analysis of Event‐Based Vision for Online Perception Onboard Flapping‐Wing Robots

08/03/2025 - A thorough analysis of event‐based vision for flapping‐wing robots is presented, with emphasis on hardware integration, suitability under challenging flight conditions, and real‐time...

Force Measurement Technology of Vision‐Based Tactile Sensor

20/01/2025 - Force sensing is an important function of marker‐type vision‐based tactile sensors. This review focuses on force measurement technologies based on visuotactile sensing from sensing principle,...

Fourier‐Based Action Recognition for Wildlife Behavior Quantification with Event Cameras

11/11/2024 - This article investigates the detection of oscillatory motion in nature, such as in a penguin colony in Antarctica, by combining neuromorphic cameras with frequency‐domain techniques. The pixels of...

Fourier‐Based Action Recognition for Wildlife Behavior Quantification with Event Cameras

DCV2I$text{DCV}^2text{I}$: Leveraging deep vision models to support geographers' visual interpretation in dune segmentation

18/10/2024 - Visual interpretation is extremely important in human geography as the primary technique for geographers to use photograph data in identifying, classifying, and quantifying geographic and topological...

Force Measurement Technology of Vision‐Based Tactile Sensor

23/09/2024 - Force sensing is an important function of marker‐type vision‐based tactile sensors. This review focuses on force measurement technologies based on visuotactile sensing from sensing principle,...

Towards robust visual understanding: A paradigm shift in computer vision from recognition to reasoning

22/09/2024 - Models that learn from data are widely and rapidly being deployed today for real‐world use, but they suffer from unforeseen failures that limit their reliability. These failures often have several...

Deformable Capsules for Object Detection

22/08/2024 - DeformCaps, a novel capsule networks, addresses scalability and deformation modeling challenges in object detection. It introduces. SplitCaps and SE‐Routing, achieving the state‐of‐the‐art...

Maximizing the Synaptic Efficiency of Ferroelectric Tunnel Junction Devices Using a Switching Mechanism Hidden in an Identical Pulse Programming Learning Scheme

17/07/2024 - This article delves into the potential of HfO2‐based ferroelectric tunnel junctions (FTJs) as synaptic devices, highlighting energy efficiency. Focusing on Zr‐doped HfO2 (HZO) FTJs, the...