Results 311 to 320 of about 11,025,050 (395)

Widened Attention‐Enhanced Atrous Convolutional Network for Efficient Embedded Vision Applications under Resource Constraints

Advanced Intelligent Systems, EarlyView.
The article introduces WACEfNet, a new convolutional neural network architecture optimized for efficient aerial image analysis under resource constraints. It creatively integrates attention mechanisms and atrous convolutions into a compact widened residual network framework.
Md Meftahul Ferdaus, Mahdi Abdelguerfi, Kendall N. Niles, Ken Pathak, Joe Tom   +4 more
wiley   +1 more source

Research on the Polishing Process of Wolter-I Type Grazing Incidence Mirrors. [PDF]

PLoS One
Ding S, Yao S, Sun L, Feng F, Chen D.
europepmc   +1 more source

A Study on the Effect of Structural Compliance Placing in Soft Contact/Collision Properties of Multirotor Micro Aerial Vehicles

Advanced Intelligent Systems, EarlyView.
This article answers the question “Where is the best place to apply compliance in a multirotor micro aerial vehicle to make it more collision‐resilient?” by analyzing the output data of more than 1200 drone collision tests for four main configurations of micro‐quadcopters each possessing a unique softness layout of physical frame.
Amirali Abazari, Alihan Bakir, Altar Sertpoyraz, Onur Ozcan   +3 more
wiley   +1 more source

Comparison of algorithms using deep reinforcement learning for optimization of hyperbolic metamaterials. [PDF]

Sci Rep
Hamada K, Hsiao HH, Kubo W.
europepmc   +1 more source

A Privacy‐Preserving System for Confidential Carpooling Services Using Homomorphic Encryption

Advanced Intelligent Systems, EarlyView.
This research presents a homomorphic encryption‐based solution to enhance privacy in carpooling without compromising functionality. Rigorous experiments confirm its robustness, accuracy, and efficiency in securely matching users while preserving confidentiality.
David Palma, Pier Luca Montessoro, Mirko Loghi, Daniele Casagrande   +3 more
wiley   +1 more source

A shape control method for soap bubble simulation using external forces. [PDF]

Sci Rep
Horimoto A, Sato S, Sakurai K, Nakamoto K.   +3 more
europepmc   +1 more source

A Physics‐Informed Neural Network as a Digital Twin of Optically Turbid Media

Advanced Intelligent Systems, EarlyView.
In this work, a physics‐informed neural network is presented as a digital twin for modeling and studying optically turbid media without requiring reference light. The model learns and integrates complex wavefront modulation behavior, ensuring robustness without explicit calibration.
Mohammadrahim Kazemzadeh, Liam Collard, Linda Piscopo, Massimo De Vittorio, Ferruccio Pisanello   +4 more
wiley   +1 more source

Bifurcation analysis and dynamical behavior of novel optical soliton solution of chiral (2 + 1) dimensional nonlinear Schrodinger equation in telecommunication system. [PDF]

Sci Rep
Saber H, Roshid MM, Bouye M, Muhyi A, Moumen A, Aldwoah K.   +5 more
europepmc   +1 more source

Deep Learning Methods in Soft Robotics: Architectures and Applications

Advanced Intelligent Systems, EarlyView.
Soft robotics has seen intense research over the past two decades and offers a promising approach for future robotic applications. However, standard industrial methods may be challenging to apply to soft robots. Recent advances in deep learning provide powerful tools to analyze and design complex soft machines that can operate in unstructured ...
Tomáš Čakurda, Monika Trojanová, Pavlo Pomin, Alexander Hošovský   +3 more
wiley   +1 more source