Article
Computer Science, Interdisciplinary Applications
Rogerio G. Negri, Alejandro C. Frery
Summary: This study proposes a novel framework for imagery dataset simulation, specifically designed for evaluating and comparing change detection methods. The framework is versatile, allowing the use of both supervised and unsupervised techniques. Applying this framework, the performance of well-known algorithms was compared using simulated data resembling deforestation in a forest area observed by the Landsat 5 TM sensor. The results provide insights into method performance and demonstrate the usefulness of the proposed framework.
COMPUTERS & GEOSCIENCES
(2023)
Article
Engineering, Mechanical
K. E. Tatsis, V. K. Dertimanis, C. Papadimitriou, E. Lourens, E. N. Chatzi
Summary: This paper presents a general framework for estimating the state and unknown inputs at the level of a system subdomain using a limited number of output measurements. The proposed framework is based on fixed and free-interface domain decomposition methods and verified by simulated Wind Turbine structure applications. The accuracy of the estimated unknown quantities is assessed to evaluate the performance.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2021)
Article
Engineering, Mechanical
Jacopo Brunetti, Walter D'Ambrogio, Annalisa Fregolent
Summary: Contact and friction forces in complex vibrating systems can lead to dynamic responses, requiring consideration of both macroscopic sliding and local vibrations at contact points. While classical techniques assume time-invariant component subsystems and coupling conditions, a proposed method allows for time-dependent coupling conditions, improving the analysis of friction-induced vibrations.
NONLINEAR DYNAMICS
(2021)
Article
Acoustics
Jun Gu Kim, Hansol Park, Munhwan Cho, David P. Song, Yeon June Kang
Summary: This study proposes an improved dynamic substructuring model that predicts the performance of a coupling system using the frequency response function information at coupling points. The method utilizes generalized coupling properties and virtual point transformation to accurately estimate the dynamic characteristics of the assembled system when direct measurements are not feasible.
JOURNAL OF VIBRATION AND CONTROL
(2022)
Article
Genetics & Heredity
Lifei Wang, Xuexia Miao, Rui Nie, Zhang Zhang, Jiang Zhang, Jun Cai
Summary: MultiCapsNet is a novel deep learning model that combines CapsNet and scCapsNet, providing easy data integration and high interpretability. Tests on multiple datasets demonstrate that MultiCapsNet shows similar classification performance to neural network models when dealing with modular inputs, while also providing direct importance scores for data sources, further proving its effectiveness and interpretability.
FRONTIERS IN GENETICS
(2021)
Article
Public, Environmental & Occupational Health
Fatma Taher, Mohamed R. Shoaib, Heba M. Emara, Khaled M. Abdelwahab, Fathi E. Abd El-Samie, Mohammad T. Haweel
Summary: This paper introduces a method for brain tumor classification using deep learning technology. By incorporating transfer learning and a self-trained neural network model, high accuracy is achieved in brain magnetic resonance image recognition.
FRONTIERS IN PUBLIC HEALTH
(2022)
Article
Environmental Sciences
Emre Ozdemir, Fabio Remondino, Alessandro Golkar
Summary: With recent advances in technologies, deep learning has been increasingly applied to various tasks. In this study, an efficient framework named TONIC is proposed, which can work with any kind of aerial data source (LiDAR or photogrammetry) without requiring high computational power, achieving accuracy on par with the current state of the art methods. The framework also demonstrates the ability to generalize and predict on unseen aerial scenarios.
Article
Biochemistry & Molecular Biology
Zuguang Gu, Matthias Schlesner, Daniel Huebschmann
Summary: Consensus partitioning is a widely used method for stability evaluation, and the cola software package offers a general framework and a new method named ATC for this process. It also provides rich functionalities for automated analysis and HTML report generation.
NUCLEIC ACIDS RESEARCH
(2021)
Article
Biology
Alasdair I. Houston, Lutz Fromhage, John M. McNamara
Summary: The behavior of animals can impact various variables, such as their energy reserves, risk of injury or mortality, and reproduction rate. This study presents a unified framework that explains how these trade-offs fit together, covering a wide range of biological phenomena and helping to identify gaps in knowledge. Additionally, a new model of foraging under the risk of predation and damage accumulation is introduced to fill one of these gaps.
BIOLOGICAL REVIEWS
(2023)
Article
Environmental Sciences
Yuan Gao, Xiaojuan Sun, Chao Liu
Summary: This paper explores the combination of self-supervised learning and remote sensing, proposes a self-supervised pre-training framework, and discovers the attribution consensus between the pre-trained model and downstream tasks. The research suggests that self-supervised learning can enhance general model representation in the remote sensing domain, and achieves higher accuracy in remote sensing scene classification tasks.
Article
Engineering, Multidisciplinary
Dave Deriso, Stephen Boyd
Summary: The goal of dynamic time warping is to approximately align two signals by transforming or warping time. This method poses the choice of warping function as an optimization problem with three objective terms, and it uses discretization and dynamic programming to find the (globally) optimal warping function. The method achieves high accuracy and is implemented as an open source Python package called GDTW.
OPTIMIZATION AND ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
Rakesh Kumar Sanodiya, Leehter Yao
Summary: Transfer learning is proposed as an effective way to learn the classifier in a new domain by utilizing supervision information from related domains. Existing methods aim to minimize geometric and distribution shifts while preserving source domain discriminant information and data similarity. However, they face challenges in preserving target domain discriminant information and may consider unnecessary data for classification.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Mathematics, Interdisciplinary Applications
Marium Mehmood, Nasser Alshammari, Saad Awadh Alanazi, Fahad Ahmad
Summary: The liver is a vital organ in the human body, but early detection of liver diseases is challenging due to hidden symptoms. Early treatment and designing disease diagnosis models are important. By extracting essential features through data mining and soft computing techniques, evaluating results using different regression algorithms, it is found that Random Forest Regression with the Wrapper Method performs the best.
Article
Engineering, Mechanical
Verena Gimpl, Alfredo Fantetti, Steven W. B. Klaassen, Christoph W. Schwingshackl, Daniel J. Rixen
Summary: The tangential contact stiffness is a significant parameter for non-linear dynamic analyses of jointed structures. This paper proposes a comparative study between hysteresis loop measurements and Frequency Based Substructuring (FBS) as two experimental techniques for contact stiffness estimations. The results validate FBS approaches against hysteresis measurements and highlight the strengths and limitations of both methods, ultimately contributing to a better understanding of contact stiffness in jointed structures.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2022)
Article
Engineering, Mechanical
Jose-Luis Blanco-Claraco, Antonio Leanza, Giulio Reina
Summary: This paper presents a novel general framework based on factor graph theory for solving kinematic and dynamic problems of multibody systems. The framework models and simulates multibody systems using factor graphs, with batch optimization or a fixed lag smoother to solve the optimization problem. The proposed framework has been extensively tested in simulations, showing potential for applications related to motion estimation and parameter identification in complex mechanical systems.
NONLINEAR DYNAMICS
(2021)