Article
Chemistry, Analytical
Min-Lung Cheng, Masashi Matsuoka
Summary: The paper proposes a systematic algorithm that combines the SC-EABRISK and ATBB methods to address robustness, number of matches, and processing efficiency issues in image matching. Experimental results show that this algorithm is faster, provides more matches, and achieves higher matching precision compared to previous methods.
Article
Computer Science, Interdisciplinary Applications
Cedric Driesen, Herve Degee, Bram Vandoren
Summary: This study introduces a finite element framework for an adaptive multiresolution multiscale technique to accurately and efficiently simulate large masonry structures. It demonstrates through testing and comparison that the model achieves accurate results with higher computational efficiency than a microscale approach.
COMPUTERS & STRUCTURES
(2021)
Article
Geochemistry & Geophysics
Yuanxin Ye, Chao Yang, Jiacheng Zhang, Jianwei Fan, Ruitao Feng, Yao Qin
Summary: This study proposes a robust matching method by using a multiscale masked structure feature representation. By extracting pixelwise gradient structure features on multiple scales of images and constructing a mask based on large contours, the proposed method significantly improves the matching performance.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2022)
Article
Engineering, Petroleum
Xiaopeng Ma, Kai Zhang, Jian Wang, Chuanjin Yao, Yongfei Yang, Hai Sun, Jun Yao
Summary: This paper introduces a deep learning-based surrogate modeling framework that can directly predict production data from high-dimensional spatial parameters. By combining this surrogate model with an improved data assimilation algorithm, a surrogate-based history-matching workflow is developed.
Article
Mathematics, Interdisciplinary Applications
Fadi Aldakheel, Elsayed S. S. Elsayed, Tarek I. I. Zohdi, Peter Wriggers
Summary: Material modeling using modern numerical methods accelerates the design process and reduces costs. The well-established homogenization techniques for multiscale modeling of heterogeneous materials are computationally expensive. This paper proposes the use of convolutional neural networks (CNNs) as a computationally efficient solution with high accuracy. The CNN model, trained on artificial/real microstructural images, accurately predicts the homogenized stresses of representative volume elements (RVEs) with reduced computation time.
COMPUTATIONAL MECHANICS
(2023)
Article
Engineering, Chemical
A. R. Khoei, A. Rezaei Sameti, H. Mofatteh
Summary: A continuum-atomistic multiscale technique (FEM-MD) is developed to model the compaction process of metallic nano-powders. Nonlinear finite element method (FEM) is used to model the nano-powders as a continuum body, while molecular dynamics (MD) method is used to model the discrete nature of nano-powders through representative volume elements (RVEs). A parametric study is conducted to evaluate the influence of various parameters on the densification behavior of nano-powders. The accuracy of the multiscale analysis is verified by comparing with experimental data, and the computational method is applied to model micro-scale metallic nano-powder components.
Article
Engineering, Electrical & Electronic
Dongxing Liang, Jinshan Ding, Yuhong Zhang
Summary: This article proposes a fast matching approach based on dominant orientation of gradient (DOG) for robust image registration in the presence of nonlinear intensity variations. The method constructs DOG feature maps and utilizes template matching with sum of cosine differences similarity measurement to determine correspondences between images. Additionally, a variable template matching (VTM) method is developed to improve matching precision and performance.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2021)
Article
Geochemistry & Geophysics
Maryam Hajebi, Ahmad Hoorfar
Summary: An efficient multiresolution inverse scattering approach is proposed for profiling high-contrast buried targets in large investigation domains. The method is based on iterative multiscale approach (IMSA) combined with global evolutionary programming (EP) optimization algorithm to guarantee the success of inversion process. The results show superior performance of the proposed technique in handling nonlinearities and outperforming standard IMSA and other well-known global optimization techniques.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2021)
Article
Geochemistry & Geophysics
Liang Zhou, Yuanxin Ye, Tengfeng Tang, Ke Nan, Yao Qin
Summary: This study employs deep learning techniques to enhance image structure features for improved matching between SAR and optical images, showing advantages over other methods in experimental results.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2022)
Article
Mathematics, Applied
Pingbing Ming, Siqi Song
Summary: This article proposes a Nitsche method for multiscale partial differential equations, which is able to retrieve both macroscopic and microscopic information simultaneously. It proves the convergence of the method for second order elliptic problems with bounded and measurable coefficients, and also derives the convergence rate for coefficients with further structures like periodicity and ergodicity. Extensive numerical results confirm the theoretical predictions.
JOURNAL OF SCIENTIFIC COMPUTING
(2022)
Article
Automation & Control Systems
Maria Flores, David Valiente, Arturo Gil, Oscar Reinoso, Luis Paya
Summary: This study evaluates an Adaptive Probability-Oriented Feature Matching method and proposes several improvements to achieve more robust matching. Performance comparisons show that these improvements outperform other methods in the visual odometry framework, such as standard visual odometry and RANSAC-based methods.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2022)
Article
Mathematics, Applied
Chengzhuan Yang, Qian Yu
Summary: Shape is an important visual characteristic that represents objects, and shape recognition is a significant research direction. The invariant multiscale triangle feature (IMTF) is a novel method for robust shape recognition, which can effectively combine boundary and interior characteristics to improve recognition accuracy.
APPLIED MATHEMATICS AND COMPUTATION
(2021)
Article
Automation & Control Systems
Jen-Wei Huang, Peng-Jung Lee, Bijay Prasad Jaysawal
Summary: This research introduces a novel multiscale control chart pattern recognition scheme, which uses histogram-based data representation and time series subsequence matching to identify abnormal patterns at various scales in long series of control charts. Experimental results show that this framework efficiently detects chart patterns at different scales and outperforms state-of-the-art time series subsequence matching algorithms.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2022)
Article
Computer Science, Information Systems
Akash Uikey, Anterpreet Kaur Bedi, Priyankar Choudhary, Wei Tsang Ooi, Mukesh Saini
Summary: Due to the proliferation of video-based applications, there is a high demand for automated systems to support various video-based tasks. This paper presents a novel approach for detecting overlap between two videos by analyzing their audio signals. The proposed framework outperforms other approaches by an average of 13.71% in terms of accuracy.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Dou Quan, Shuang Wang, Ning Huyan, Yi Li, Ruiqi Lei, Jocelyn Chanussot, Biao Hou, Licheng Jiao
Summary: This article focuses on end-to-end image matching through joint key-point detection and descriptor extraction. By improving the network structure and optimization, the proposed approach achieves better matching performance. The proposed concurrent multiscale detector network and rank consistent losses contribute to the improvement in image matching accuracy.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)