• Journal of Korea Multimedia Society
• /
• 제24권3호
• /
• pp.423-430
• /
• 2021

Aiming Graph convolutional networks (GCNs) have achieved outstanding performances on skeleton-based action recognition. However, several problems remain in existing GCN-based methods, and the problem of low recognition rate caused by single input data information has not been effectively solved. In this article, we propose a Dual-stream fusion method that combines video data and skeleton data. The two networks respectively identify skeleton data and video data and fuse the probabilities of the two outputs to achieve the effect of information fusion. Experiments on two large dataset, Kinetics and NTU-RGBC+D Human Action Dataset, illustrate that our proposed method achieves state-of-the-art. Compared with the traditional method, the recognition accuracy is improved better.

• Journal of Information Processing Systems
• /
• 제18권4호
• /
• pp.480-488
• /
• 2022

Images of tobacco retail licenses have complex unstructured characteristics, which is an urgent technical problem in the robot process automation of tobacco marketing. In this paper, a novel recognition approach using a double attention mechanism is presented to realize the automatic recognition and information extraction from such images. First, we utilized a DenseNet network to extract the license information from the input tobacco retail license data. Second, bi-directional long short-term memory was used for coding and decoding using a continuous decoder integrating dual attention to realize the recognition and information extraction of tobacco retail license images without segmentation. Finally, several performance experiments were conducted using a largescale dataset of tobacco retail licenses. The experimental results show that the proposed approach achieves a correction accuracy of 98.36% on the ZY-LQ dataset, outperforming most existing methods.

• East Asian mathematical journal
• /
• 제25권1호
• /
• pp.55-68
• /
• 2009

In this paper we propose new large-update primal-dual inte-rior point algorithms for $P_*(\kappa)$ linear complementarity problems(LCPs). New search directions and proximity measures are proposed based on the kernel function$\psi(t)=\frac{t^{p+1}-1}{p+1}+\frac{e^{\frac{1}{t}}-e}{e}$,$p{\in}$[0,1]. We showed that if a strictly feasible starting point is available, then the algorithm has $O((1+2\kappa)(logn)^{2}n^{\frac{1}{p+1}}log\frac{n}{\varepsilon}$ complexity bound.

• Steel and Composite Structures
• /
• 제45권4호
• /
• pp.535-546
• /
• 2022

The memory response of nonlocal systematical formulation size-dependent coupling of viscoelastic deformation and thermal fields for piezoelectric materials with dual-phase lag heat conduction law is constructed. The method of the matrix exponential, which constitutes the basis of the state-space approach of modern control theory, is applied to the non-dimensional equations. The resulting formulation together with the Laplace transform technique is applied to solve a problem of a semi-infinite piezoelectric rod subjected to a continuous heat flux with constant time rates. The inversion of the Laplace transforms is carried out using a numerical approach. Some comparisons of the impacts of nonlocal parameters and time-delay constants for various forms of kernel functions on thermal spreads and thermo-viscoelastic response are illustrated graphically.

• Proceedings of the Korea Information Processing Society Conference
• /
• 한국정보처리학회 2023년도 춘계학술발표대회
• /
• pp.496-498
• /
• 2023

Brain tumor segmentation problem has challenges in the tumor diversity of location, imbalance, and morphology. Attention mechanisms have recently been used widely to tackle medical segmentation problems efficiently by focusing on essential regions. In contrast, the fusion approaches enhance performance by merging mutual benefits from many models. In this study, we proposed a 3D dual fusion attention network to combine the advantages of fusion approaches and attention mechanisms by residual self-attention and local blocks. Compared to fusion approaches and related works, our proposed method has shown promising results on the BraTS 2018 dataset.

• Current Optics and Photonics
• /
• 제8권2호
• /
• pp.170-182
• /
• 2024

The dimensional accuracy and consistency of a dual oil circuit centrifugal fuel nozzle are important for fuel distribution and combustion efficiency in an engine combustion chamber. A point cloud measurement method was proposed to solve the geometric accuracy detection problem for the fuel nozzle. An improved variance focus measure operator was used to extract the depth point cloud. Compared with other traditional sharpness evaluation functions, the improved operator can generate the best evaluation curve, and has the least noise and the shortest calculation time. The experimental results of point cloud slicing measurement show that the best window size is 24 × 24 pixels. In the height measurement experiment of the standard sample block, the relative error is 2.32%, and in the fuel nozzle cone angle measurement experiment, the relative error is 2.46%, which can meet the high precision requirements of a dual oil circuit centrifugal fuel nozzle.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제7권2호
• /
• pp.288-307
• /
• 2013

This paper formulates resource allocation for decode-and-forward (DF) relay assisted multi-cell orthogonal frequency division multiple (OFDM) networks as an optimization problem taking into account of inter-cell interference and users fairness. To maximize the transmit rate of system we propose a joint interference coordination, subcarrier and power allocation algorithm. To reduce the complexity, this semi-distributed algorithm divides the primal optimization into three sub-optimization problems, which transforms the mixed binary nonlinear programming problem (BNLP) into standard convex optimization problems. The first layer optimization problem is used to get the optimal subcarrier distribution index. The second is to solve the problem that how to allocate power optimally in a certain subcarrier distribution order. Based on the concept of equivalent channel gain (ECG) we transform the max-min function into standard closed expression. Subsequently, with the aid of dual decomposition, water-filling theorem and iterative power allocation algorithm the optimal solution of the original problem can be got with acceptable complexity. The third sub-problem considers dynamic co-channel interference caused by adjacent cells and redistributes resources to achieve the goal of maximizing system throughput. Finally, simulation results are provided to corroborate the proposed algorithm.

• Journal of the Korean Operations Research and Management Science Society
• /
• 제23권2호
• /
• pp.29-46
• /
• 1998

In this paper, we develop detailed algorithms for implementing the so-called Limited Column Generation procedure for Local Access Telecommunication Network(LATN) design problem. We formulate the problem into a tree-partitioning problem with an exponential number of variables. Its linear programming relaxation has all integral vertices, and can be solved by the Limited Column Generation procedure in just n pivots, where n is the number of nodes in the network. Prior to each pivot. an entering variable is selected by detecting the Locally Most Violated(LMV) reduced cost, which can be obtained by solving a subproblem in pseudo-polynomial time. A critical step in the Limited Column Generation is to find all the LMV reduced costs. As dual variables are updated at each pivot, the reduced costs have to be computed in an on-line fashion. An efficient implementation is developed to execute such a task so that the LATN design problem can be solved in O(n$^2$H), where H is the maximum concentrator capacity. Our computational experiments indicate that our algorithm delivers an outstanding performance. For instance, the LATN design problem with n＝150 and H＝1000 can be solved in approximately 67 seconds on a SUN SPARC 1000 workstation.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 대한산업공학회/한국경영과학회 1995년도 춘계공동학술대회논문집; 전남대학교; 28-29 Apr. 1995
• /
• pp.54-68
• /
• 1995

In this paper, we develop detailed algorithms for implementing the so-called Limited Column Generation procedure for Local Access Telecommunication Network (LATN) Design problem. We formulate the problem into a tree-partitioning problem with an exponential number of variables. Its linear programming relaxation has all integral vertices, and can be solved by the Limited Column. Generation procedure in just n pivots, where n is the number of nodes in the network. Prior to each pivot, an entering variable is selected by detecting the Locally Most Violated (LMV) reduced cost, which can be obtained by solving a subproblem in pseudo-polynomial time. A critical step in the Limited Column Generation is to find all the LMV reduced costs. As dual variables are updated at each pivot, the reduced costs have to be computed in an on-line fashion. An efficient implementation is developed to execute such a task so that the LATN Design problem can be solved in O(n$^{2}$H), where H is the maximum concentrator capacity. Our computational experiments indicate that our algorithm delivers an outstanding performance. For instance, the LATN Design problem with n = 150 and H = 1000 can be solved in approximately 67 seconds on a SUN SPARC 1000 workstation.

• Journal of applied mathematics & informatics
• /
• 제17권1_2_3호
• /
• pp.121-144
• /
• 2005

In this paper, we propose unconstrained and constrained posynomial Geometric Programming (GP) problem with negative or positive integral degree of difficulty. Conventional GP approach has been modified to solve some special type of GP problems. In specific case, when the degree of difficulty is negative, the normality and the orthogonality conditions of the dual program give a system of linear equations. No general solution vector exists for this system of linear equations. But an approximate solution can be determined by the least square and also max-min method. Here, modified form of geometric programming method has been demonstrated and for that purpose necessary theorems have been derived. Finally, these are illustrated by numerical examples and applications.