• 한국멀티미디어학회논문지
• /
• 제24권5호
• /
• pp.659-666
• /
• 2021

With the continuous development of deep learning, human behavior recognition algorithms have achieved good results. However, in a multi-person recognition environment, the complex behavior environment poses a great challenge to the efficiency of recognition. To this end, this paper proposes a multi-person pose estimation model. First of all, the human detectors in the top-down framework mostly use the two-stage target detection model, which runs slow down. The single-stage YOLOv3 target detection model is used to effectively improve the running speed and the generalization of the model. Depth separable convolution, which further improves the speed of target detection and improves the model's ability to extract target proposed regions; Secondly, based on the feature pyramid network combined with context semantic information in the pose estimation model, the OHEM algorithm is used to solve difficult key point detection problems, and the accuracy of multi-person pose estimation is improved; Finally, the Euclidean distance is used to calculate the spatial distance between key points, to determine the similarity of postures in the frame, and to eliminate redundant postures.

• 한국전자통신학회논문지
• /
• 제16권1호
• /
• pp.109-116
• /
• 2021

RDF(Resource Description Framework) 데이터 구조는 그래프로 모델링하기 때문에, 관계형 데이터베이스와 XML 기술의 기존 솔루션은 RDF 모델에 바로 적용하기 어렵다. 우리는 링크 데이터를 더욱 효과적으로 저장하고, 인덱스하고, 검색하기 위해 융합 인덱싱 방법을 제안한다. 이 방법은 HDD(Hard Disk Drive) 와 SSD(Solid State Drive) 디바이스에 기반한 하이브리드 스토리지 시스템을 사용하고, 불필요한 데이터를 필터하고 중간 결과를 정제하기 위해 분리된 필터 및 정제 인덱스 구조를 사용한다. 우리는 3개의 표준 조인 검색알고리즘에 대한 성능 비교를 수행했는데, 실험 결과 제안된 방법이 Quad와 Darq와 같은 다른 기존 방법들에 비해 뛰어난 성능을 보인다.

• 전기전자학회논문지
• /
• 제24권4호
• /
• pp.991-997
• /
• 2020

강화학습은 주로 순차적인 의사 결정 문제에 적용되어 왔다. 특히 최근에는 신경망과 결합한 형태로 기존에는 해결하지 못한 분야에서도 성공적인 결과를 내고 있다. 하지만 신경망을 이용하는 강화학습은 현장에서 즉각적으로 사용하기엔 너무 복잡하다는 단점이 있다. 본 논문에서는 학습이 쉬운 강화학습 알고리즘 중 하나인 Q-learning을 이용하여 이동 로봇의 경로를 생성하는 알고리즘을 구현하였다. Q-table을 미리 만드는 방식의 Q-learning은 명확한 한계를 가지기 때문에 실시간으로 Q-table을 업데이트하는 실시간 Q-learning을 사용하였다. 탐험 전략을 조정하여 실시간 Q-learning에 필요한 학습 속도를 얻을 수 있었다. 마지막으로 실시간 Q-learning과 DQN의 성능을 비교하였다.

• Steel and Composite Structures
• /
• 제38권5호
• /
• pp.497-521
• /
• 2021

In this study, an effective numerical method is introduced for nonlinear inelastic analyses of rectangular concrete-filled steel tubular (CFST) frames for the first time. A steel-concrete composite fiber beam-column element model is developed that considers material, and geometric nonlinearities, and residual stresses. This is achieved by using stability functions combined with integration points along the element length to capture the spread of plasticity over the composite cross-section along the element length. Additionally, a multi-spring element with a zero-length is employed to model the nonlinear semi-rigid beam-to-column connections in CFST frame models. To solve the nonlinear equilibrium equations, the generalized displacement control algorithm is adopted. The accuracy of the proposed method is firstly verified by a large number of experiments of CFST members subjected to various loading conditions. Subsequently, the proposed method is applied to investigate the nonlinear inelastic behavior of rectangular CFST frames with fully rigid, semi-rigid, and hinged connections. The accuracy of the predicted results and the efficiency pertaining to the computation time of the proposed method are demonstrated in comparison with the ABAQUS software. The proposed numerical method may be efficiently utilized in practical designs for advanced analysis of the rectangular CFST structures.

• Current Optics and Photonics
• /
• 제6권3호
• /
• pp.313-322
• /
• 2022

Remote-sensing optical payloads, especially hyperspectral imagers, have particular issues with stray light because they often encounter high-contrast target/background conditions, such as sun glint. While developing an optical payload, we usually apply several stray-light analysis methods, including forward and backward analyses, separately or in combination, to support lens design and optomechanical design. In addition, we often characterize the stray-light response over a full field to support calibration, or when developing an algorithm to correct stray-light errors. For this purpose, we usually use forward analysis across the entire field, but this requires a tremendous amount of computational time. In this paper, we propose a sequence of forward-backward-forward analyses to more effectively investigate the through-field response of stray light, utilizing the combined advantages of the individual methods. The application is an airborne hyperspectral imager for creating hyperspectral maps from 900 to 1700 nm in a 5-nm-continuous band. With the proposed method, we have investigated the through-field response of stray light to an effective accuracy of 0.1°, while reducing computation time to 1/17^th of that for a conventional, forward-only stray-light analysis.

• Structural Engineering and Mechanics
• /
• 제83권1호
• /
• pp.93-108
• /
• 2022

Real-time hybrid simulation (RTHS) was applied to investigate the train-bridge interaction of a high-speed railway system, where the railway bridge was selected as the numerical substructure, and the train was physically tested. The interaction between the two substructures was reproduced by a servo-hydraulic shaking table. To accurately reproduce the high-frequency interaction responses ranging from 10-25Hz using the hydraulic shaking table with an inherent delay of 6-50ms, an adaptive time series (ATS) compensation algorithm combined with the linear quadratic Gaussian (LQG) was proposed and implemented in the RTHS. Testing cases considering different train speeds, track irregularities, bridge girder cross-sections, and track settlements featuring a wide range of frequency contents were conducted. The performance of the proposed ATS+LQG delay compensation method was compared to the ATS method and RTHS without any compensation in terms of residual time delays and root mean square errors between commands and responses. The effectiveness of the ATS+LQG method to compensate time delay in RTHS with high-frequency responses was demonstrated and the proposed ATS+LQG method outperformed the ATS method in yielding more accurate responses with less residual time delays.

• Wind and Structures
• /
• 제33권6호
• /
• pp.423-435
• /
• 2021

Although existing algorithms can predict wind speed using historical observation data, for engineering feasibility, most use moment methods and probability density functions to estimate fitted parameters. However, extreme wind speed prediction accuracy for long-term return periods is not always dependent on how the optimized frequency distribution curves are obtained; long-term return periods emphasize general distribution effects rather than marginal distributions, which are closely related to potential extreme values. Moreover, there are different wind speed parent sample types; how to theoretically select the proper extreme value distribution is uncertain. The influence of different sampling time intervals has not been evaluated in the fitting process. To overcome these shortcomings, updated steps are introduced, involving parameter sensitivity analysis for different sampling time intervals. The extreme value prediction accuracy of unknown parent samples is also discussed. Probability analysis of mean wind is combined with estimation of the probability plot correlation coefficient and the maximum likelihood method; an iterative estimation algorithm is proposed. With the updated steps and comparison using a Monte Carlo simulation, a fitting policy suitable for different parent distributions is proposed; its feasibility is demonstrated in extreme wind speed evaluations at Longhua and Chuansha meteorological stations in Shanghai, China.

• Journal of Multimedia Information System
• /
• 제8권4호
• /
• pp.233-242
• /
• 2021

Various anatomical MRI imaging biomarkers for Alzheimer's Disease (AD) identification have been recognized so far. Cortical and subcortical volume, hippocampal, amygdala volume, and genetics patterns have been utilized successfully to diagnose AD patients from healthy. These fundamental sMRI bio-measures have been utilized frequently and independently. The entire possibility of anatomical MRI imaging measures for AD diagnosis might thus still to analyze fully. Thus, in this paper, we merge different structural MRI imaging biomarkers to intensify diagnostic classification and analysis of Alzheimer's. For 54 clinically pronounce Alzheimer's patients, 58 cognitively healthy controls, and 99 Mild Cognitive Impairment (MCI); we calculated 1. Cortical and subcortical features, 2. The hippocampal subfield, amygdala nuclei volume using Freesurfer (6.0.0) and 3. Genetics (APoE ε4) biomarkers were obtained from the ADNI database. These three measures were first applied separately and then combined to predict the AD. After feature combination, we utilize the sequential feature selection [SFS (wrapper)] method to select the top-ranked features vectors and feed them into the Multi-Kernel SVM for classification. This diagnostic classification algorithm yields 94.33% of accuracy, 95.40% of sensitivity, 96.50% of specificity with 94.30% of AUC for AD/HC; for AD/MCI propose method obtained 85.58% of accuracy, 95.73% of sensitivity, and 87.30% of specificity along with 91.48% of AUC. Similarly, for HC/MCI, we obtained 89.77% of accuracy, 96.15% of sensitivity, and 87.35% of specificity with 92.55% of AUC. We also presented the performance comparison of the proposed method with KNN classifiers.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권4호
• /
• pp.1187-1208
• /
• 2022

One of the main goals of wireless sensor networks (WSNs) is to utilize the energy of sensor nodes effectively and maximize the network lifetime. Thus, this paper proposed a routing protocol for WSNs based on virtual force disturbing mobile Sink node (VFMSR). According to the number of sensor nodes in the cluster, the average energy and the centroid factor of the cluster, a new cluster head (CH) election fitness function was designed. At the same time, a hexagonal fixed-point moving trajectory model with the best radius was constructed, and the virtual force was introduced to interfere with it, so as to avoid the frequent propagation of sink node position information, and reduce the energy consumption of CH. Combined with the improved ant colony algorithm (ACA), the shortest transmission path to Sink node was constructed to reduce the energy consumption of long-distance data transmission of CHs. The simulation results showed that, compared with LEACH, EIP-LEACH, ANT-LEACH and MECA protocols, VFMSR protocol was superior to the existing routing protocols in terms of network energy consumption and network lifetime, and compared with LEACH protocol, the network lifetime was increased by more than three times.

• Smart Structures and Systems
• /
• 제29권2호
• /
• pp.321-336
• /
• 2022

Model uncertainty is a key factor that could influence the accuracy and reliability of numerical model-based analysis. It is necessary to acquire an appropriate updating approach which could search and determine the realistic model parameter values from measurements. In this paper, the Bayesian model updating theory combined with the transitional Markov chain Monte Carlo (TMCMC) method and K-means cluster analysis is utilized in the updating of the structural model parameters. Kriging and polynomial chaos expansion (PCE) are employed to generate surrogate models to reduce the computational burden in TMCMC. The selected updating approaches are applied to three structural examples with different complexity, including a two-storey frame, a ten-storey frame, and the national stadium model. These models stand for the low-dimensional linear model, the high-dimensional linear model, and the nonlinear model, respectively. The performances of updating in these three models are assessed in terms of the prediction uncertainty, numerical efforts, and prior information. This study also investigates the updating scenarios using the analytical approach and surrogate models. The uncertainty quantification in the Bayesian approach is further discussed to verify the validity and accuracy of the surrogate models. Finally, the advantages and limitations of the surrogate model-based updating approaches are discussed for different structural complexity. The possibility of utilizing the boosting algorithm as an ensemble learning method for improving the surrogate models is also presented.