• Journal of Electrical Engineering and Technology
• /
• 제10권4호
• /
• pp.1851-1863
• /
• 2015

– In the paper, to handle the problem of traditional bag-of-features model ignoring the spatial relationship of local features in human action recognition, we proposed a Multiscale Spatial Position Coding under Locality Constraint method. Specifically, to describe this spatial relationship, we proposed a mixed feature combining motion feature and multi-spatial-scale configuration. To utilize temporal information between features, sub spatial-temporal-volumes are built. Next, the pooled features of sub-STVs are obtained via max-pooling method. In classification stage, the Locality-Constrained Group Sparse Representation is adopted to utilize the intrinsic group information of the sub-STV features. The experimental results on the KTH, Weizmann, and UCF sports datasets show that our action recognition system outperforms the classical local ST feature-based recognition systems published recently.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제5권12호
• /
• pp.2374-2391
• /
• 2011

A variety of reconstruction methods has been developed to convert a set of scattered points generated from real models into explicit forms, such as polygonal meshes, parametric or implicit surfaces. In this paper, we present a method to construct multi-scale implicit surfaces from scattered points using multiscale kernels based on kernel and multi-resolution analysis theories. Our approach differs from other methods in that multi-scale reconstruction can be done without additional manipulation on input data, calculated functions support level of detail representation, and it can be naturally expanded for n-dimensional data. The method also works well with point-sets that are noisy or not uniformly distributed. We show features and performances of the proposed method via experimental results for various data sets.

• Coupled systems mechanics
• /
• 제10권6호
• /
• pp.521-544
• /
• 2021

In this work, we present the development of a 3D lattice-type model at microscale based upon the Voronoi-cell representation of material microstructure. This model can capture the coupling between mechanic and electric fields with non-linear constitutive behavior for both. More precisely, for electric part we consider the ferroelectric constitutive behavior with the possibility of domain switching polarization, which can be handled in the same fashion as deformation theory of plasticity. For mechanics part, we introduce the constitutive model of plasticity with the Armstrong-Frederick kinematic hardening. This model is used to simulate a complete coupling of the chosen electric and mechanics behavior with a multiscale approach implemented within the same computational architecture.

• Interaction and multiscale mechanics
• /
• 제3권4호
• /
• pp.389-403
• /
• 2010

Based on the Model Coupled Method (MCM), a case study has been carried out on a Concrete-Filled Steel Tubular (CFST) tied arch bridge to investigate the vibration problem. The mathematical model assumed a finite element representation of the bridge together with beam, shell, and link elements, and the vehicle simulation employed a three dimensional linear vehicle model with seven independent degrees-of-freedom. A well-known power spectral density of road pavement profiles defined the road surface roughness for Perfect, Good and Poor roads respectively. In virtue of a home-code program, the dynamic interaction between the bridge and vehicle model was simulated, and the dynamic amplification factors were computed for displacement and internal force. The impact effects of the vehicle on different bridge members and the influencing factors were studied. Meanwhile the acceleration responses of some of the components were analyzed in the frequency domain. From the results some valuable conclusions have been drawn.

• Communications for Statistical Applications and Methods
• /
• 제22권2호
• /
• pp.159-172
• /
• 2015

This paper considers the problem of reconstructing an underlying signal from noisy data. This paper presents a reconstruction method based on synchrosqueezed wavelet transform recently developed for multiscale representation. Synchrosqueezed wavelet transform based on continuous wavelet transform is efficient to estimate the instantaneous frequency of each component that consist of a signal and to reconstruct components. However, an objective selection method for the optimal number of intrinsic mode type functions is required. The proposed method is obtained by coupling the synchrosqueezed wavelet transform with cross-validation scheme. Simulation studies and musical instrument sounds are used to compare the empirical performance of the proposed method with existing methods.

• Journal of Information Processing Systems
• /
• 제17권2호
• /
• pp.411-425
• /
• 2021

In this paper, a vehicle recognition algorithm based on deep convolutional neural network and compression dictionary is proposed. Firstly, the network structure of fine vehicle recognition based on convolutional neural network is introduced. Then, a vehicle recognition system based on multi-scale pyramid convolutional neural network is constructed. The contribution of different networks to the recognition results is adjusted by the adaptive fusion method that adjusts the network according to the recognition accuracy of a single network. The proportion of output in the network output of the entire multiscale network. Then, the compressed dictionary learning and the data dimension reduction are carried out using the effective block structure method combined with very sparse random projection matrix, which solves the computational complexity caused by high-dimensional features and shortens the dictionary learning time. Finally, the sparse representation classification method is used to realize vehicle type recognition. The experimental results show that the detection effect of the proposed algorithm is stable in sunny, cloudy and rainy weather, and it has strong adaptability to typical application scenarios such as occlusion and blurring, with an average recognition rate of more than 95%.

• Journal of Information Processing Systems
• /
• 제19권3호
• /
• pp.370-376
• /
• 2023

Aiming at the problems of low image enhancement accuracy, long enhancement time and poor image quality in the traditional robot sequence motion image enhancement methods, an adaptive enhancement method for robot sequence motion image is proposed. The feature representation of the image was obtained by Karhunen-Loeve (K-L) transformation, and the nonlinear relationship between the robot joint angle and the image feature was established. The trajectory planning was carried out in the robot joint space to generate the robot sequence motion image, and an adaptive homomorphic filter was constructed to process the noise of the robot sequence motion image. According to the noise processing results, the brightness of robot sequence motion image was enhanced by using the multi-scale Retinex algorithm. The simulation results showed that the proposed method had higher accuracy and consumed shorter time for enhancement of robot sequence motion images. The simulation results showed that the image enhancement accuracy of the proposed method could reach 100%. The proposed method has important research significance and economic value in intelligent monitoring, automatic driving, and military fields.

• 한국정보전자통신기술학회논문지
• /
• 제11권3호
• /
• pp.229-234
• /
• 2018

텍스춰는 영상을 분류하거나 분할하는데 사용되는 유용한 특징이다. 기존에 제안되었던 LBP는 텍스춰 영상의 지역적인 특징을 간단한 연산을 통해 성공적으로 추출하는 기법으로서 많은 응용 분야에서 높은 성능을 보인 것으로 확인되었지만 오직 화소값의 차이만을 토대로 특징을 기술하기 때문에 잡음에 약하고 특히 이웃화소의 수가 증가함에 따라 특징벡터의 차원이 기하급수적으로 증가하는 문제점으로 인해 멀티스테일 텍스춰 기술자로서 사용하기에는 제약이 크다. 본 논문은 이런 LBP의 단점을 극복하기 위하여 화소값의 범위를 구간별로 양자화하여 양자화영상의 화소값의 상관관계를 3차원 히스토그램으로 표현하는 기법을 제시한다. 이와 같이 3차원 히스토그램을 이용하여 화소값 사이의 상관 관계를 추출하면 특징벡터의 차원이 선형적으로 증가하는 특성을 가지므로 멀티스케일 텍스춰 기술자로 다양하게 응용될 수 있다. 제안하는 방법을 텍스춰 실험영상을 통해 실험한 결과 텍스춰를 분류하는 문제에 있어서 LBP와 비교하여 유의 수준의 성능의 향상을 확인하였다.

• 한국정보과학회논문지:컴퓨팅의 실제 및 레터
• /
• 제16권7호
• /
• pp.768-774
• /
• 2010

흩어진 점 데이터의 집합으로부터 음함수 곡면을 생성하는 기법들이 다양한 과학과 공학 분야에서 개발되어 활용되고 있다. 본 논문에서는 다중스케일 커널을 이용하여 2차원 점 데이터를 함수 형태로 표현하고, 이것이 실시간 데이터 액세스가 필요한 그래픽스 응용에 효과적으로 활용될 수 있음을 보인다. 전처리 단계에서 계산되고 저장된 함수 계수들을 이용해, 실시간 응용 프로그램에서 임의의 위치에 대한 함수 값을 액세스하는 과정은 기존의 연구 방법들과 유사하지만, 실시간 처리 과정에서 사용자가 원하는 섬세한 레벨의 함수 값을 자유롭게 선택할 수 있다는 점에서 본 기법은 다른 기법들과 차별된다. 내재적으로 다중해상도 표현을 지원하는 함수를 계산할 수 있는 것은 멀티 스케일 커널이 갖는 수학적인 특성에 기인하며, 이 커널은 2차원뿐만 아니라 n차원 데이터의 다중해상도 표현을 위해 확장가능하다.

• Interaction and multiscale mechanics
• /
• 제1권2호
• /
• pp.231-250
• /
• 2008

The statistical two-order and two-scale method is developed for predicting the mechanics parameters, such as stiffness and strength of core-shell particle-filled polymer composites. The representation and simulation on meso-configuration of random particle-filled polymers are stated. And the major statistical two-order and two-scale analysis formulation is briefly given. The two-order and two-scale expressions for the strains and stresses of conventionally strength experimental components, including the tensional or compressive column, the twist bar and the bending beam, are developed by means of their classical solutions with orthogonal-anisotropic coefficients. Then a new effective mesh generation algorithm is presented. The mechanics parameters of core-shell particle-filled polymer composites, including the expected stiffness parameters, minimum stiffness parameters, and the expected elasticity limit strength and the minimum elasticity limit strength, are defined by means of the stiffness coefficients and elasticity strength criterions for core, shell and matrix. Finally, the numerical results for predicting both stiffness and elasticity limit strength parameters are compared with the experimental data.