• Journal of Information Processing Systems
• /
• 제17권4호
• /
• pp.787-800
• /
• 2021

Three-dimensional (3D) object classification tasks using point clouds are widely used in 3D modeling, face recognition, and robotic missions. However, processing raw point clouds directly is problematic for a traditional convolutional network due to the irregular data format of point clouds. This paper proposes a pointwise convolution neural network (CNN) structure that can process point cloud data directly without preprocessing. First, a 2D convolutional layer is introduced to percept coordinate information of each point. Then, multiple 2D convolutional layers and a global max pooling layer are applied to extract global features. Finally, based on the extracted features, fully connected layers predict the class labels of objects. We evaluated the proposed pointwise CNN structure on the ModelNet10 dataset. The proposed structure obtained higher accuracy compared to the existing methods. Experiments using the ModelNet10 dataset also prove that the difference in the point number of point clouds does not significantly influence on the proposed pointwise CNN structure.

• 대한수학회보
• /
• 제52권5호
• /
• pp.1711-1719
• /
• 2015

We obtain a new criterion for the boundedness and compactness of the weighted composition operators ${\psi}C_{\varphi}$ from ${\ss}^{{\alpha}}$(0 < ${\alpha}$ < 1) to ${\ss}^{{\beta}}$ in terms of the sequence $\{{\psi}{\varphi}^n\}$. An estimate for the essential norm of ${\psi}C_{\varphi}$ is also given.

• Smart Structures and Systems
• /
• 제10권4_5호
• /
• pp.375-391
• /
• 2012

This paper reports the structural health monitoring benchmark study results for the Canton Tower using Bayesian methods. In this study, output-only modal identification and finite element model updating are considered using a given set of structural acceleration measurements and the corresponding ambient conditions of 24 hours. In the first stage, the Bayesian spectral density approach is used for output-only modal identification with the acceleration time histories as the excitation to the tower is unknown. The modal parameters and the associated uncertainty can be estimated through Bayesian inference. Uncertainty quantification is important for determination of statistically significant change of the modal parameters and for weighting assignment in the subsequent stage of model updating. In the second stage, a Bayesian model updating approach is utilized to update the finite element model of the tower. The uncertain stiffness parameters can be obtained by minimizing an objective function that is a weighted sum of the square of the differences (residuals) between the identified modal parameters and the corresponding values of the model. The weightings distinguish the contribution of different residuals with different uncertain levels. They are obtained using the Bayesian spectral density approach in the first stage. Again, uncertainty of the stiffness parameters can be quantified with Bayesian inference. Finally, this Bayesian framework is applied to the 24-hour field measurements to investigate the variation of the modal and stiffness parameters under changing ambient conditions. Results show that the Bayesian framework successfully achieves the goal of the first task of this benchmark study.

• The Korean Journal of Physiology and Pharmacology
• /
• 제20권4호
• /
• pp.347-355
• /
• 2016

Cryptotanshinone (CPT) is a natural compound isolated from traditional Chinese medicine Salvia miltiorrhiza Bunge. In the present study, the regulatory effect and potential mechanisms of CPT on tumor necrosis factor alpha ($TNF-{\alpha}$) induced lectin-like receptor for oxidized low density lipoprotein (LOX-1) were investigated. Human umbilical vein endothelial cells (HUVECs) were cultured and the effect of $TNF-{\alpha}$ on LOX-1 expression at mRNA and protein levels was determined by Real-time PCR and Western blotting respectively. The formation of intracellular ROS was determined with fluorescence probe $CM-DCFH_2-DA$. The endothelial ox-LDL uptake was evaluated with DiI-ox-LDL. The effect of CPT on LOX-1 expression was also evaluated with SD rats. $TNF-{\alpha}$ induced LOX-1 expression in a dose- and time- dependent manner in endothelial cells. $TNF-{\alpha}$ induced ROS formation, phosphorylation of $NF-{\kappa}B$ p65 and ERK, and LOX-1 expression, which were suppressed by rotenone, DPI, NAC, and CPT. $NF-{\kappa}B$ inhibitor BAY11-7082 and ERK inhibitor PD98059 inhibited $TNF-{\alpha}-induced$ LOX-1 expression. CPT and NAC suppressed $TNF-{\alpha}-induced$ LOX-1 expression and phosphorylation of $NF-{\kappa}B$ p65 and ERK in rat aorta. These data suggested that $TNF-{\alpha}$ induced LOX-1 expression via ROS activated $NF-{\kappa}B/ERK$ pathway, which could be inhibited by CPT. This study provides new insights for the anti-atherosclerotic effect of CPT.

• Journal of Information Processing Systems
• /
• 제14권6호
• /
• pp.1445-1456
• /
• 2018

Environment perception and three-dimensional (3D) reconstruction tasks are used to provide unmanned ground vehicle (UGV) with driving awareness interfaces. The speed of obstacle segmentation and surrounding terrain reconstruction crucially influences decision making in UGVs. To increase the processing speed of environment information analysis, we develop a CPU-GPU hybrid system of automatic environment perception and 3D terrain reconstruction based on the integration of multiple sensors. The system consists of three functional modules, namely, multi-sensor data collection and pre-processing, environment perception, and 3D reconstruction. To integrate individual datasets collected from different sensors, the pre-processing function registers the sensed LiDAR (light detection and ranging) point clouds, video sequences, and motion information into a global terrain model after filtering redundant and noise data according to the redundancy removal principle. In the environment perception module, the registered discrete points are clustered into ground surface and individual objects by using a ground segmentation method and a connected component labeling algorithm. The estimated ground surface and non-ground objects indicate the terrain to be traversed and obstacles in the environment, thus creating driving awareness. The 3D reconstruction module calibrates the projection matrix between the mounted LiDAR and cameras to map the local point clouds onto the captured video images. Texture meshes and color particle models are used to reconstruct the ground surface and objects of the 3D terrain model, respectively. To accelerate the proposed system, we apply the GPU parallel computation method to implement the applied computer graphics and image processing algorithms in parallel.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2003년도 ISIS 2003
• /
• pp.284-287
• /
• 2003

In recent years, most of the researches on pattern recognition are for medical diagnosis or for characters recognition. In fact its applications are very wide. In this paper, the pattern recognition is employed by linguistic translation, i.e. the output of Pattern Recognition is translated into another language. In this paper, it focuses on several fields: (1) System overview-explicate the functions of each part individually; (2) Criteria on the system-discuss the difficulties in each part; (3) System implementation-discuss how to design the approaches for constructing the system. Furthermore, intelligent approaches are considered be use on the system in different parts. They are discussed in the late paper, and also we concentrate on user interface, which can make a serious of processes in order, and easy control-just only pressing a few buttons. It is a new and creative attempt in digital system.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제7권11호
• /
• pp.2754-2768
• /
• 2013

Video fingerprints generated from global features are usually vulnerable against general geometric transformations. In this paper, a novel video fingerprinting algorithm is proposed, in which a new spatio-temporal gradient is designed to represent the spatial and temporal information for each frame, and a new partition scheme, based on concentric circle and rings, is developed to resist the attacks efficiently. The centroids of spatio-temporal gradient orientations (CSTGO) within the circle and rings are then calculated to generate a robust fingerprint. Our experiments with different attacks have demonstrated that the proposed approach outperforms the state-of-the-art methods in terms of robustness and discrimination.

• 대한수학회지
• /
• 제54권5호
• /
• pp.1573-1594
• /
• 2017

This paper deals with the existence and energy estimates of solutions for a class of degenerate nonlocal problems involving sub-linear nonlinearities, while the nonlinear part of the problem admits some hypotheses on the behavior at origin or perturbation property. In particular, for a precise localization of the parameter, the existence of a non-zero solution is established requiring the sublinearity of nonlinear part at origin and infinity. We also consider the existence of solutions for our problem under algebraic conditions with the classical Ambrosetti-Rabinowitz. In what follows, by combining two algebraic conditions on the nonlinear term which guarantees the existence of two solutions as well as applying the mountain pass theorem given by Pucci and Serrin, we establish the existence of the third solution for our problem. Moreover, concrete examples of applications are provided.

• International Journal of Control, Automation, and Systems
• /
• 제6권4호
• /
• pp.559-570
• /
• 2008

This paper proposes an improved particle swarm optimization named PSO with Controllable Random Exploration Velocity (PSO-CREV) behaving an additional exploration behavior. Different from other improvements on PSO, the updating principle of PSO-CREV is constructed in terms of stochastic approximation diagram. Hence a stochastic velocity independent on cognitive and social components of PSO can be added to the updating principle, so that particles have strong exploration ability than those of conventional PSO. The conditions and main behaviors of PSO-CREV are described. Two properties in terms of "divergence before convergence" and "controllable exploration behavior" are presented, which promote the performance of PSO-CREV. An experimental method based on a complex test function is proposed by which the proper parameters of PSO-CREV used in practice are figured out, which guarantees the high exploration ability, as well as the convergence rate is concerned. The benchmarks and applications on FCRNN training verify the improvements brought by PSO-CREV.

• 한국수학교육학회지시리즈D:수학교육연구
• /
• 제16권3호
• /
• pp.155-176
• /
• 2012

It is an important issue to explore classroom environments which are conducive to developing students' mathematical performance. This study explores the effects of different classroom environments (solution-demand and corresponding-time setting) on mathematical performances. Fourteen and eighteen prospective teachers were required to prove a task under different conditions respectively: a) Cognitive demand of multiple-solution corresponding time of three hours, and b) Cognitive demand of a right solution corresponding time of 20 minutes. We used SOLO as the assessment tool for mathematical performance from quality perspective. Significant differences were found in the quantity and quality of mathematical performance. The regular environment focusing on speed and accuracy were found to be directly linked to low levels of performance. The findings above provide implications to the cognitive benefits of multiple-solution demand and corresponding time setting.