检索结果-南通市图书馆

面向微博短文本的社交与概念化语义扩展搜索方法

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

计算机研究与发展 2018年第8期55卷 1641-1652页

作者：崔婉秋杜军平寇菲菲李志坚 lee jangmyung 智能通信软件与多媒体北京市重点实验室(北京邮电大学) 北京100876 釜山国立大学电子工程系韩国釜山46241

充分挖掘微博短文本的语义以实现精准搜索是一项重要任务.由于微博文本内容具有稀疏性和语义局限性的特点,使得仅通过分析字面语义来进行短文本理解和相似性匹配的传统搜索方法受到了一定的限制.因此提出了一种社交与概念化语义结合的... 详细信息

充分挖掘微博短文本的语义以实现精准搜索是一项重要任务.由于微博文本内容具有稀疏性和语义局限性的特点,使得仅通过分析字面语义来进行短文本理解和相似性匹配的传统搜索方法受到了一定的限制.因此提出了一种社交与概念化语义结合的扩展搜索方法,通过挖掘社交网络独特的社交属性如#标签#、"@"和链接信息URL,对微博短文本实现进一步的社交语义扩展.该方法将文本字面分析获取的概念词语和社交关系中潜在的关联标签信息相结合,对短文本进行2种角度下的语义特征表示,实现了基于微博短文本语义充分理解的精准搜索.在微博数据集上的对比实验表明,与已有的扩展搜索方法相比所提方法能捕捉更多的语义特征,微博搜索的性能也得到了显著的提升.

关键词：微博短文本社交与概念化语义扩展搜索概念词语关联标签

在线全文

学校读者我要写书评

暂无评论

基于卷积神经网络的微博话题内容搜索方法

计算机科学与探索 2019年第5期13卷 753-764页

作者：周南杜军平姚旭梁美玉薛哲 lee jangmyung 北京邮电大学智能通信软件与多媒体北京市重点实验室计算机学院北京100876 釜山国立大学电子工程系韩国釜山46241

针对由于微博文本的数据特性造成的传统信息搜索方法无法直接实现微博话题内容搜索的问题,提出了一种基于卷积神经网络的微博话题内容搜索方法,对微博安全话题内容进行搜索和匹配排序。该方法包括基于深度卷积神经网络的微博内容筛选和... 详细信息

针对由于微博文本的数据特性造成的传统信息搜索方法无法直接实现微博话题内容搜索的问题,提出了一种基于卷积神经网络的微博话题内容搜索方法,对微博安全话题内容进行搜索和匹配排序。该方法包括基于深度卷积神经网络的微博内容筛选和微博内容匹配两部分。微博内容筛选依据深度卷积特征表示进行微博内容筛选,微博内容匹配通过卷积特征非线性变换对筛选结果进行匹配排序。微博内容筛选和微博内容匹配对国民安全话题相关的微博文本内容局部特征进行处理,对筛选结果进行相似度计算从而实现相似度匹配。实验结果表明该方法在微博搜索性能上优于现有同类方法,并验证了所提出方法针对安全话题的微博文本内容搜索的有效性。

关键词：微博搜索深度卷积神经网络深度学习搜索排序信息搜索

Robust visual tracking based on scale invariance and deep learning

在线全文

学校读者我要写书评

暂无评论

Frontiers of Computer Science 2017年第2期11卷 230-242页

作者： Nan REN Junping DU Suguo ZHU Linghui LI Dan FAN jangmyung lee Beijing Key Laboratory of Intelligent Telecommunication Software and Multimedia School of Computer Science Beijing University of Posts and Telecommunications Beijing 100876 China Department of Electronics Engineering Pusan National University Busan 46241 Korea

Visual tracking is a popular research area in com- puter vision, which is very difficult to actualize because of challenges such as changes in scale and illumination, rota- tion, fast motion, and occlusion. Consequently, the focus in this research area is to make tracking algorithms adapt to these changes, so as to implement stable and accurate vi- sual tracking. This paper proposes a visual tracking algorithm that integrates the scale invariance of SURF feature with deep learning to enhance the tracking robustness when the size of the object to be tracked changes significantly. Particle filter is used for motion estimation. The co^fidence of each parti- cle is computed via a deep neural network, and the result of particle filter is verified and corrected by mean shift because of its computational efficiency and insensitivity to external interference. Both qualitative and quantitative evaluations on challenging benchmark sequences demonstrate that the pro- posed tracking algorithm performs favorably against several state-of-the-art methods throughout the challenging factors in visual tracking, especially for scale variation.

关键词： visual tracking SURF mean shift particle filter neural network

Spatio-Temporal Adaptive Super-Resolution Reconstruction Model Based on Zemike Moment for Spatial Video Sequences

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

China Communications 2012年第12期9卷 93-107页

作者： Liang Meiyu Du Junping jangmyung lee Liu Honggang Zhang Yun Beijing Key Laboratory of Intelligent Telecommunication Software and Multimedia School of Computer Science Beijing University of Posts and Telecommunications Beijing 100876 P. R. China Department of Electronics Engineering Pusan Nation University Busan Korea

Video Super-Resolution(SR) reconstruction produces video sequences with High Resolution(HR) via the fusion of several Low-Resolution(LR) video frames.Traditional methods rely on the accurate estimation of subpixel motion,which constrains their applicability to video sequences with relatively simple motions such as global translation.We propose an efficient iterative spatio-temporal adaptive SR reconstruction model based on Zernike Moment(ZM),which is effective for spatial video sequences with arbitrary motion.The model uses region correlation judgment and self-adaptive threshold strategies to improve the effect and time efficiency of the ZM-based SR method.This leads to better mining of non-local selfsimilarity and local structural regularity,and is robust to noise and rotation.An efficient iterative curvature-based interpolation scheme is introduced to obtain the initial HR estimation of each LR video frame.Experimental results both on spatial and standard video sequences demonstrate that the proposed method outperforms existing methods in terms of both subjective visual and objective quantitative evaluations,and greatly improves the time efficiency.

关键词：自适应阈值视频序列重建模型超分辨率空间时空运动估计时间效率

Balance control of a 12-DOF mobile manipulator based on two-wheel inverted pendulum robot

在线全文

学校读者我要写书评

暂无评论

Journal of Measurement Science and Instrumentation 2013年第1期4卷 52-57页

作者： Gang Wang Seunghwan Choi jangmyung lee Dept.of Electrical Engineering Pusan UniversityBusan 609-735Korea

Humanoid mobile manipulator which is based on two-wheel inverted pendulum robot has been studied.Balance control is a key problem for this kind of centroid-variable robot.Due to the principle of two wheel inverted pendulum,a timely angle compensation is necessary to make the system keep balance when the centroid changes.In this paper,a method based on coordinate transformation is introduced to get the compensatory angle and a 12-DOF mobile manipulator is also used to check the method.Simulation and experimental results show the effectiveness of the method.

关键词： angle compensation inverted pendulum variable centroid humanoid

在线全文

学校读者我要写书评

暂无评论

基于时空特征和神经网络的视频超分辨率算法

北京邮电大学学报 2016年第4期39卷 1-6页

作者：李玲慧杜军平梁美玉任楠 jangmyung lee 北京邮电大学计算机学院韩国釜山国立大学电子工程系

提出了一种基于时空特征和神经网络的视频超分辨率重建算法,实现了视频视觉分辨率质量和细节清晰度的提升.该算法综合考虑了外部图像块之间的关联映射关系和内部图像块间的相似性,利用深度卷积神经网络学习得到的拟合系数快速地重建视... 详细信息

提出了一种基于时空特征和神经网络的视频超分辨率重建算法,实现了视频视觉分辨率质量和细节清晰度的提升.该算法综合考虑了外部图像块之间的关联映射关系和内部图像块间的相似性,利用深度卷积神经网络学习得到的拟合系数快速地重建视频细节.采用时空非局部特征相似性优化重建结果,将相邻视频帧间的非局部互补冗余信息融入学习视频帧结果中,解决了误匹配等问题,进一步提升了超分辨率性能.实验结果表明,所提方法在客观评价指标和主观视觉效果上均取得了较好的重建效果.

关键词：超分辨率重建深度卷积神经网络时空特征非局部相似性

Balancing control of unicycle robot by using sliding control

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Journal of Measurement Science and Instrumentation 2012年第4期3卷 383-388页

作者： Inwoo Han Jaewon An Hyunwoo Kim jangmyung lee Dept.of Electrical Engineering Pusan National University

This paper discusses about balancing control of unicycle robot.Unicycle robot consists of pitch which acts like inverted pendulum and roll which acts like reaction wheel pendulum.The robot which does not have actuator located in yaw axis is made to derive the simple dynamics.Lagrange equation is applied to deriving dynamic equations.Obtained dynamic equations are used to design the sliding mode control.State variables of the designed control are pitch angle and roll angle.Sliding mode control has chattering problem,which is eliminated by using the sigmoid function as switching function.Finally the control performance and eliminated chattering problem is verified by simulation.

关键词：自动化系统数据处理数据收集自动分类

Outdoor position estimation based on a combination system of GPS-INS by using UPF

在线全文

学校读者我要写书评

暂无评论

Journal of Measurement Science and Instrumentation 2013年第1期4卷 47-51页

作者： Yunki Kim Jaehyun Park jangmyung lee Dept.of Electrical Engineering Pusan National UniversityPusan 609-735Korea

This paper proposes a technique that global positioning system(GPS)combines inertial navigation system(INS)by using unscented particle filter(UPF)to estimate the exact outdoor position.This system can make up for the weak point on position estimation by the merits of GPS and INS.In general,extended Kalman filter(EKF)has been widely used in order to combine GPS with INS.However,UPF can get the position more accurately and correctly than EKF when it is applied to real-system included non-linear,irregular distribution errors.In this paper,the accuracy of UPF is proved through the simulation experiment,using the virtual-data needed for the test.

关键词： global positioning system(GPS) unscented particle filter(UPF) navigation inertial navigation system(INS) strapdown inertial navigation system(SDINS)

Object-tracking robot using ultrasonic sensor and servo motor

在线全文

学校读者我要写书评

暂无评论

Journal of Measurement Science and Instrumentation 2012年第4期3卷 379-382页

作者： Kyounghwan Kim Hyunseop Lim Yoseop Hwang jangmyung lee Dept.of Electrical Engineering Pusan National University

This paper proposes a method that rotation angle of servo motor and distance values of ultrasonic sensor are used for tracking an object in real-time while the robot keeps regular distance.Object detection distance widens by using ultrasonic sensors and object recognition,and movement of robot is controlled by angle of servo motor and distance of ultrasonic sensors.Not adopting the existing tracking methods:camera,laser-infrared(LRF)and many ultrasonic sensors,the proposed method proves that it is possible to track object using ultrasonic sensor and servo motor.Trajectory of robot is represented and analysed according to movement of object in limited conditions.

关键词：自动化系统数据处理数据收集自动分类