检索结果-南通市图书馆

Retrieval-and-alignment based large-scale indoor point cloud semantic segmentation

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2024年第4期67卷 164-180页

作者： Zongyi XU Xiaoshui HUANG Bo YUAN Yangfu WANG Qianni ZHANG Weisheng LI xinbo gao School of Computer Science and Technology Chongqing University of Posts and Telecommunications Shanghai Artificial Intelligence Laboratory School of Electronic Engineering and Computer Science Queen Mary University of London

Current methods for point cloud semantic segmentation depend on the extraction of descriptive features. However, unlike images, point clouds are irregular and often lack texture information, making it demanding to extract discriminative features. In addition, noise, outliers, and uneven point distribution are commonly present in point clouds, which further complicates the segmentation task. To address these problems, a novel architecture is proposed for direct and accurate large-scale point cloud segmentation based on point cloud retrieval and alignment. The proposed approach involves using a feature-based point cloud retrieval method for searching for reference point clouds with annotations from a dataset. In the following segmentation stage, an overlap-based point cloud registration method has been developed to align the target and reference point clouds. For accurate and robust alignment, an overlap region estimation module is trained to locate the optimal overlap region between two pieces of point clouds in a coarse-to-fine manner. In the detected overlap region, the global and local features of the points are extracted and combined for featuremetric registration to obtain accurate transformation parameters between the target and reference point clouds. After alignment, the annotated segmentation of the reference is transferred to the target point clouds to obtain accurate segmentation results. Extensive experiments are conducted to show that the developed method outperforms the state-of-the-art approaches in terms of both accuracy and robustness against noise and outliers.

关键词： point cloud semantic segmentation large-scale indoor point clouds point cloud alignment overlap estimation label transfer

一种单相Quasi-Z源逆变器二次谐波电流抑制策略

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

太阳能学报 2024年第6期45卷 349-358页

作者：刘欣高鑫波袁静华北电力大学(保定)电力工程系保定071003

为抑制单相qZSI直流侧二次谐波电流,提出一种基于虚拟阻抗的二次谐波电流抑制策略。该控制策略通过引入串联虚拟阻抗附加控制以提高单相qZSI在2f_(0)处的闭环输出阻抗,用以有效抑制直流侧的二次谐波电流。同时,综合考虑二次谐波电流抑... 详细信息

为抑制单相qZSI直流侧二次谐波电流,提出一种基于虚拟阻抗的二次谐波电流抑制策略。该控制策略通过引入串联虚拟阻抗附加控制以提高单相qZSI在2f_(0)处的闭环输出阻抗,用以有效抑制直流侧的二次谐波电流。同时,综合考虑二次谐波电流抑制要求以及保证系统稳定裕度等约束条件,提出该控制策略的闭环参数设计方法。最后通过数字仿真和基于RT-LAB的硬件在环仿真验证了所提二次谐波电流抑制策略以及其闭环参数设计方法的正确性。

关键词：光伏发电准Z源逆变器谐波分析虚拟阻抗

维普期刊数据库博看期刊

无模型自适应控制在大型12Cr2Mo1V钢热处理中的应用

在线全文

学校读者我要写书评

暂无评论

金属热处理 2024年第9期49卷 290-296页

作者：高新勃覃庆良青岛科技大学自动化与电子工程学院山东青岛266061

针对大口径厚壁12Cr2Mo1V钢的加热及焊后热处理,采用中频感应加热并选择无模型自适应控制(MFAC)算法代替传统的PID算法对其加热温度进行控制。首先分析工件的物理参数,计算出加热频率、线圈电流、电源配置等参数,再通过COMSOL软件将得... 详细信息

针对大口径厚壁12Cr2Mo1V钢的加热及焊后热处理,采用中频感应加热并选择无模型自适应控制(MFAC)算法代替传统的PID算法对其加热温度进行控制。首先分析工件的物理参数,计算出加热频率、线圈电流、电源配置等参数,再通过COMSOL软件将得到的各项参数带入,对现场条件进行磁-热耦合的仿真分析,验证计算数据的可行性;之后在SIMULINK仿真平台上搭建MFAC模型并与传统PID模型作比较,并将两种算法实际应用到现场观测加热效果。结果表明,将大型工件加热到目标温度700℃,采用MFAC算法加热的温度波动更小,稳定性更高,均匀性更好。

关键词：中频感应加热 12Cr2Mo1V钢 COMSOL有限元仿真无模型自适应控制算法

在线全文

学校读者我要写书评

暂无评论

基于多边形特征池化与融合的复杂文本检测

西安电子科技大学学报 2024年第3期51卷 113-123页

作者：张相南高新波田春娜西安电子科技大学电子工程学院陕西西安710071 重庆邮电大学计算机科学与技术学院图像认知重庆市重点实验室重庆400065

文本检测在图像理解中发挥着重要的作用。基于深度学习的文本检测是当前的主流算法,包括单阶段方法和双阶段方法两类,而且后者的检测精度往往高于前者。双阶段的检测方法通常包含感兴趣区域特征池化操作,为进一步的检测和识别任务提供... 详细信息

文本检测在图像理解中发挥着重要的作用。基于深度学习的文本检测是当前的主流算法,包括单阶段方法和双阶段方法两类,而且后者的检测精度往往高于前者。双阶段的检测方法通常包含感兴趣区域特征池化操作,为进一步的检测和识别任务提供特定维度的局部区域特征。然而对于弯曲文本等复杂文本区域来说,现有的基于矩形感兴趣区域的池化方法不再适用,而基于点特征替代区域特征的方法又损失了空间信息。针对该问题,提出了一种基于多边形特征池化和Transformer的复杂文本区域检测方法。首先,将复杂文本区域检测中感兴趣区域进行多边形特征池化,将池化操作的区域形状从矩形拓展到多边形并且不需要借助其他形状进行拟合,即可将多边形区域对应的特征池化为固定维度的特征序列,避免了拟合过程中出现误差。进而,将池化后的特征视为具有空间关系的序列,然后利用Transformer融合视觉特征之间的上下文关系,降低训练难度,提升检测精确度。在包含弯曲文本等复杂文本情况的ICDAR2015、MLT、Total Text和CTW1500数据集上的测试实验结果表明,提出的双阶段检测算法能更好地提取感兴趣区域特征,并取得了比现有方法更好的检测结果。

关键词：文本检测双阶段方法多边形特征池化 Transformer

Fast interactive segmentation algorithm of image sequences based on relative fuzzy connectedness

同方期刊数据库博看期刊评论

在线全文

学校读者我要写书评

暂无评论

Journal of Systems Engineering and Electronics 2005年第4期16卷 750-755页

作者： Tian Chunna gao xinbo School of Electronic Engineering Xidian Univ. Xi'an 710071 P. R. China

A fast interactive segmentation algorithm of image-sequences based on relative fuzzy connectedness is presented. In comparison with the original algorithm, the proposed one, with the same accuracy, accelerates the segmentation speed by three times for single image. Meanwhile, this fast segmentation algorithm is extended from single object to multiple objects and from single-image to image-sequences. Thus the segmentation of multiple objects from complex hackground and batch segmentation of image-sequences can be achieved. In addition, a post-processing scheme is incorporated in this algorithm, which extracts smooth edge with one-pixel-width for each segmented object. The experimental results illustrate that the proposed algorithm can obtain the object regions of interest from medical image or image-sequences as well as man-made images quickly and reliably with only a little interaction.

关键词： fuzzy connectedness interactive image segmentation image-sequences segmentation multiple objects segmentation fast algorithm.

Video Saliency Detection via Pairwise Interaction

在线全文

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2020年第3期29卷 427-436页

作者： QIU Wenliang gao xinbo HAN Bing Visual Information Processing Laboratory School of Electronic Engineering Xidian University

We propose a novel video saliency detection method based on pairwise interaction learning in this paper. Different from the traditional video saliency detection methods, which mostly combine spatial and temporal features, we adopt Least squares Conditional random field(LS-CRF) to capture the interaction information of regions within a frame or between video ***, dual graph-connection models are built on superpixels structure of each frame for training and testing, respectively. In order to extract the essential scene structure from video sequences, LS-CRF is introduced to learn the background texture, object components and the various relationships between foreground and background regions through the training set, and each region will be distributed an inferred saliency value in testing *** from the learned diverse relations among scene regions, the proposed approach achieves reliable results especially on multiple objects scenes or under highly complicated scenes. Further, we substitute weak saliency maps for pixel-wise annotations in training phase to verify the expansibility and practicability of the proposed method. Extensive quantitative and qualitative experiments on various video sequences demonstrate that the proposed algorithm outperforms conventional saliency detection algorithms.

关键词： Video saliency detection Conditional random fields Dual graph-connection model Pairwise interaction

Adaptive deinterlacing algorithm based on motion compensation

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 Systems Engineering and Electronics 2005年第4期16卷 931-936页

作者： gao xinbo Gu Junxia School of Electronic Engineering Xidian Univ. Xi'an 710071 P. R. China Dept. of Electronic Engineering Tsinghua Univ. Beijing 100084 P. R. China

An adaptive de-interlacing algorithm based on motion compensation is presented. It consists of the detection of motion blocks, the adaptive motion estimation with Kalman filtering, and the motion compensation for motion blocks and field repetition for static blocks. The detection of motion blocks can accurately identify the motion blocks by using successive 4-field images. The motion estimation module with Kalman filtering searches motion vectors only for motion blocks, and the search model is adaptive to motion velocity and acceleration. Two de-interlacing methods are adopted to satisfy the different requirements of motion blocks and static blocks. Compared with full search algorithm, the proposed algorithm greatly reduces the computational amount while keeping the performance approximately.

关键词： de-interlace adaptive motion estimation Kalman filter motion compensation.

Cooperative Jamming for Enhancing Security of Cognitive Radio Networks with Multiple Primary Users

在线全文

学校读者我要写书评

暂无评论

China Communications 2017年第7期14卷 93-107页

作者： Xiaofeng Feng xinbo gao Ru Zong State Key Laboratory of Integrated Services Networks School of Electronic Engineering Xidian University

In cognitive radio networks(CRNs), through recruiting secondary user(SU) as friendly jammer, the secrecy rate obtained by primary user(PU) can be improved. Previous work only considered a simple scenario with a single PU in their frameworks. In this paper, we will consider a more complicated scenario with multiple PUs and try to investigate the cooperative jamming between multiple PUs and a single SU. When there are multiple PUs in CRN, in order to obtain more spectrum for data transmission, SU will cooperate with multiple PUs at the same time. Considering that both PU and SU are rational and selfish individuals, the interaction between PUs and SU is formulated as a multi-leaders and single-follower Stackelberg game, wherein PU is the leader and SU is the follower. And the Stackelberg Equilibrium(SE) is considered as the final decisions accepted by all PUs and SU. Furthermore, we also prove that when a specific condition is satisfied, the existence of SE can be guaranteed. And a Gauss-Jacobi iterative algorithm is proposed to compute a SE. Finally, simulation results are given to verify the performance and demonstrate that both of the PUs' secrecy rate and the SU's transmission rate can be improved through cooperation.

关键词： cognitive radio networks cooperative jamming physical layer security spectrum sharing stackelberg game

在线全文

学校读者我要写书评

暂无评论

图像重定向质量评价的研究进展

中国图象图形学报 2024年第1期29卷 22-44页

作者：胡波谢国庆李雷达李静杨嘉琛路文高新波重庆邮电大学图像认知重庆市重点实验室重庆400065 广阳湾实验室重庆脑与智能科学中心重庆400064 西安电子科技大学人工智能学院西安710071 阿里巴巴集团北京100020 天津大学电气自动化与信息工程学院天津300072 西安电子科技大学电子工程学院西安710071

随着移动显示设备的普及,人们可以使用不同的终端设备来获取图像信息。为了适配不同尺寸的设备,图像重定向技术应运而生,并成为当前研究的热点。尽管该技术在近年来取得了长足进步,但没有算法能够保证在不降低视觉内容质量的前提下满足... 详细信息

随着移动显示设备的普及,人们可以使用不同的终端设备来获取图像信息。为了适配不同尺寸的设备,图像重定向技术应运而生,并成为当前研究的热点。尽管该技术在近年来取得了长足进步,但没有算法能够保证在不降低视觉内容质量的前提下满足多种显示设备的要求。换句话说,在图像重定向的过程中,不可避免地会引入失真,降低用户的视觉体验。因此,如何客观、准确地评价重定向图像质量对图像重定向算法的选择、优化及发展至关重要。当前,研究人员已针对图像重定向质量评价开展了一系列研究,并取得了一定的研究成果。但是缺乏对图像重定向质量评价的综述,因此本文对当前图像重定向质量评价的研究进展进行了较为全面的回顾和总结。首先简单介绍了图像重定向以及传统图像质量评价;然后分别介绍图像重定向质量评价的数据集和客观质量评价方法,将现有的客观质量评价分为基于传统特征相似性的方法和基于图像配准的方法两类,并分析了两类方法的优缺点;接着在3个数据集上对具有代表性的算法的性能进行了对比分析;最后,总结了当前图像重定向质量评价领域面临的问题和挑战,并指明了未来可能的发展方向。

关键词：图像质量评价(IQA) 图像重定向图像配准内容损失几何失真