检索结果-内蒙古大学图书馆

Research in Adaptive and Convergent Systems, RACS 2015

作者： Song, Enmin Pan, Ning Hung, Chih-Cheng Li, Xiang Jin, Lianghai School of Computer Science and Technology Huazhong University of Science and Technology Key Laboratory of Education Ministry for Image Processing and Intelligent Control Wuhan430074 China College of Biomedical Engineering South-Central University for Nationalities Wuhan430074 China Sino-US Intelligent Information Processing Lab Anyang Normal University Kennesaw State University MariettaGA30060 United States

ISBN: (纸本)9781450337380

In this study we modified the local binary pattern operator (LBP) to obtain the robust invariant texture patterns for image texture classification. The modified method will be able to calculate patterns which are invariant for translation, scaling, rotation and reflection. Therefore, the modified LBP is called R-LBP. Although many variation of LBP have been proposed, most of them cannot detect and recognize the patterns of reflection. Both clockwise and counter clockwise coding is used in the proposed R-LBP operator in order to derive a minimum code to representing the pattern. Experimental results show that the proposed method is effective in determining the invariant patterns for image texture classification. © 2015 ACM.

关键词： Local binary pattern

来源：评论

学校读者我要写书评

暂无评论

A novel path planning method based on path network

A novel path planning method based on path network

引用

MIPPR 2009 - Medical Imaging, Parallel processing of images, and Optimization Techniques: 6th International Symposium on Multispectral image processing and Pattern Recognition

作者： Li, Shidong Ding, Mingyue Cai, Chao Institute for Pattern Recognition and Artificial Intelligence National Lab. for Multi-spectral Information Processing Technologies Huazhong University of Science and Technology Wuhan 430074 China Hubei University for Nationalities Enshi Hubei 445000 China College of Life Science and Technology Image Processing and Intelligent Control Key Laboratory of Education Ministry of China Huazhong University of Science and Technology Wuhan 430074 China

ISBN: (纸本)9780819478085

Traditional path planning methods are too slow to meet the real-time requirement in practical applications. In order to solve this problem, an idea of path net was proposed in this paper. The path planning procedure is divided into two steps: network segment planning and segment assembling. The first step was done with Fast Marching Method, including port selection of segment and network segment planning. Second step, the A* searching method was chosen to select segments for assembling. Experiments demonstrated that our method can obtain an optimal route in a few of seconds after the start and goal are given while several minutes are needed for traditional methods. © 2009 Copyright SPIE - The International Society for Optical Engineering.

关键词： Motion planning

来源：评论

学校读者我要写书评

暂无评论

Path planning using FMM with direction and curvature constrained

Path planning using FMM with direction and curvature constra...

引用

MIPPR 2009 - Remote Sensing and GIS Data processing and Other Applications: 6th International Symposium on Multispectral image processing and Pattern Recognition

ISBN: (纸本)9780819478092

It is difficult to meet both direction and curvature constraints for traditional Fast Marching (FM) method in path planning. Based on adjusting the cost function in Eiknoal equation-the control equation for FM, a new model for computing the integrated cost function was presented in this paper. A relationship formula about curvature radius was obtained and three kinds of adjusting strategies were given;two of them were used to modify the route to meet with the requires of turning constraint in this paper. Experiments showed that the improved model can be used to plan the path with FMM for agent such as unmanned aerial vehicle (UAV) or robot, which is limited to pass through the scene matching areas. And our preliminary experiments demonstrated that the strategies are feasible and efficient to obtain path with certain curvature radius. The model can also be used to represent the problem such as an aircraft flying in a flow field. © 2009 Copyright SPIE - The International Society for Optical Engineering.

关键词： Motion planning

来源：评论

学校读者我要写书评

暂无评论

A parallel HEVC encoder scheme based on Multi-core platform

A parallel HEVC encoder scheme based on Multi-core platform

引用

2015 4th National Conference on Electrical,Electronics and Computer Engineering(NCEECE 2015)

作者： Shu Jun Hu Dong Education Ministry’s Key Lab of Broadband Wireless Communication and Sensor Network Technology Education Ministry’s Engineering Research Center of Ubiquitous Network and Health Service System Jiangsu Province’s Key Lab of Image Procession and Image Communications Nanjing University of Post and Telecommunications

ISBN: (纸本)9781510819085

In this paper, we propose a parallel HEVC encoder scheme based on multi-core platform, which provides maximized parallel scalability by exploiting two-level parallelism, namely, the frame level parallelism and the CTB level parallelism. Inspired by the intra-CTB row level parallelism of WPP in HEVC, we investigate the inter-frame CTB prediction dependency to its reference CTBs, and find the inter-CTB correlation. Using this inter-correlation, we divide a frame into CTB units and create CTB-row level coding threads when their corresponding reference CTBs are available. Each thread is bonded to a processing core, therefore, both intra- and inter-CTB rows can be encoded in parallel. Moreover, we introduce a priority scheduling mechanism to control the coding threads. Experiments on Tilera-Gx36 multi-core platform show that, compared with serial execution, the proposed method achieves 3.6 and 4.3 times speedup for 1080 P and 720 P video sequences, respectively.

关键词： HEVC multi-core platform parallel processing frame-level CTB-level

来源：评论

学校读者我要写书评

暂无评论

NiuEM: A Nested-iterative Unsupervised Learning Model for Single-particle Cryo-EM image processing

NiuEM: A Nested-iterative Unsupervised Learning Model for Si...

引用

2020 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2020

作者： Hu, Rui Cai, Jiaming Zheng, Wangjie Yang, Yang Shen, Hong-Bin Shanghai Jiao Tong University Key Lab. of Shanghai Education Commission for Intelligent Interaction and Cognitive Engineering Department of Computer Science and Engineering Shanghai200240 China Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University Key Laboratory of System Control and Information Processing Ministry of Education of China Shanghai200240 China Shanghai Jiao Tong University Department of Bioinformatics and Biostatistics Shanghai200240 China

ISBN: (纸本)9781728162157

Cryo-electron microscopy (cryo-EM) has become a mainstream technology for solving spatial structures of biomacromolecules, while the processing of cryo-EM images is a very challenging task. One of the great challenges is the high noise in the images. A common method is to cluster the images with close projecting angles to get mean images, which are used for 3D reconstruction. However, due to the extremely low signal-to-noise-ratio, common clustering methods often fail to obtain high-quality mean images, leading to poorly reconstructed structures. In this study, we present a new unsupervised learning framework, called NiuEM, to discriminate images captured from different angles and yield cluster-mean images. NiuEM first generates pseudo-labels and then exploits both contrastive loss and cross-entropy loss for training convolutional layers to learn feature representations. Moreover, the pseudo-labels are updated iteratively to enhance the reliability of labels. We assess the performance of NiuEM on four data sets via both visualized and quantitative experiments. Especially, two kinds of metrics are adopted to measure the performance, regarding the clustering quality and the resolution of reconstructed 3D models, respectively. The experimental results show that NiuEM achieves very competitive clustering accuracy in the comparison with the state-of-the-art image clustering methods. Moreover, the cluster mean images yielded by NiuEM lead to better initial 3D models compared with the mainstream reconstruction tools. © 2020 IEEE.

关键词： image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Physics-Environment Interaction Network for Dense Crowd Behavior Recognition

SSRN

引用

SSRN 2024年

作者： Yu, Jiaqi Zhou, Yanshan Pan, Renjie Lai, Pingrui Yang, Hua Institute of Image Communication and Network Engineering Shanghai Key Lab of Digital Media Processing and Transmission Shanghai Jiao Tong University Shanghai200240 China

The analysis of large-scale crowd behavior plays a crucial role in public safety. However, intelligent systems face three major challenges in analyzing dense crowd behavior: the severe occlusion between individuals, the variability in behavior patterns, and the complexity of behavioral evolution. To address these challenges, we propose the Physics-Environment Interaction Network (PEIN), which directly models the motion characteristics of a group with its physics attributes. Specifically, our method consists of two streams. The first stream is the physics-informed crowd property stream, which leverages on the similarity between dense crowd motion and fluid dynamics, using the Navier-Stokes (N-S) equation from fluid mechanics as the modeling framework to describe crowd motion. Considering the inherent relationship between the terms in the N-S equation and various crowd properties (collectiveness, conflict, uniformity and stability), we model these terms with operators and neural networks guided by these crowd properties, enabling the modeling of crowd motion characteristic without relying on the extraction of individual motion information. The second stream is the environment perception stream. Considering that the physics-informed crowd property stream mainly focuses on instantaneous information and that scenes with dense crowd behavior have variability, we introduce a 3D network to enhance the model's robustness. This stream can extract global spatiotemporal information from input video frames, providing a comprehensive perception of the surrounding crowd environment. Since the two streams process different data sources, we design a dual cross-attention mechanism to to enable between features from different modalities, resulting in a joint learnable representation for the final crowd behavior recognition. By incorporating physical laws as constraints, we design a physics-informed loss function combined with a crowd behavior loss function to optimize the model. Consi

关键词： Navier Stokes equations

来源：评论

学校读者我要写书评

暂无评论

Improving Mandarin End-to-End Speech Synthesis by Self-Attention and Learnable Gaussian Bias

Improving Mandarin End-to-End Speech Synthesis by Self-Atten...

引用

IEEE Workshop on Automatic Speech Recognition and Understanding

作者： Fengyu Yang Shan Yang Pengcheng Zhu Pengju Yan Lei Xie Shaanxi Provincial Key Laboratory of Speech and Image Information Processing School of Computer Science Northwestern Polytechnical University Xian China Tongdun AI Lab

ISBN: (数字)9781728103068

ISBN: (纸本)9781728103075

Compared to conventional speech synthesis, end-to-end speech synthesis has achieved much better naturalness with more simplified system building pipeline. End-to-end framework can generate natural speech directly from characters for English. But for other languages like Chinese, recent studies have indicated that extra engineering features are still needed for model robustness and naturalness, e.g, word boundaries and prosody boundaries, which makes the front-end pipeline as complicated as the traditional approach. To maintain the naturalness of generated speech and discard language-specific expertise as much as possible, in Mandarin TTS, we introduce a novel self-attention based encoder with learnable Gaussian bias in Tacotron. We evaluate different systems with and without complex prosody information and results show that the proposed approach has the ability to generate stable and natural speech with minimum language-dependent front-end modules.

关键词： Decoding Speech synthesis Pipelines Robustness Spectrogram Natural languages

来源：评论

学校读者我要写书评

暂无评论

An image fusion method of multi-scale image based on HMM and genetic algorithms

引用

Journal of Computational Information Systems 2015年第12期11卷 4367-4374页

作者： Jiang, Zetao Wu, Donghui The Institute of Computer Science and Engineering Guilin University of Electronic Technology Guilin China The Key Lab of Image Processing and Pattern Recognition of Jiangxi Province Nanchang Hangkong University Nanchang China

At present, in the field of pixel-level image fusion, researchers tend to treat each pixel independently, which destroys the relationship between the images to be fused. In view of this defects, this paper aims to propose a method for multi-scale image fusion based on Genetic Algorithm (GA) under the guidance of Hidden Markov Model (HMM). This fusion method is set on the following principles. First of all, the wavelet transformation (WT) of the two images to be fused will be conducted. And then establish HMM for the fusion coefficients from high to low in descending hierarchical order. Finally, guided by HMM, the author applies GA to searching hierarchically for the optimal fusion coefficient. With this method, the best fusion effect will be achieved. This method take good advantage of the correlation information between the images to be fused so as to improve the fidelity and accuracy of the fusion. Given this, the author makes an empirical analysis on this fusion method, which proves to be feasible and effective. ©, 2015, Binary Information Press. All right reserved.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Learning Hierarchical Representations for Expressive Speaking Style in End-to-End Speech Synthesis

Learning Hierarchical Representations for Expressive Speakin...

引用

IEEE Workshop on Automatic Speech Recognition and Understanding

作者： Xiaochun An Yuxuan Wang Shan Yang Zejun Ma Lei Xie Shaanxi Provincial Key Laboratory of Speech and Image Information Processing School of Computer Science Northwestern Polytechnical University Xi'an China ByteDance AI Lab

ISBN: (数字)9781728103068

ISBN: (纸本)9781728103075

Although Global Style Tokens (GSTs) are a recently-proposed method to uncover expressive factors of variation in speaking style, they are a mixture of style attributes without explicitly considering the factorization of multiple-level speaking styles. In this work, we introduce a hierarchical GST architecture with residuals to Tacotron, which learns multiple-level disentangled representations to model and control different style granularities in synthesized speech. We make hierarchical evaluations conditioned on individual tokens from different GST layers. As the number of layers increases, we tend to observe a coarse to fine style decomposition. For example, the first GST layer learns a good representation of speaker IDs while finer speaking style or emotion variations can be found in higher-level layers. Meanwhile, the proposed model shows good performance of style transfer.

关键词： Speech Spectrogram Speech synthesis Data models Training data Predictive models Training

来源：评论

学校读者我要写书评

暂无评论

A new technique for high-fidelity cutting technology for hydrate samples

引用

Journal of Zhejiang University-Science A(Applied Physics & Engineering) 2022年第1期23卷 40-54页

作者： Hai ZHU Jia-wang CHEN Zi-qiang REN Pei-hao ZHANG Qiao-ling GAO Xiao-ling LE Chun-ying XU Kai HE Peng ZHOU Feng GAO Yu-ping FANG Institute of Ocean Engineering and Technology Ocean CollegeZhejiang UniversityZhoushan316021China Southern Marine Science and Engineering Guangdong Laboratory(Guangzhou) Guangzhou511458China Key Lab of Digital Signal and Image Processing of Guangdong Province School of EngineeringShantou UniversityShantou515063China

Designing a high-fidelity cutting device is one of the difficulties in hydrate samples pressure-holding *** to the limitations of the existing mechanical system,there is much damage to the cut surface of hydrate samples,with many chips produced,which seriously affects the quality of *** this paper,a new cutting device utilizes two servo motors to achieve a high degree of *** the Archimedes spiral,it achieves low disturbance of the cut surface and provides accurate control of the *** addition,due to the operation of the sample long-stroke push unit,cutting hydrate samples of any length with almost no chips within a short cutting time can be *** and sea tests have achieved all design requirements of the equipment and strongly demonstrate its benefit and *** is concluded that this new high-fidelity cutting technology is practically *** physical state of the hydrate can be maintained to the greatest extent,and thus the new equipment provides significant support for the exploration and development of hydrate resources.

关键词： High-fidelity Cutting technology Hydrate samples Automation

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：