检索结果-内蒙古大学图书馆

Saliency region and motion characteristics combined video quality assessment

Journal of Information Hiding and Multimedia Signal processing 2016年第4期7卷 881-888页

作者： Zhou, Ying Liang, Yongsheng Liu, Wei He, Cheng Zhu, Yue Shenzhen Key Laboratory of Visual Media Processing and Transmission Shenzhen Institute of Information Technology Shenzhen518172 China CNNC Lanzhou Uranium EnrichmentCo. Ltd Lanzhou730060 China

In order to accurately evaluate the video quality and make it consistent with the subjective evaluation result, a saliency region and motion characteristics combined video quality assessment is proposed in this paper. This method is based on traditional SSIM and improves its assessment performance. According to the relationship between human visual attention and video’s region of saliency, frame saliency is weighted by spatial saliency and temporal motion characteristics, in which the spatial saliency is extracted by spectrum analysis and the temporal one got by motion visual attention model. This saliency is weighted by SSIM index for single frame quality assessment. Finally, based on HVS characteristics and motion characteristics, the entire video quality SMW-SSIM can be got by weighting single-frame quality. Experiment results prove that this method is faster, more convenient and accurate than other SSIM-based assessments. © 2016.

关键词： Spectrum analysis

来源：评论

学校读者我要写书评

暂无评论

An improved non-linear ARED algorithm based on queue length 4

An improved non-linear ARED algorithm based on queue length

引用

4th International Conference on Multimedia Technology, ICMT 2015

作者： Zhou, Ying Zhang, Jihong Liu, Wei Liang, Yongsheng Hu, Tao Ren, Xianyi Chen, Xiaoping College of Information Engineering Shenzhen University Shenzhen China Shenzhen Key Laboratory of Visual Media Processing and Transmission Shenzhen Institute of Information Technology Shenzhen China SZMOBI Shenzhen Company Shenzhen China

ISBN: (纸本)9781138027947

Active queue management algorithm can effectively control the network congestion, among which ARED and its improved algorithm has been widely used in recent years. In order to timely response to unexpected network congestion and match the non-linear relationship between packet loss rate and queue length, this paper improved ARED algorithm and proposed a queue length based nonlinear ARED algorithm. This algorithm was analyzed and compared with others in NS2 simulation environment. Experimental results show that compared to other ARED algorithm, this algorithm can guarantee the stability of the queue, respond to unexpected network congestion timely, reduce drop rate and improve network throughput. © 2015 Taylor & Francis Group.

关键词： Queueing theory

来源：评论

学校读者我要写书评

暂无评论

Shape detector for generic ball detection

Shape detector for generic ball detection

引用

IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB)

作者： Yuncheng Li Yukun Zhu Rui Zhang Jun Zhou Shanghai Key Lab of Digital Media Processing and Transmission Shanghai Jiao Tong University Shanghai China

This paper describes a novel approach for generating object proposals for ball detection. Our method, called shape detector, captures the possible contours of balls and then transfers them into proposal bounding boxes which may contain the target object. These proposal bounding boxes can be further used in class-specific object detection task. Our experiment results on part of ILSVRC dataset show that shape detector can achieve 71.13% recall and a mean average best overlap of 0.648 using less than 300 proposals. It also shows strong performance in object detection, in which we get a mean average precision of 34.33%.

关键词： Proposals Object detection Shape Feature extraction Detectors Image edge detection Sports equipment

来源：评论

学校读者我要写书评

暂无评论

Video motion characteristics based spatial-temporal salient region extraction method

引用

Journal of Information Hiding and Multimedia Signal processing 2015年第2期6卷 225-233页

作者： Zhou, Ying Zhang, Ji-Hong Liang, Yong-Sheng Liu, Wei College of Information Engineering Shenzhen University Shenzhen518060 China Shenzhen Key Laboratory of Visual Media Processing and Transmission Shenzhen Institute of Information Technology Shenzhen518172 China

The human eyes only observe the salient regions of the video. According to this, the motion characteristics based spatial-temporal salient region extraction method was proposed. Spatial saliency map was extracted by analyzing the log spectrum of each frame in the frequency domain. Temporal saliency map was obtained by global motion estimation, block matching and then Gaussian filter. According to the human visual characteristics and the subjective perception of different motion characteristics, the final saliency region was fused dynamically by spatial and temporal saliency map. The experiment was analyzed from both subjective and objective indicators. visual observation and quantitative indicators show that the method proposed in this paper can reflect the human visual attention area more accurately than other classical extraction methods. © 2015.

关键词： Extraction

来源：评论

学校读者我要写书评

暂无评论

A combined motion characteristics and video saliency map extraction method

A combined motion characteristics and video saliency map ext...

引用

2014 International Conference on Computer Science and Systems Engineering（CSSE2014）

作者： Y.Zhou J.H.Zhang College of Information Engineering Shenzhen University Shenzhen Key Laboratory of Visual Media Processing and Transmission Shenzhen Institute of Information Technology

ISBN: (纸本)9781784660543

A combined motion characteristics and video saliency map extraction method is proposed according to the human visual system. A spatial saliency map was extracted by analyzing the log spectrum of each frame in the frequency domain. Temporal saliency was obtained by global motion estimation and block matching. According to the human visual characteristics and the subjective perception of different motion characteristics, the region of saliency was fused dynamically by the spatial and temporal saliency map. The experiment was analyzed from both subjective and objective indicators. visual observations and quantitative indicators show that the method proposed in this paper can reflect the human visual attention area more accurately than other classical extraction methods.

关键词： region of saliency visual attention model temporal saliency spatial saliency motion characteristics

来源：评论

学校读者我要写书评

暂无评论

Region based saliency detection by learning background information

Region based saliency detection by learning background infor...

引用

IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB)

作者： Di Chen Lianyang Ma Rui Zhang Shanghai Key Lab. of Digital Media Processing and Transmission Shanghai Jiao Tong University Shanghai China

visual saliency detection has become a challenging area in computer vision. In this paper, we propose a novel region based saliency detection model which considers background priors. The proposed method consists of two successive steps - region weighting and contrast computing. In the step of region weighting, we calculate the region weight for each region by region-level image feature and a log-linear prediction model. In the step of contrast computing, we propose a modified contrast computing algorithm by exploiting the advantage of region weights for bottom-up saliency detection. The experimental results on two datasets prove that our method effectively improves the performance on visual saliency detection.

关键词： Image color analysis Feature extraction Mathematical model Equations Computational modeling Image segmentation Predictive models

来源：评论

学校读者我要写书评

暂无评论

An improved algorithm for extraction of exact boundaries and boundaries inclusion relationship

An improved algorithm for extraction of exact boundaries and...

引用

19th International Computing and Combinatorics Conference, COCOON 2013

作者： Hu, Tao Ren, Xianyi Zhang, Jihong Institute of Information Technology Shenzhen Institute of Information Technology 518171 Shenzhen China Shenzhen Key Laboratory of Visual Media Processing and Transmission 518171 Shenzhen China

ISBN: (纸本)9783642387678

Boundary extraction algorithm proposed by Capson can get the same or even better performance as the commercial software such as VisionPro andHalcon. Unfortunately, the algorithm cannot extract the inclusion relationship between boundaries, which greatly reduce its attractiveness. Two improvements is proposed to improve the original algorithm in this paper, one is to improve the precision of boundaries by splitting points and merging points, another one is to design new data-structure and rules to implement acquiring deep inclusion relationship between external and internal boundaries. Experimental results show that the improvements increase a little consuming time but make the original algorithm more attractive. © 2013 Springer-Verlag Berlin Heidelberg.

关键词： Merging

来源：评论

学校读者我要写书评

暂无评论

Network flow based collective behavior analysis

引用

2013 International Workshop on Behavior and Social Informatics and Computing, BSIC 2013

作者： Hu, Yuting Xie, Rong Zhang, Wenjun Institute of Image Communication and Network Engineering Shanghai Key Lab. of Digital Media Processing and Transmission Shanghai Jiao Tong University Shanghai China

ISBN: (纸本)9783319040479

With the large-scale activities increasing gradually, the intelligent video surveillance system becomes more and more popular and important. The trajectory identification and behavior analysis are very important techniques for the intelligent video surveillance system. This paper focuses on the trajectory identification and behavior analysis framework for video surveillance system. The framework is implemented on footbridge video and queuing video of Shanghai World Expo 2010 video surveillance system. The experimental results show the efficiency of our proposed framework. © Springer International Publishing Switzerland 2013.

关键词： Integer programming

来源：评论

学校读者我要写书评

暂无评论

A novel architecture of multi-gpu computing card

引用

Telkomnika - Indonesian Journal of Electrical Engineering 2013年第8期11卷 4379-4384页

作者： Guo, Sen Chen, Sanfeng Liang, YongSheng ShenZhen Institute of Information Technology ShenZhen China Shenzhen Key Laboratory of Visual Media Processing and Transmission Shenzhen Guangdong 518172 China

AThe data transmission between GPUS in the existing multi_GPU computing card is often through PCIE which is in relative low speed, so the PCIE has become bottleneck of Overall performance. A novel architecture of multi_GPU computing card have been proposed in this paper: A multi-channel memory which have multiple interfaces is added, including one common interface shared by different GPUs, which is connected with a FPGA arbitration circuit and several other interfaces connected with dedicated GPUs frame buffer independently, and this multi-channel memory is called "global shared memory". The result of a simulation of accelerating computer tomography algebraic reconstruction on multi-GPU demonstrates effectiveness of this approach. © 2013 Universitas Ahmad Dahlan.

关键词： Program processors

来源：评论

学校读者我要写书评

暂无评论

IMAGE RESTORATION VIA EFFICIENT GAUSSIAN MIXTURE MODEL LEARNING

IMAGE RESTORATION VIA EFFICIENT GAUSSIAN MIXTURE MODEL LEARN...

引用

IEEE International Conference on Image processing

作者： Jianzhou Feng Li Song Xiaoming Huo Xiaokang Yang Wenjun Zhang Shanghai Digital Media Processing and Transmission Key Lab Shanghai Jiaotong University School of Industrial and Systems Engineering Georgia Institute of Technology

ISBN: (纸本)9781479923427

Expected Patch Log Likelihood (EPLL) framework using Gaussian Mixture Model (GMM) prior for image restoration was recently proposed with its performance comparable to the state-of-the-art algorithms. However, EPLL uses generic prior trained from offline image patches, which may not correctly represent statistics of the current image patches. In this paper, we extend the EPLL framework to an adaptive one, named A-EPLL, which not only concerns the likelihood of restored patches, but also trains the GMM to fit for the degraded image. To efficiently estimate GMM parameters in A-EPLL framework, we improve a recent Expectation- Maximization (EM) algorithm by exploiting specific structures of GMM from image patches, like Gaussian Scale Models. Experiment results show that A-EPLL outperforms the original EPLL significantly on several image restoration problems, like inpainting, denoising and deblurring.

关键词： Image restoration Expected patch log likelihood Gaussian mixture model image restoration technique Gm(m) Mixture models Gauss Current Pick image patch deblurring

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：