检索结果-南通市图书馆

Salt and pepper noise removal in surveillance video based on low-rank matrix recovery

在线全文

学校读者我要写书评

暂无评论

Computational Visual media 2015年第1期1卷 59-68页

作者： Yongxia Zhang Yi Liu Xuemei Li Caiming Zhang School of Computer Science and Technology Shandong University Shandong University of Finance and Economics Shandong Provincial Key Laboratory of Digital Media Technology

This paper proposes a new algorithm based on low-rank matrix recovery to remove salt &pepper noise from surveillance video. Unlike single image denoising techniques, noise removal from video sequences aims to utilize both temporal and spatial information. By grouping neighboring frames based on similarities of the whole images in the temporal domain, we formulate the problem of removing salt &pepper noise from a video tracking sequence as a lowrank matrix recovery problem. The resulting nuclear norm and L1-norm related minimization problems can be efficiently solved by many recently developed methods. To determine the low-rank matrix, we use an averaging method based on other similar images. Our method can not only remove noise but also preserve edges and details. The performance of our proposed approach compares favorably to that of existing algorithms and gives better PSNR and SSIM results.

关键词： multimedia computing noise cancellation signal denoising sparse matrices video signal processing video surveillance

Science Letters:Human-centered modeling for style-based adaptive games

在线全文

学校读者我要写书评

暂无评论

Journal of Zhejiang University-Science A(Applied Physics & Engineering) 2009年第4期10卷 530-534页

作者： Chee-onn WONG Jongin KIM Eunjung HAN Keechul JUNG Human-Centered Interfaces Lab Department of Digital Media Soongsil UniversitySeoul 156743 Korea

This letter proposes a categorization matrix to analyze the playing style of a computer game player for a shooting game genre. Our aim is to use human-centered modeling as a strategy for adaptive games based on entertainment measure to evaluate the playing experience. We utilized a self-organizing map (SOM) to cluster the player's style with the data obtained while playing the game. We further argued that style-based adaptation contributes to higher enjoyment, and this is reflected in our experiment using a supervised multilayered perceptron (MLP) network.

关键词： Player modeling Categorization matrix Adaptive games Human-centered modeling Data clustering

Enlarging Image by Constrained Least Square Approach with Shape Preserving

在线全文

学校读者我要写书评

暂无评论

Journal of Computer Science & Technology 2015年第3期30卷 489-498页

作者：张帆张新秦学英张彩明 School of Computer Science and Technology Shandong University Jinan 250101 China Shandong Provincial Key laboratory of digital media Technology Shandong University of Finance and Economics Jinan 250014 China

From a visual point of view, the shape of an image is mainly determined by the edges. Conventional polynomial interpolation of image enlarging methods would produce blurred edges, while edge-directed interpolation based methods would cause distortion in the non-edge areas. A new method for image enlarging is presented. The image is enlarged in two steps. In the first step, a fitting surface is constructed to interpolate the image data. To remove the zigzagging artifact for each pixel, a fitting patch is constructed using edge information as constraints. The combination of all the patches forms the fitting surface which has the shape suggested by image data. Each point on the fitting surface can be regarded as a sampling point taken from a unit square domain, which means that when the fitting surface is used to enlarge the image, each sampling domain of the enlarged pixels is also a unit square, causing the enlarged image to lose some details. To make the enlarged image keep the details as many as possible, the sampling domain of the enlarged pixels should be less than a unit square. Then, in the second step, using the points taken from the fitting surface, new pixels are computed using constrained optimization technique to form the enlarged image, and the size of the sampling domain of the enlarged pixels is inversely proportional to the size of the enlarged image. The image enlarged by the new method has a quadratic polynomial precision. Comparison results show that the new method produces resized image with better quality.

关键词： image resizing fitting quadratic polynomial shape preserving edge information

Depth estimation using an improved stereo network

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

Frontiers of Information Technology & Electronic Engineering 2022年第5期23卷 777-789页

作者： Wanpeng XU Ling ZOU Lingda WU Yue QI Zhaoyong QIAN Science and Technology on Complex Electronic System Simulation laboratory Space Engineering UniversityBeijing 101416China Peng Cheng laboratory Shenzhen 518055China digital media School Beijing Film AcademyBeijing 100088China

Self-supervised depth estimation approaches present excellent results that are comparable to those of the fully supervised approaches,by employing view synthesis between the target and reference images in the training ***,which serves as a backbone network,has some structural deficiencies when applied to downstream fields,because its original purpose was to cope with classification *** low-texture area also deteriorates the *** address these problems,we propose a set of improvements that lead to superior ***,we boost the information flow in the network and improve the ability to learn spatial structures by improving the network ***,we use a binary mask to remove the pixels in low-texture areas between the target and reference images to more accurately reconstruct the ***,we input the target and reference images randomly to expand the dataset and pre-train it on ImageNet,so that the model obtains a favorable general feature *** demonstrate state-of-the-art performance on an Eigen split of the KITTI driving dataset using stereo pairs.

关键词： Monocular depth estimation Self-supervised Image reconstruction

First-Principles Study of Magnetic Properties of TM_(13) and TM_(13)@Au_(32) Clusters(TM = Mn, Co)

在线全文

学校读者我要写书评

暂无评论

Chinese Physics Letters 2018年第10期35卷 22-26页

作者： Yi-Bo Li Li-Jin Zeng Chun-Xi＆ng Zhao Chun-Yao Niu International laboratory for Quantum Fhnctional Materials of Henan School of Physics and Engineering Zhengzhou UniversityZhengzhou 450001 School of digital media Shenzhen Institute of Information TechnologyShenzhen 518172

The structural and magnetic properties of TM13and TM_（13）@Au32clusters（TM=Mn,Co）are studied by firstprinciples *** find that the Au32cluster can tune not only the magnetic moment but also the magnetic coupling properties between the TM atoms of the TM *** Au32cluster can increase the net magnetic moment of Mn13clusters while reducing that of *** interaction between Au and Mn atoms induces more Mn atoms to form spin parallel coupling,resulting in an increase of the total magnetic moment of Mn13clusters,while for the Co13clusters,the interaction between Au and Co atoms does not change the magnetic coupling states between the Co atoms,but reduces the magnetic moment of the Co atoms,leading to a decrease of the total magnetic moment of this *** findings indicate that the encapsulation of Au32clusters can not only raise the chemical stability of TM clusters,but also can tune their magnetic coupling properties and magnetic moment,which enables such systems to be widely applied in fields of spintronics and medical science.

关键词： net Mn Clusters Au Co

A survey of photon mapping state-of-the-art research and future challenges

在线全文

学校读者我要写书评

暂无评论

Frontiers of Information Technology & Electronic Engineering 2016年第3期17卷 185-199页

作者： Chun-meng KANG Lu WANG Yan-ning XU Xiang-xu MENG Department of Computer Science and Technology Shandong University Engineering Research Center of digital media Technology Ministry of Education

Global illumination is the core part of photo-realistic rendering. The photon mapping algorithm is an effective method for computing global illumination with its obvious advantage of caustic and color bleeding rendering. It is an active research field that has been developed over the past two decades. The deficiency of precise details and efficient rendering are still the main challenges of photon mapping. This report reviews recent work and classifies it into a set of categories including radiance estimation, photon relaxation, photon tracing, progressive photon mapping, and parallel methods. The goals of our report are giving readers an overall introduction to photon mapping and motivating further research to address the limitations of existing methods.

关键词： Global illumination Photon mapping Radiance estimation Photon relaxation Progressive photon mapping

Nonlocal image denoising using edge-based similarity metric and adaptive parameter selection

在线全文

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2018年第4期61卷 253-261页

作者： Linwei FAN Xuemei LI Qiang GUO Caiming ZHANG School of Computer Science and Technology Shandong University School of Computer Science and Technology Shandong University of Finance and Economics Shandong Provincial Key laboratory of digital media Technology

Dear editor,We propose a nonlocal image denoising method with an edge-based similarity metric and adaptive parameter *** proposed denoising method uses a two-stage scheme to refine the denoising *** first produces the... 详细信息

关键词： Nonlocal image denoising using edge-based similarity metric and adaptive parameter selection Figure

Fiber laser for on-demand mode generation in 1550 nm band

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Photonics Research 2017年第3期5卷 256-260页

作者： CHENGHUI TIAN SONG YU SHANYONG CAI MINGYING LAN WANYI GU State Key laboratory of Information Photonics and Optical Communications Beijing University of Posts and Telecommunications School of digital media and Design Arts Beijing University of Posts and Telecommunications

Transverse mode characteristics of a laser are related to a variety of interesting applications. An on-demand mode solid laser in the 1064 nm band was proposed previously. In this paper, we provide a fiber laser for on-demand modes in the 1550 nm band to prescribe the pure and high-quality emission of a higher-order transverse laser mode, based on a simple construction with one spatial light modulator(SLM) and a single-mode erbium-doped fiber(SM-EDF). The SLM is designated to generate the desired higher-order mode and separate the higher-order mode and the fundamental mode. The fundamental mode oscillates in the fiber ring laser, and therefore the SM-EDF can be pumped with a single-mode 980 nm laser, no matter what higher-order mode is *** this proof-of-principle experiment, high-quality higher-order modes are observed from LP01 to *** emission and real-time switching between modes can be easily realized by altering the phase on the SLM. In addition, the propagation of the LP_(01), LP_(11), LP_(21), and LP_(02) modes from the fiber laser is also demonstrated in a four-mode few-mode fiber.

关键词： mode as is Fiber laser for on-demand mode generation in 1550 nm band EDF in for nm on

Efficient page layout analysis on small devices

在线全文

学校读者我要写书评

暂无评论

Journal of Zhejiang University-Science A(Applied Physics & Engineering) 2009年第6期10卷 800-804页

作者： Eun-jung HAN Chee-onn WONG Kee-chul JUNG Kyung-ho LEE Eun-yi KIM Department of Naval Architecture and Ocean Engineering Inha University Human-Centered Interfaces Lab Department of Digital MediaSoongsil University Internet and media Engineering Konkuk University

Previously we have designed and implemented new image browsing facilities to support effective offiine image contents on mobile devices with limited capabilities： low bandwidth, small display, and slow processing. In this letter, we fulfill the automatic production of cartoon contents fitting small-screen display, and introduce a clustering method useful for various types of cartoon images as a prerequisite stage for preserving semantic meaning. The usage of neural networks is to properly cut the various forms of pages. Texture information that is useful for grayscale image segmentation gives us a good clue for page layout analysis using the multilayer perceptron （MLP） based x-y recursive algorithm. We also automatically frame the segment MLP using agglomerative segmentation. Our experimental results show that the combined approaches yield good results of segmentation for several cartoons.

关键词： Efficient page layout analysis MLP-based segmentation Mobile devices Image segmentation Neural network