检索结果-内蒙古大学图书馆

International Conference on Computer Science & Education (ICCSE)

作者： Jiancheng Zou Honggen Zhang Institute of Image Processing and Pattern Recognition orth China University of Technology Beijing China

With the rapid development of artificial intelligence technology, more and more intelligent devices are beginning to be manufactured. At the same time, the demand of the image resolution is getting higher. Super-resolution of the image has become a research hotspot in the field of artificial intelligence. The same time, convolution neural networks have been successful in many fields of computer vision, including image super-resolution. Single image super-resolution has received a lot of attention and made good progress recently. However, there are few neural network models for multi-image super-resolution. Array images are a completely new experience and application for super-resolution. In this paper, we use the microarray camera (3*3) to get nine pictures in the same scene and use nine images to realize multiple images super-resolution. A super-resolution model combining software with hardware is proposed. The structure of image registration, V- system image fusion and convolution operation are constructed. The super-resolution model can extract the feature of the target image more efficiently, and the array camera has the characteristic of trading space for the time change. The model can be more accurate and efficient than single image super-resolution. In the PSNR and SSIM indexes, 1.01dB and 0.03 are increased respectively. We use ISO12233 camera resolution chart to detect the experimental pictures. In the experimental results, the resolution of the algorithm is the best.

关键词： image resolution Cameras Convolution Feature extraction Interpolation Neural networks image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Neural networks in image pyramids

Neural networks in image pyramids

引用

International Joint Conference on Neural Networks (IJCNN)

作者： H. Bischof A.J. Pinz Institute for Automation Department for Pattern Recognition and Image Processing Technical University of of Vienna Austria

The authors present a novel neural network model for visual information processing. The model uses a hierarchical network with local connectivity as a stem network. This network generates hypotheses about the expected image content, and then selectively uses small neural network modules on parts of the image to check these hypotheses. The resulting neural network is able to use different spatial resolutions, and is both modular and hierarchical. Applying this model to remotely sensed image classification (Landsat TM) is described. A slightly better classification accuracy was achieved at reduced computational cost, compared to classification without the model.< >

关键词： Neural networks Intelligent networks Spatial resolution pattern recognition Information processing Satellites Remote sensing Humans Visual system Computer vision

来源：评论

学校读者我要写书评

暂无评论

Interior tomography using the truncated Hilbert transform with the total variation constraint

Interior tomography using the truncated Hilbert transform wi...

引用

International Conference on BioMedical Engineering and Informatics (BMEI)

作者： Qiong Xu Xuanqin Mou Institute of Image Processing and Pattern Recognition Xi' an Jiaotong University Xi'an Shaanxi China

Interior tomography is to reconstruct the interior region of interest (ROI) from the projection data just across the ROI. One kind of interior reconstruction methods is based on the inversion of truncated Hilbert transform (THT) when there is a known sub-region inside ROI. However, the result via this method is usually to be degraded by noise in real data case. In this paper, we propose to incorporate the total variation (TV) minimization constraint into the THT-based interior tomography to improve the reconstruction quality. Therein, we first carry out projection-on-convex-sets (POCS) iteration on each chord, and then we perform a soft-threshold based TV minimization on the intermediate image. In order to validate the proposed method, we conduct both simulated and real data experiments. The results show that with TV constraint the proposed method can lead to better ROI with less noise.

关键词： TV image reconstruction Minimization Transforms Noise Computed tomography

来源：评论

学校读者我要写书评

暂无评论

A Novel Approach to Detect Ship-Radiated Signal Based on HMT

A Novel Approach to Detect Ship-Radiated Signal Based on HMT

引用

International Conference on pattern recognition

作者： Yue Zhou Zhibin Niu Chenghao Wang Institute of Image Processing and Pattern Recognition Department of Automation Shanghai Jiaotong University China

ISBN: (纸本)9781424475421

In the presence of non-gaussian noise, we propose a method for the detection of underwater ship-radiated signal. The wavelet decomposition of the underwater signal yields a natural tree structure, which is further modeled by the Hidden Markov Tree (HMT). Therefore, the signal is represented as the parameter of the correspondent HMT. We analysis the likelihood defined on the parameters and form the new detection criteria. Experimental results demonstrate a reliable and robust solution of our method.

关键词： Noise Hidden Markov models Oceans Underwater acoustics Wavelet coefficients Marine vehicles

来源：评论

学校读者我要写书评

暂无评论

SSIM-based CTU-level joint optimal bit allocation and rate distortion optimization

arXiv

引用

arXiv 2020年

作者： Li, Yang Mou, Xuanqin Institute of Image Processing and Pattern Recognition Xi'an Jiaotong University Xi'an710049 China

Structural similarity (SSIM)-based distortion DSSIM is more consistent with human perception than the traditional mean squared error DMSE. To achieve better video quality, many studies on optimal bit allocation (OBA) and rate-distortion optimization (RDO) used DSSIM as the distortion metric. However, many of them failed to optimize OBA and RDO jointly based on SSIM, thus causing a non-optimal R-DSSIM performance. This problem is due to the lack of an accurate R-DSSIM model that can be used uniformly in both OBA and RDO. To solve this problem, we propose a DSSIM-DMSE model first. Based on this model, the complex R-DSSIM cost in RDO can be calculated as simpler RDMSE cost with a new SSIM-related Lagrange multiplier. This not only reduces the computation burden of SSIM-based RDO, but also enables the R-DSSIM model to be uniformly used in OBA and RDO. Moreover, with the new SSIM-related Lagrange multiplier in hand, the joint relationship of R-DSSIM-λSSIM (the negative derivative of R-DSSIM) can be built, based on which the R-DSSIM model parameters can be calculated accurately. With accurate and unified R-DSSIM model, SSIM-based OBA and SSIM-based RDO are unified together in our scheme, called SOSR. Compared with the HEVC reference encoder HM16.20, SOSR saves 4%, 10%, and 14% bitrate under the same SSIM in all-intra, hierarchical and non-hierarchical low-delay-B configurations, which is superior to other state-of-the-art schemes. An improved version of this manuscript has been accepted by IEEE Transactions on Broadcasting DOI 10.1109/TBC.2021.3068871. The project page is located at http://***/web/xqmou/sosr. Copyright © 2020, The Authors. All rights reserved.

关键词： Signal distortion

来源：评论

学校读者我要写书评

暂无评论

Registration-based auto-detection of the optimal cross sections in 3D echocardiographic images

Registration-based auto-detection of the optimal cross secti...

引用

2010 International Conference on Bioinformatics and Biomedical Technology, ICBBT 2010

作者： Liu, Xiaoping Yang, Xin Wu, Lanping Tan, Xiao Computer Center East China Normal University Shanghai China Institute of Pattern Recognition and Image Processing Shanghai Jiaotong University Shanghai China Childerns' Medical Center Shanghai Jiaotong University Shanghai China

ISBN: (纸本)9781424467761

In most cases, echocardiographers examine threedimensional (3D) or four-dimensional (4D) echocardiographic images by studying several important cross sections which are defined as the optimal cross sections. It is a time-consuming and tedious task for echocardiographers to manually search the optimal cross sections by existing real-time 3D echocardiography (RT3DE) systems or off-line 3D softwares. To assist doctors' diagnosis, we design a registration-based method to automatically detect the optimal cross sections in three-dimensional echocardiographic (3DE) images. Firstly, a normal end-diastolic (ED) volumetric data is chosen, in which its optimal cross section is predetermined. Secondly, instead of the whole ED volume, only part of it is selected as the template. Finally, all the optimal cross sections of a 3DE image are detected by aligning the 3DE image with the template image. As a preliminary study, we performed this method on 28 normal apical datasets and the error rate is 15%. The results show good prospects for implementing automatic detection, which might offer better repeatability and save time. Furthermore, it can be a prior procedure for measurement and the computer-aided diagnosis. © 2010 IEEE.

关键词： Echocardiography

来源：评论

学校读者我要写书评

暂无评论

Blind background subtraction in dental panoramic X-ray images: An application approach

Blind background subtraction in dental panoramic X-ray image...

引用

27th DAGM (German Association for pattern recognition) Symposium, DAGM 2005

作者： Goebel, Peter Michael Belbachir, Nabil Ahmed Truppe, Michael Vienna University of Technology Institute of Computer Aided Automation Pattern Recognition and Image Processing Group Austria Fh-campus Wien Technical Project- and Processmanagement Austria Karl Landsteiner Institute for Biotelematics Danube University Krems Austria

Dental Panoramic X-ray images are images having complex content, because several layers of tissue, bone, fat, etc. are superimposed. Non-uniform illumination, stemming from the X-ray source, gives extra modulation to the image, which causes spatially varying X-ray photon density. The interaction of the X-ray photons with the density of matter causes spatially coherent varying noise contribution. Many algorithms exist to compensate background effects, by pixel based or global methods. However, if the image is contaminated by a non-negligible amount of noise, that is usually non-Gaussian, the methods cannot approximate the background efficiently. In this paper, a dedicated approach for background subtraction is presented, which operates blind, that means the separation of a set of independent signals from a set of mixed signals, with at least, only little a priori information about the nature of the signals, using the A-Trous multiresolution transform to alleviate this problem. The new method estimates the background bias from a reference scan, which is taken without a patient. The background values are rescaled by a polynomial compensation factor, given by mean square error criteria, thus subtracting the background will not produce additional artifacts in the image. The energy of the background estimate is subtracted from the energy of the mixture. The method is capable to remove spatially varying noise also, allocating an appropriate spatially noise estimate. This approach has been tested on 50 images from a database of panoramic X-ray images, where the results are cross validated by medical experts. © Springer-Verlag Berlin Heidelberg 2005.

关键词： Medical imaging

来源：评论

学校读者我要写书评

暂无评论

Dual-channel noise reduction via sprase representations

Dual-channel noise reduction via sprase representations

引用

IEEE Workshop on Multimedia Signal processing

作者： Jiancheng Zou Jingsai Liang Xin Yang Institute of Image Processing and Pattern Recognition North-China University of Technology Beijing China

An effective dual-channel noise reduction algorithm is proposed based on sparse representations. The algorithm is composed of the following steps. Firstly, overlapping patches sampled from two channels together instead of each channel one by one are trained to be a dictionary via K-SVD. Secondly, OMP(Orthogonal-Matching-Pursuit) reconstruction algorithm is applied to obtain the sparse coefficients of patches using the dictionary. Thirdly, the denoising speech can be obtained by the updated coefficients. Lastly, the above three steps are iterated to get clearer speech until some conditions are reached. Experimental results show that this algorithm performs better than that with single channel.

关键词： Dictionaries Speech Noise reduction Matching pursuit algorithms Sparse matrices Gaussian noise Discrete cosine transforms

来源：评论

学校读者我要写书评

暂无评论

3D shape retrieval using point bidirectional features

3D shape retrieval using point bidirectional features

引用

International Conference on Computer Science and Network Technology (ICCSNT)

作者： Hui Chen Jingqi Yan Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University Shanghai China

ISBN: (纸本)9781479905607

In this paper, we describe a novel 3D shape retrieval method based on new features. The features are extracted for 3D points based on a 2D attribute space which consists of two bidirectional 3D shape attributes, one of which along the shape surface direction and the other along the shape content direction perpendicular to the former. We define this features as Point Bidirectional Features (PBFs), which can reflect not only the global shape but also the local content. For simplicity, we choose the geodesic distance (GD) and the shape diameter function (SDF) to construct PBFs. Given a database of 3D shapes and a query shape, the proposed retrieval method adopts a shape similarity measurement based on 3D points matching with PBFs to decide which is the most similar shape from the database. To reduce cost of computation and feature storage, a shape simplification algorithm is integrated into this method. Additionally, a simple but effective point correspondence mechanism with K-Nearest Neighbor (KNN) assignment is designed for this retrieval method. Experimental results demonstrate the effectiveness of the proposed features and method.

关键词： Shape Three-dimensional displays Histograms Solid modeling Computational modeling Feature extraction Databases

来源：评论

学校读者我要写书评

暂无评论

Nonlinear multi-scale contrast enhancement for chest radiograph

Nonlinear multi-scale contrast enhancement for chest radiogr...

引用

IEEE International Conference on image processing

作者： Xuanqin Mou Min Zhang Institute of Image Processing and Pattern Recognition Xi'an Jiaotong University Xi'an Shaanxi China

Recently, extensive research and application on contrast enhancement of radiographs based on multi-scale decomposition of the images have validated its higher performance than regular techniques. However, to some extent, conventional multi-scale methods suffered from the introduction of visible artifacts. In this work, we present an algorithm for nonlinear chest radiograph contrast enhancement algorithm within the multi-scale decomposition architecture in spatial domain. In particular, one kind of nonlinear enhancement function is designed by exploiting local contrast information. The main contribution of this model is the local adaptive enhancement ability, which can avoid visible artifacts, while keeping the same detail enhancement ability. In the meantime, no excessive noise is amplified, comparing to conventional methods. Finally, an evaluation using a chest image is provided to demonstrate the effectiveness of the proposed algorithm.

关键词： Computer architecture Diagnostic radiography Humans Rendering (computer graphics) Band pass filters X-ray imaging Kernel image processing pattern recognition Cardiology

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：