检索结果-内蒙古大学图书馆

A saliency model based on wavelet transform and visual attention

Science China(Information Sciences) 2010年第4期53卷 738-751页

作者： LI ZhiQiang, FANG Tao* & HUO Hong institute of {1. processing and pattern recognition, Shanghai Jiaotong University, Shanghai 200240, China 1. Institute of Image Processing and Pattern Recognition Shanghai Jiaotong University Shanghai 200240 China

This paper presents a novel wavelet transform saliency model to detect salient objects. In this model, a saliency map is generated by combining orientation feature maps obtained from wavelet transform of different scale images derived from the same image. Then, the order map of a saliency map is obtained by using Fourier descriptor, which could be used as a guidance to process the most important objects. Experiments indicate that this saliency model is robust to noise and superior to other saliency models in the literature.

关键词： bottom-up attentional mechanism center-surround inhibition phenomenon Fourier descriptor top-down attentional mechanism visual attention wavelet transform

来源：评论

学校读者我要写书评

暂无评论

Regularities of formation of ternary intermetallic compounds between two transition elements and one non-transition element

引用

Science China Technological Sciences 2001年第1期44卷 42-46页

作者：姚莉秀陈念贻杨杰陈瑞亮钦佩 1. Institute of Image Processing and Pattern Recognition Shanghai Jiaotong University 200030 Shanghai China 2. Shanghai Institute of Metallurgy Chinese Academy of Science 200050 Shanghai China

The pattern {1. methods and a four-parameter model, based on extended Miede-ma’s cellular model of alloy phases, are used to study the regularities of formation of ternary compounds between two transition elements (T,T ’) and one non-transition element (N) (T-T ’-N system) . The criterion of formation can be expressed as some functions of φ (electronegativity), n1.3ws (valence electron density in Wagner-Seitz cell), R (Pauling’s metallic radii) and Z (number of valence electrons in atom).

关键词： T-T ’-N alloy system criterion of formation ternary intermetallic compounds.

来源：评论

学校读者我要写书评

暂无评论

Automatic segmentation of bladder in CT images

引用

Journal of Zhejiang University-Science A(Applied Physics & Engineering) 2009年第2期10卷 239-246页

作者： Feng SHI Jie YANG Yue-min ZHU Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University Shanghai 200240 China CREATIS. LNSA of Lyon University of Lyon 1 Villeurbanne France

Segmentation of the bladder in computerized tomography(CT) images is an important step in radiation therapy planning of prostate cancer. We present a new segmentation scheme to automatically delineate the bladder contour in CT images with three major steps. First,we use the mean shift algorithm to obtain a clustered image containing the rough contour of the bladder,which is then extracted in the second step by applying a region-growing algorithm with the initial seed point selected from a line-by-line scanning process. The third step is to refine the bladder contour more accurately using the rolling-ball algorithm. These steps are then extended to segment the bladder volume in a slice-by-slice manner. The obtained results were compared to manual segmentation by radiation oncologists. The average values of sensitivity,specificity,positive predictive value,negative predictive value,and Hausdorff distance are 86.5%,96.3%,90.5%,96.5%,and 2.8 pixels,respectively. The results show that the bladder can be accurately segmented.

关键词： image segmentation Computerized tomography (CT) Mean shift Bladder Rolling ball

来源：评论

学校读者我要写书评

暂无评论

Restoring magnetic anomalies

引用

Archaeological Prospection 1996年第4期3卷 185-197页

作者： Eder-Hinterleitner, A. Neubauer, W. Melichar, P. Institute for Automation Dept. For Pattern Recognition and Image Processing Vienna A-1040 Treitlstr. 3/1832 Austria Institute for Prehistory University of Vienna Vienna A-1190 F. Kleing. 1 Austria Central Institute for Meteorology and Geodynamics Vienna A-1190 Hohe Warte 38 Austria

A caesium magnetometer attached to a handcart enables accurate magnetic prospection of large areas within short times. The roughness of the earth's surface, minor differences in handling the handcart and a lot of other reasons are possible sources for various errors during a survey. Standard image pre{1. methods, such as median or low-pass filtering, remove not only errors but also important information and deform the shape of magnetic anomalies. We present four special pre{1. methods to remove errors automatically by protecting and restoring the real value and shape of anomalies and compare them with standard pre{1. methods. The four methods are equalizing subgrid shifts, equalizing line shifts, removing spikes and removing displacement errors. © 1.96 John Wiley and Sons, Ltd.

关键词： image pre{1. image restoration Magnetics Prospection

来源：评论

学校读者我要写书评

暂无评论

Head detection and localization from sparse 3D data

Head detection and localization from sparse 3D data

引用

24th German Association for pattern {1. Symposium, DAGM 2002

作者： Clabian, Markus Rötzer, Harald Bischof, Horst Kropatsch, Walter Advanced Computer Vision GmbH Donau City Str.1 ViennaA-1220 Austria Institute for Computer Graphics and Vision Graz University of Technology Inffeldgasse 16 GrazA-8010 Austria Pattern Recognition and Image Processing Group Institute of Computer Aided Automation Computer Science Department Vienna University of Technology Favoritenstr. 9 ViennaA-1040 Austria

ISBN: (纸本)354044209X

Head detection is an important, but difficult task, if no restrictions such as static illumination, frontal face appearance or uniform background can be assumed. We present a system that is able to perform head detection under very general conditions by employing a 3D measurement system namely a structured light distance measurement. An algorithm of head detection from sparse 3D data (1.x1. data points) is developed that reconstructs a 3D surface over the image plane and detects head hypotheses of ellipsoidal shape. We demonstrate that detection and rough localization is possible in up to 90% of the images. © Springer-Verlag Berlin Heidelberg 2002.

关键词： Computers

来源：评论

学校读者我要写书评

暂无评论

On the detection of individuals and vehicles on an airport's apron

On the detection of individuals and vehicles on an airport's...

引用

IEE International Conference on Visual Information Engineering (VIE 2005)

作者： Josep Aguilera Antequera Martin Kampel Vienna University of Technology Institute of Computer Aided Automation Pattern Recognition and Image Processing Group Favoritenstr.9 183-2 A-1 040 Vienna Austria

This paper gives a robust motion detection and tracking solution for a video surveillance application on an airport's apron. As an outdoor application, the system must be capable of adapting to a wide range of weather conditions and illumination changes, furthermore, the achromaticity of the scene and the presence of occlusions in the tracking process are issues considered in the selection of the motion detector and tracking system respectively. We propose an adapted mixture of Gaussiani model with RGB colour normalisation to detect mobile objects in the scene and a region tracking method based on significant mobile object features to track individuals and vehicles on the selected airport's apron. The performance of the proposed motion detector is evaluated using pixel-based performance metrics and compared with other existing methods. The capability of the application to handle partial occlusions is tested on the region tracker.

关键词： Tracking Airports Surveillance Adaptation models Visualization Motion detection Atmospheric modeling

来源：评论

学校读者我要写书评

暂无评论

Measurement of Imaging Indexes of Distal Radius based on deep Learning Key Point Detection 23

Measurement of Imaging Indexes of Distal Radius based on dee...

引用

9th International Conference on Computing and Artificial Intelligence, ICCAI 2023

作者： Jiachao, Niu Xin, Zhang Xufeng, Ling Yang, Jie Yin, Yong Shanghai Normal University Tianhua College AI School No. 1661 North Sheng Xin Road China Department of Orthopedics Jiading District Central Hospital Affiliated Shanghai University of Medicine &Health Sciences No.1 Chengbei Road China Institute of Image Processing and Pattern Recognition Shanghai Jiaotong University 800 Dongchuan Road 200240 China

ISBN: (纸本)9781450399029

The measurement of imaging indexes of distal radius is the basic work of diagnosis and recovery evaluation. Due to the problems of large morphological differences and insufficient clarity in radial images, the accurate measurement of indexes relies heavily on the clinical experience of doctors, with defects such as low accuracy and poor consistency. In order to solve the above problems, this paper proposes an intelligent technology to achieve accurate measurement of imaging indicators of the distal radius. Firstly, 1. key points needed for distal imaging indexes were determined: radius left vertex, radius right vertex, radius left neck point, radius right neck point, radius left lower point, radius right lower point, ulna left vertex, ulna right vertex, ulna left neck point, right neck point, ulna left lower point, ulna right lower point. Then, the deep convolutional neural network was used to estimate the geometric transformation features and regress the thermal map of the feature points, so as to achieve accurate positioning of the 1. feature points. Finally, each imaging index was calculated accurately by the preset computational method. The experimental results show that the proposed measurement method has advantages of high accuracy, good consistency and fast speed. The test accuracy is higher than the manual measurement results of inexperienced doctors, and the network model has good robustness and generalization ability. This method has practical value and can be applied in the field of medical imaging to help doctors improve the efficiency of diagnosis. © 2023 Copyright held by the owner/author(s).

关键词： Diagnosis

来源：评论

学校读者我要写书评

暂无评论

AMIL: Adversarial multi instance learning for human pose estimation

arXiv

引用

arXiv 2020年

作者： SHAMSOLMOALI, POURYA ZAREAPOOR, MASOUMEH ZHOU, HUIYU YANG, JIE Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University Shanghai China Department of Informatics University of Leicester LeicesterLE1 7RH United Kingdom

Human pose estimation has an important impact on a wide range of applications from human-computer interface to surveillance and content-based video retrieval. For human pose estimation, joint obstructions and overlapping upon human bodies result in departed pose estimation. To address these problems, by integrating priors of the structure of human bodies, we present a novel structure-aware network to discreetly consider such priors during the training of the network. Typically, learning such constraints is a challenging task. Instead, we propose generative adversarial networks as our learning model in which we design two residual multiple instance learning (MIL) models with the identical architecture, one is used as the generator and the other one is used as the discriminator. The discriminator task is to distinguish the actual poses from the fake ones. If the pose generator generates the results that the discriminator is not able to distinguish from the real ones, the model has successfully learnt the priors. In the proposed model, the discriminator differentiates the ground-truth heatmaps from the generated ones, and later the adversarial loss back-propagates to the generator. Such procedure assists the generator to learn reasonable body configurations and is proved to be advantageous to improve the pose estimation accuracy. Meanwhile, we propose a novel function for MIL. It is an adjustable structure for both instance selection and modeling to appropriately pass the information between instances in a single bag. In the proposed residual MIL neural network, the pooling action adequately updates the instance contribution to its bag. The proposed adversarial residual multi-instance neural network that is based on pooling has been validated on two datasets for the human pose estimation task and successfully outperforms the other state-of-arts models. The code will be made available on https://***/pshams55/AMIL. Copyright © 2020, The Authors. All rights reserve

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Combining Feature Extraction and Hidden Markov Modelling for Automatic Speech recognition

Combining Feature Extraction and Hidden Markov Modelling for...

引用

International Conference on Communication Technology(ICCT’94)

作者： WU Jianxiong CHEN Peifang CHAN Chorkin DENG Li Institute of Image Processing and Pattern Recognition Shanghai Jiaotong University Department of Computer Science University of Hong Kong Dcpartment of Electrical and Computer Engineering University of WaterlooWaterlooOntarioCanada N2L 3G1.

＜正＞Feature extraction is very important for the classifier design and the overall performance of *** recognition ***,due to the lack of theoretical guidances,feature extraction and classifier design are usually treated separately in current speech recognition *** *** proposes an approach to combine linear feature extraction with continuous density hidden Markov modeling（HMM） which is currently the most successful speech pattern classifier.A maximumlikelihood based algorithm is derived to iteratively train HMM parameters as well as the parameters of the feature *** algorithm is an exteusion of the Baum-Welcli parameter re-estimation algorithm for conventional HMMs and thus has a nice property of guara, nteed convergence.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multi-patch feature pyramid network for weakly supervised object detection in optical remote sensing images

arXiv

引用

arXiv 2021年

作者： Shamsolmoali, Pourya Chanussot, Jocelyn Zareapoor, Masoumeh Zhou, Huiyu Yang, Jie The Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University Shanghai China LJK CNRS Inria Grenoble INP Université Grenoble Alpes Grenoble38000 France The School of Informatics University of Leicester LeicesterLE1 7RH United Kingdom

To read the paper please go to IEEE Transactions on Geoscience and Remote Sensing on IEEE Xplore. Object detection is a challenging task in remote sensing because objects only occupy a few pixels in the images, and the models are required to simultaneously learn object locations and detection. Even though the established approaches well perform for the objects of regular sizes, they achieve weak performance when analyzing small ones or getting stuck in the local minima (e.g. false object parts). Two possible issues stand in their way. First, the existing methods struggle to perform stably on the detection of small objects because of the complicated background. Second, most of the standard methods used hand-crafted features, and do not work well on the detection of objects parts of which are missing. We here address the above issues and propose a new architecture with a multiple patch feature pyramid network (MPFP-Net). Different from the current models that during training only pursue the most discriminative patches, in MPFP-Net the patches are divided into class-affiliated subsets, in which the patches are related and based on the primary loss function, a sequence of smooth loss functions are determined for the subsets to improve the model for collecting small object parts. To enhance the feature representation for patch selection, we introduce an effective method to regularize the residual values and make the fusion transition layers strictly norm-preserving. The network contains bottom-up and crosswise connections to fuse the features of different scales to achieve better accuracy, compared to several state-of-the-art object detection models. Also, the developed architecture is more efficient than the baselines. © 2021. CC BY-NC-SA.

关键词： Object recognition

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：