检索结果-内蒙古大学图书馆

International conference on Pattern Recognition

作者： M v Rohith Chandra Kambhamettu University of Delaware Newark DE US Dept. of Computer and Information Sciences University of Delaware Delaware USA

Estimation of human motion has been improved by recent advances in depth sensors such as the Microsoft Kinect. However, they often have limited range of depths and a large number of such sensors are necessary to estimate motion in large areas. In this paper, we explore the possibility of estimating motion from monocular data using initial and intermittent 3D models provided by the depth sensor. We use motion segmentation to divide the scene into several rigidly moving components. The orientation of individual components are estimated and these reconstructions are synthesized to provide a coherent estimate of the scene. We demonstrate our algorithm on three sequences from a real video sequence. Quantitative comparison with depth sensor reconstructions show that the proposed method can accurately estimate motion even with a single 3D initialization.

关键词： Motion segmentation image reconstruction Sensors Trajectory Solid modeling Humans Cameras

来源：评论

学校读者我要写书评

暂无评论

visual-inertial curve SLAM

Visual-inertial curve SLAM

引用

IEEE/RSJ International conference on Intelligent Robots and Systems

作者： Kevin Meier Soon-Jo Chung Seth Hutchinson Department of Electrical and Computer Engineering University of Illinois at Urbana-Champaign 61801 USA Department of Aerospace Engineering University of Illinois at Urbana-Champaign 61801 USA

ISBN: (纸本)9781509037636

We present a simultaneous localization and mapping (SLAM) algorithm that uses Bezier curves as static landmark primitives rather than sparse feature points. Our approach allows us to estimate the full 6-DOF pose of a robot while providing a structured map which can be used to assist a robot in motion planning and control. We demonstrate how to reconstruct the 3-D location of curve landmarks from a stereo pair without searching for point-based stereo correspondences and how to compare the 3-D shape of curve landmarks between chronologically sequential stereo frames to solve the data association problem. We present a method to combine curve landmarks for mapping purposes, resulting in a map with a continuous set of curves that contain fewer landmark states than conventional sparse point-based SLAM algorithms. Note, to combine curves, we assume the curved landmarks are fixed to a larger curved object naturally occurring in the scene. While our algorithm is less accurate than point-based SLAM algorithms, we are able to create maps with considerably less landmark states and our algorithm can operate in settings lacking texture.

关键词： Simultaneous localization and mapping Cameras image reconstruction Algorithm design and analysis Roads simultaneous localization and mapping Algorithm design and analysis image reconstruction Cameras Curves Road landmarks supersonic low altitude missile data association Robot maps Stereo pairs

来源：评论

学校读者我要写书评

暂无评论

Robust Object Co-Detection

Robust Object Co-Detection

引用

IEEE conference on Computer vision and Pattern Recognition

作者： Xin Guo Dong Liu Brendan Jou Mojun Zhu Anni Cai Shih-Fu Chang Beijing University of Posts and Telecommunications Dept. of Electrical Engineering Columbia University

ISBN: (纸本)9781467364102

Object co-detection aims at simultaneous detection of objects of the same category from a pool of related images by exploiting consistent visual patterns present in candidate objects in the images. The related image set may contain a mixture of annotated objects and candidate objects generated by automatic detectors. Co-detection differs from the conventional object detection paradigm in which detection over each test image is determined one-by-one independently without taking advantage of common patterns in the data pool. In this paper, we propose a novel, robust approach to dramatically enhance co-detection by extracting a shared low-rank representation of the object instances in multiple feature spaces. The idea is analogous to that of the well-known Robust PCA [28], but has not been explored in object co-detection so far. The representation is based on a linear reconstruction over the entire data set and the low-rank approach enables effective removal of noisy and outlier samples. The extracted low-rank representation can be used to detect the target objects by spectral clustering. Extensive experiments over diverse benchmark datasets demonstrate consistent and significant performance gains of the proposed method over the state-of-the-art object co-detection method and the generic object detection methods without co-detection formulations.

关键词： Object detection dataset image set candidate species Noise images

来源：评论

学校读者我要写书评

暂无评论

Results from the SPI Imaging Test Setup

引用

AIP conference Proceedings 2001年第1期587卷 831-835页

作者： Cornelia B. Wunderer R. Diehl R. Georgii A. v. Kienlin G. G. Lichti v. Schönfelder A. Strong P. Connell F. Sanchez G. vedrenne 1Max-Planck-Institute for extraterrestrial Physics Garching Germany 2University of Birmingham Birmingham United Kingdom 3University of Valencia Valencia Spain 4Centre d’etudes spatiales des rayonnements Toulouse France

The SPI Imaging Test Setup (SPITS) was built at MPE to allow experimental verification of the imaging properties of the SPectrometer onboard INTEGRAL (SPI). Of special importance is the possibility to validate simulations—which are needed for SPI image reconstruction—with laboratory measurements. SPITS consists of a coded mask and two Germanium detectors. The coded mask has the same tungsten-alloy HURA mask coding as SPI and is made of SPI flight model materials. The two hexagonal Ge-detectors are from the SPI flight detector manufacturing line. Housed in a common Al vacuum cap and mounted on an XY-table, they can be moved to cover the 19 Ge detector positions of the SPI camera. We have measured the response of SPITS to radioactive sources (60 kev to 1.8 Mev) at a distance of 9 m from the detector plane. We use both image deconvolution algorithms foreseen for SPI data analysis (spiros and spiskymax) for our analysis. We present our findings for the angular resolution and the point-source-location capability of SPITS for several photon energies and for several source geometries relative to the mask coding.

关键词：

来源：评论

学校读者我要写书评

暂无评论

The Photometry of Intrinsic images

The Photometry of Intrinsic Images

引用

IEEE conference on Computer vision and Pattern Recognition

作者： Marc Serra Olivier Penacchio Robert Benavente Maria vanrell Dimitris Samaras Computer Vision Center/Computer Science Dept. Universitat Autonoma de Barcelona Image Analysis Lab Computer Science Dept. Stony Brook University

ISBN: (纸本)9781479951192

Intrinsic characterization of scenes is often the best way to overcome the illumination variability artifacts that complicate most computer vision problems, from 3D reconstruction to object or material recognition. This paper examines the deficiency of existing intrinsic image models to accurately account for the effects of illuminant color and sensor characteristics in the estimation of intrinsic images and presents a generic framework which incorporates insights from color constancy research to the intrinsic image decomposition problem. The proposed mathematical formulation includes information about the color of the illuminant and the effects of the camera sensors, both of which modify the observed color of the reflectance of the objects in the scene during the acquisition process. By modeling these effects, we get a "truly intrinsic" reflectance image, which we call absolute reflectance, which is invariant to changes of illuminant or camera sensors. This model allows us to represent a wide range of intrinsic image decompositions depending on the specific assumptions on the geometric properties of the scene configuration and the spectral properties of the light source and the acquisition system, thus unifying previous models in a single general framework. We demonstrate that even partial information about sensors improves significantly the estimated reflectance images, thus making our method applicable for a wide range of sensors. We validate our general intrinsic image framework experimentally with both synthetic data and natural images.

关键词： Cameras image color analysis image sensors Light sources Standards Sensor phenomena and characterization

来源：评论

学校读者我要写书评

暂无评论

Design of 2D-multiple notch filter and its application in reducing blocking artifact from DCT coded image

Design of 2D-multiple notch filter and its application in re...

引用

Annual International conference of the IEEE Engineering in Medicine and Biology Society (EMBC)

作者： v.K. Srivastava G.C. Ray Department of Electrical Engineering Indian Institute of Technology Kanpur India

ISBN: (纸本)0780364651

The reconstructed image from highly compressed JPEG data has noticeable degradation due to blocking artifacts. Linear, space-invariant filtering is inadequate to remove these artifacts. So, space-variant or adaptive filtering is required. These schemes are based on edge information extracted from the received blocky image, which contain false edges due to blocking artifacts. Based on analysis in frequency domain, it is found that artifacts in decoded image have periodic structure. These periodic-textures become more obvious after gradient operation since discontinuities due to block coding are highlighted in gradient image. According to signal processing theory, periodic texture will result in some peaks in frequency domain. Here, peaks are due to false edges in gradient image or due to blocking discontinuities in decoded image. A two-dimensional (2D) multiple-notch filter can kill these peaks to reduce the artificial edges due to blocking artifacts in gradient image. Here, a 2D IIR multiple-notch filter is designed from one-dimensional multiple bandpass filter using simple algebraic method. This approach is an extension of method used by Pei at el. (1994) in the domain of 2D. Discontinuities due to blocking artifacts are more in monotone area so this filter can be applied directly. Thus, improvement in the performance of space-variant/adaptive filter by removing of false edges from edge of blocky decoded image and reduction of blocking artifacts directly in monotone area, are the new possible application of multiple-notch filter which is considered in this work. Computational complexity is low because filtering is performed in DCT domain.

关键词： Band pass filters Adaptive filters Decoding IIR filters Filtering Frequency domain analysis image reconstruction image coding Transform coding Degradation

来源：评论

学校读者我要写书评

暂无评论

Perceptually Driven Conditional GAN for Fourier Ptychography

Perceptually Driven Conditional GAN for Fourier Ptychography

引用

Asilomar conference on Signals, Systems & Computers

作者： Abhinau Kumar v. Shashank Gupta Sumohana S. Channappayya Department of Electrical Engineering Indian Institute of Technology Hyderabad Hyderabad India

ISBN: (数字)9781728143002

ISBN: (纸本)9781728143019

Fourier Ptychography (FP) is a computational imaging technique which artificially increases the effective numerical aperture of an imaging system. In FP, the object is imaged using an array of Light Emitting Diodes (LEDs), each from a different illumination angle. A high resolution image is synthesized from this low resolution stack, typically using iterative phase retrieval algorithms. However, such algorithms are time consuming and fail when the overlap between the spectra of images is low, leading to high data requirements. At the crux of FP lies a phase retrieval problem. In this paper, we propose a Deep Learning (DL) algorithm to perform this synthesis under low spectral overlap between samples, and show a significant improvement in phase reconstruction over existing DL algorithms.

关键词：

来源：评论

学校读者我要写书评

暂无评论

LEARNING OvERCOMPLETE SPARSIFYING TRANSFORMS FOR SIGNAL PROCESSING

LEARNING OVERCOMPLETE SPARSIFYING TRANSFORMS FOR SIGNAL PROC...

引用

IEEE International conference on Acoustics, Speech and Signal Processing

作者： Saiprasad Ravishankar Yoram Bresler Department of Electrical and Computer Engineering and the Coordinated Science Laboratory University of Illinois Urbana-Champaign IL 61801 USA

ISBN: (纸本)9781479903573

Adaptive sparse representations have been very popular in numerous applications in recent years. The learning of synthesis sparsifying dictionaries has particularly received much attention, and such adaptive dictionaries have been shown to be useful in applications such as image denoising, and magnetic resonance image reconstruction. In this work, we focus on the alternative sparsifying transform model, for which sparse coding is cheap and exact, and study the learning of tall or overcomplete sparsifying transforms from data. We propose various penalties that control the sparsifying ability, condition number, and incoherence of the learnt transforms. Our alternating algorithm for transform learning converges empirically, and significantly improves the quality of the learnt transform over the iterations. We present examples demonstrating the promising performance of adaptive overcomplete transforms over adaptive overcomplete synthesis dictionaries learnt using K-SvD, in the application of image denoising.

关键词： Sparsifying transform learning Sparse representations dictionary learning Overcomplete representations signal processing Transform sparse representation Dictionaries as Topic image denoising Learning incoherence Self tuning Sparse coding penalty iterative methods

来源：评论

学校读者我要写书评

暂无评论

First Results from a High-Resolution Small Animal SiPM PET Insert for PET/MR Imaging at 7T

引用

IEEE TRANSACTIONS ON NUCLEAR SCIENCE 2016年第5期63卷 2424-2433页

作者： Goertzen, Andrew L. Stortz, Greg Thiessen, Jonathan D. Bishop, Daryl Khan, Muhammad Salman Kozlowski, Piotr Retiere, Fabrice Schellenberg, Graham Shams, Ehsan Sossi, vesna Thompson, Christopher J. Univ Manitoba Dept Radiol Winnipeg MB R3E 3P4 Canada Univ Manitoba Dept Phys & Astron Winnipeg MB R3E 3P4 Canada Hlth Sci Ctr Sect Nucl Med PET CT Imaging Program Winnipeg MB R3E 3P4 Canada Univ British Columbia Dept Phys & Astron Vancouver BC V5T 2E4 Canada Western Univ Imaging Program Lawson Hlth Res Inst London ON N6A 3K7 Canada Western Univ Dept Med Biophys London ON N6A 3K7 Canada TRIUMF Detector Dev Grp Vancouver BC V6T 2A3 Canada Univ Manitoba Dept Elect & Comp Engn Winnipeg MB R3E 3P4 Canada Univ British Columbia Dept Radiol Vancouver BC R3T 2N2 Canada Univ Manitoba Biomed Engn Grad Program Winnipeg MB R3E 3P4 Canada Winnipeg Reg Hlth Author Winnipeg MB R3B 1E2 Canada McGill Univ Montreal Neurol Inst Montreal PQ H3A 0G4 Canada

We present the initial results from a small animal PET insert designed to be operated inside a 7T MRI. The insert fits within the 114 mm inner diameter of the Bruker BGA-12S gradient coil while accommodating the Bruker 35 mm volume RF coil (outer diameter 60 mm), both used in the Bruker 70/20 MRI systems. The PET insert is a ring comprising 16 detectors. Each detector has a dual-layer offset (DLO) lutetium-yttrium oxyorthsilicate (LYSO) scintillator array read out by two SensL SPMArray4B SiPM arrays. The DLO scintillator has bottom (top) layers of: 22 x 10 (21 x 9) crystals of size 1.2 x 1.2 x6 (4) mm(3) for a total of 409 crystals per block, providing an axial extent of 28.17 mm. The detector outputs are multiplexed to four signals using a custom readout board and digitized using the OpenPET data acquisition platform. The detector flood images successfully resolve over 99% of the crystals, with average energy resolution of 12.5 +/- 2.0% at 511 kev. Testing of the PET system inside the MRI showed that the PET insert had no effect on MRI image homogeneity and only a small effect on echo planar images (EPI) signal to noise ratio (SNR) (-9%), with neither PET nor MRI images showing obvious artefacts. These acquisitions used the OpenPET operating in "oscilloscope mode" with USB2.0 interface, allowing a maximum total singles event rate of 280 kcps, strongly limiting the count rate capabilities of the system. The PET radial spatial resolution (as measured with a Na-22 point source and FBP-3DRP reconstruction) is 1.17 mm at the centre, degrading to 1.86 mm at a 15 mm radial offset. Simultaneous phantom and mouse PET/MR imaging produced good quality images that were free of any obvious artefacts.

关键词： Dual-layer offset scintillator PET insert PET/MRI silicon photomultiplier small animal PET

来源：评论

学校读者我要写书评

暂无评论

Quantitative CT-less PET Imaging with Energy-based Scatter Estimations

Quantitative CT-less PET Imaging with Energy-based Scatter E...

引用

IEEE Symposium on Nuclear Science (NSS/MIC)

作者： M. Teimoorisichani J. H. Hamill H. Sari v. Panin A. Rominger M. Conti Siemens Medical Solutions Inc. Knoxville Tennessee United States of America Department of Nuclear Medicine Bern University Hospital Bern Switzerland Advanced Clinical Imaging Technology Siemens Healthcare AG Lausanne Switzerland

There has been an increasing interest in low-dose PET scans, in particular after the introduction of long-axial field-of-view PET/CT scanners. Performing a CT scan in studies where CT is merely done for the attenuation and scatter correction of PET data undermines the true low-dose premise of such PET studies. We have shown in the past that one can potentially use the background radiation from lutetium in long-axial field-of-view PET scanners with lutetium-based scintillators to reconstruct an estimate of attenuation maps which can then be improved in a joint reconstruction algorithm such as maximum likelihood reconstruction of activity and attenuation (MLAA). However, the lack of an accurate scatter estimates in such algorithms can lead to a bias and quantitative inaccuracies in the reconstructed PET images. In this work, we propose and evaluate the use of an energy-based scatter estimation algorithm in CT-less reconstruction for a long-axial field-of-view PET/CT scanner. Using the data from a NEMA NU-2 image quality phantom scan on a Biograph vision Quadra PET/CT scanner (Siemens Healthineers), reconstructed images with and without the use CT and with and without the use of energy-based scatter estimates are shown. The results suggest that images reconstructed with CT-based correction and standard ordered-subset expectation maximization reconstructions match well with CT-less reconstructions and energy-based scatter. In conclusions, the use of an energy-based scatter estimation algorithm is a promising solution to the issue of scatter in CT-less reconstruction methods.

关键词：

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：