检索结果-内蒙古大学图书馆

ieee 17th International symposium on Biomedical Imaging (ISBI)

作者： Zhang, Yuxi Liu, Risheng Li, Zi Liu, Zhu Fan, Xin Luo, Zhongxuan Dalian Univ Technol DUT RU Int Sch Informat Sci & Engn Dalian Peoples R China Key Lab Ubiquitous Network & Serv Software Liaoni Dalian Peoples R China Dalian Univ Technol Sch Math Sci Dalian Peoples R China Guilin Univ Elect Technol Inst Artificial Intelligence Dalian Peoples R China Peng Cheng Lab Shenzhen Peoples R China

ISBN: (纸本)9781538693308

Deformable 3D medical image registration is challenging due to the complicated transformations between image pairs. Traditional approaches estimate deformation fields by optimizing a task-guided energy embedded with physical priors, achieving high accuracy while suffering from expensive computational loads for the iterative optimization. Recently, deep networks, encoding the information underlying data examples, render fast predictions but severely dependent on training data and have limited flexibility. In this study, we develop a paradigm integrating the principled prior into a bidirectional deep estimation process. Inheriting from the merits of both domain knowledge and deep representation, our approach achieves a more efficient and stable estimation of deformation fields than the state-of-the-art, especially when the testing pairs exhibit great variations with the training.

关键词： Medical image registration optimization deep learning 3D neural network

来源：评论

学校读者我要写书评

暂无评论

MoNuSAC2020: A Multi-Organ Nuclei Segmentation and Classification Challenge

引用

ieee TRANSACTIONS ON MEDICAL IMAGING 2021年第12期40卷 3413-3423页

作者： Verma, Ruchika Kumar, Neeraj Patil, Abhijeet Kurian, Nikhil Cherian Rane, Swapnil Graham, Simon Quoc Dang Vu Zwager, Mieke Raza, Shan E. Ahmed Rajpoot, Nasir Wu, Xiyi Chen, Huai Huang, Yijie Wang, Lisheng Jung, Hyun Brown, G. Thomas Liu, Yanling Liu, Shuolin Jahromi, Seyed Alireza Fatemi Khani, Ali Asghar Montahaei, Ehsan Baghshah, Mahdieh Soleymani Behroozi, Hamid Semkin, Pavel Rassadin, Alexandr Dutande, Prasad Lodaya, Romil Baid, Ujjwal Baheti, Bhakti Talbar, Sanjay Mahbod, Amirreza Ecker, Rupert Ellinger, Isabella Luo, Zhipeng Dong, Bin Xu, Zhengyu Yao, Yuehan Lv, Shuai Feng, Ming Xu, Kele Zunair, Hasib Ben Hamza, Abdessamad Smiley, Steven Yin, Tang-Kai Fang, Qi-Rui Srivastava, Shikhar Mahapatra, Dwarikanath Trnavska, Lubomira Zhang, Hanyun Narayanan, Priya Lakshmi Law, Justin Yuan, Yinyin Tejomay, Abhiroop Mitkari, Aditya Koka, Dinesh Ramachandra, Vikas Kini, Lata Sethi, Amit Case Western Reserve Univ Dept Biomed Engn Cleveland OH 44106 USA Univ Alberta Dept Comp Sci Edmonton AB T6G 2R3 Canada Alberta Machine Intelligence Inst Edmonton AB T5J 3B1 Canada Indian Inst Technol Dept Elect Engn Mumbai 400076 Maharashtra India HBNI Tata Mem Ctr ACTREC Dept Pathol Mumbai 400012 Maharashtra India Univ Warwick Dept Comp Sci Coventry CV4 7AL W Midlands England Univ Groningen Dept Pathol NL-9712 Groningen Netherlands Shanghai Jiao Tong Univ Inst Image Proc & Pattern Recognit Dept Automat Shanghai 200240 Peoples R China NCI Adv Biomed Comp Sci Frederick Natl Lab Canc Res Frederick MD 21702 USA Anhui Univ Dept Elect Engn & Automat Hefei 230039 Peoples R China Sharif Univ Technol Dept Comp Engn Tehran 113658639 Iran Sharif Univ Technol Dept Elect Engn Tehran 113658639 Iran Xperience AI Nizhnii Novgorod 603155 Russia SGGS Inst Engn & Technol Ctr Excellence Signal & Image Proc Nanded 431606 India Med Univ Vienna Inst Pathophysiol & Allergy Res A-1090 Vienna Austria TissueGnostics GmbH Dept Res & Dev Vienna Austria DeepBlue Technol Shanghai Co Ltd Shanghai 200336 Peoples R China Guangdong Univ Technol Sch Automat Guangzhou 510006 Peoples R China Tongji Univ Sch Elect & Informat Engn Shanghai 200092 Peoples R China Natl Univ Def Technol Natl Key Lab Parallel & Distributed Proc Changsha 410073 Peoples R China Concordia Univ Concordia Inst Informat Syst Engn Montreal PQ H4B 1R6 Canada Jensen Hughes Site Risk Management Engineer Liverpool NY 13212 USA Natl Univ Kaohsiung Dept Comp Sci & Informat Engn Kaohsiung 811 Taiwan Incept Inst Artificial Intelligence Abu Dhabi U Arab Emirates Slovak Univ Technol Bratislava Fac Informat & Informat Technol Bratislava 81107 Slovakia Inst Canc Res Ctr Evolut & Canc London SW7 3RP England Inst Canc Res Div Mol Pathol London SW7 3RP England Onward Hlth Hyderabad 500032 India

Detecting various types of cells in and around the tumor matrix holds a special significance in characterizing the tumor micro-environment for cancer prognostication and research. Automating the tasks of detecting, segmenting, and classifying nuclei can free up the pathologists' time for higher value tasks and reduce errors due to fatigue and subjectivity. To encourage the computer vision research community to develop and test algorithms for these tasks, we prepared a large and diverse dataset of nucleus boundary annotations and class labels. The dataset has over 46,000 nuclei from 37 hospitals, 71 patients, four organs, and four nucleus types. We also organized a challenge around this dataset as a satellite event at the International symposium on Biomedical Imaging (ISBI) in April 2020. The challenge saw a wide participation from across the world, and the top methods were able to match inter-human concordance for the challenge metric. In this paper, we summarize the dataset and the key findings of the challenge, including the commonalities and differences between the methods developed by various participants. We have released the MoNuSAC2020 dataset to the public.

关键词： Annotations image segmentation Tumors Computer architecture Training Task analysis Semantics Multi-organ dataset nucleus classification computational pathology instance segmentation panoptic quality

来源：评论

学校读者我要写书评

暂无评论

Accelerating Towards Larger Deep Learning Models and Datasets - A System Platform View Point 34

Accelerating Towards Larger Deep Learning Models and Dataset...

引用

34th ieee International Parallel and Distributed processing symposium (IPDPS)

作者： Vinod, Saritha Naveen, M. Patra, Asis K. John, Anto Ajay Raj IBM Corp India Syst Dev Lab Bangalore Karnataka India

ISBN: (纸本)9781728174457

Deep Learning (DL) is a rapidly evolving field under the umbrella of Artificial intelligence (AI) with proven real-world use cases in supervised and unsupervised learning tasks. As the complexity of the learning tasks increases, the DL models become deeper or wider with millions of parameters and use larger datasets. Neural networks like AmoebaNet with 557M parameters and GPT-2 with 1.5 billion parameters are some of the recent examples of large models. DL trainings are generally run on accelerated hardware such as GPUs, TPUs or FPGAs which can satisfy the high computational demands of the neural network training. But accelerators are limited in their memory capacities. Larger the models, larger the memory required while training them. Hence, large DL models and large datasets cannot fit into the limited memory available on GPUs. However, there are techniques designed to overcome this limitation like compression, using CPU memory as a data swap, recomputations within the GPUs etc. But the efficiency of each of these techniques also depends on the underneath system platform capabilities. In this paper we present the observations from our study of training large DL models using data swap method on different system platforms. This study showcases the characteristics of large models and presents the system viewpoint of large deep learning model training by studying the relation of the software techniques to the system platform used underneath. The results presented in the paper show that for training large Deep Learning models, communication link between CPU and GPU is critical and the training performance can be improved by using a platform with high bandwidth link for this communication. The results presented are based on two DL models, 3DUnetCNN model for medical image segmentation and DeepLabV3+ model for semantic image segmentation.

关键词： Deep Learning Large Model Support Performance 3DUnetCNN Medical image Segmentation Semantic Segmentation DeepLabV3+ GPUs TensorFlow

来源：评论

学校读者我要写书评

暂无评论

The Effect of Spectral Resolution Upon the Accuracy of Brain Tumor Segmentation from Multi-Spectral MRI Data 18

The Effect of Spectral Resolution Upon the Accuracy of Brain...

引用

18th ieee World symposium on Applied Machine intelligence and Informatics (SAMI)

作者： Gyorfi, Agnes Fulop, Timea Kovacs, Levente Szilagyi, Laszlo Obuda Univ Univ Res Innovat & Serv Ctr EKIK Becsi Ut 96-b H-1034 Budapest Hungary Sapientia Univ Dept Elect Engn Calea Sighisoarei 1-C Targu Mures 540485 Romania Sapientia Univ Dept Math Informat Calea Sighisoarei 1-C Targu Mures 540485 Romania

ISBN: (纸本)9781728131498

Ensemble learning methods are frequently employed for brain tumor segmentation from multi-spectral MRI data. These techniques often require involving several hundreds of computed features for the characterization of the voxels, causing a rise in the necessary storage space by two order of magnitude. processing such amounts of data also represents a serious computational burden. Under such circumstances it is useful to optimize the feature generation process. This paper proposes to establish the optimal spectral resolution of multi-spectral MRI data based feature values that allows for the best achievable brain tumor segmentation accuracy without causing unnecessary computational load and storage space waste. Experiments revealed that an 8-bit spectral resolution of the MRI-based feature data is sufficient to obtain the best possible accuracy of ensemble learning methods, while it allows for 50% reduction of the storage space required by the segmentation procedure, compared to the usually deployed featured encoding techniques.

关键词： Magnetic resonance imaging spectral resolution image segmentation ensemble learning

来源：评论

学校读者我要写书评

暂无评论

Raindrops on Windshield: Dataset and Lightweight Gradient-Based Detection Algorithm

Raindrops on Windshield: Dataset and Lightweight Gradient-Ba...

引用

ieee symposium Series on computational intelligence (SSCI)

作者： Vera Soboleva Oleg Shipitko Evocargo LLC Moscow Russia Vision Systems Laboratory Institute for Information Transmission Problems Moscow Russia Moscow Institute of Physics and Technology Dolgoprudny Russia

ISBN: (纸本)9781728190495

Autonomous vehicles use cameras as one of the primary sources of information about the environment. Adverse weather conditions such as raindrops, snow, mud, and others, can lead to various image artifacts. These artifacts significantly degrade the quality and reliability of the obtained visual data and can lead to accidents if they are not detected in time. This paper presents ongoing work on a new dataset for training and assessing vision algorithms' performance for different tasks of image artifacts detection on either camera lens or car windshield. At the moment, we present a publicly available set of images containing 8190 images, of which 3390 contain raindrops. images are annotated with the binary mask representing areas with raindrops. We demonstrate the applicability of the dataset in the problems of raindrops presence detection and raindrop region segmentation. To augment the data, we also propose a data augmentation algorithm that allows the generation of synthetic raindrops on images. Apart from the dataset, we present a novel gradient-based algorithm for raindrop presence detection in a video sequence. The experimental evaluation proves that the algorithm reliably detects raindrops. Moreover, compared with the state-of-the-art cross-correlation-based algorithm, the proposed algorithm showed a higher quality of raindrop presence detection and image processing speed, making it applicable for the self-check procedure of real autonomous systems. The dataset is available at ***/EvoCargo/RaindropsOnWindshield.

关键词： Training Measurement image segmentation Visualization Cameras Classification algorithms Reliability

来源：评论

学校读者我要写书评

暂无评论

image classification of cyanobacteria Microcystis aeruginosa in raw water samples in Curitiba's region 33

Image classification of cyanobacteria Microcystis aeruginosa...

引用

33rd ieee International symposium on Computer-Based Medical Systems (CBMS)

作者： Melo, Geisla de Albuquerque Sobieranski, Antonio Carlos Veiga Ludwig, Thelma Alvim de Oliveira, Lucas Ferrari Univ Fed Parana Informat Dept Curitiba Parana Brazil Univ Fed Santa Catarina Comp Engn Dept Ararangua Brazil Univ Fed Parana Bot Dept Curitiba Parana Brazil

ISBN: (纸本)9781728194295

Surveys studies carried out in different countries demonstrated that about 75% of lake water samples contain toxic cyanobacteria [4, 5]. The toxin that is must recurrent in Brazilian water bodies are the microcystin [6]. Microcystins are hepatotoxins produced by some species of cyanobacteria, and act by inflicting damage to cells from the liver and other organs [7]. Cases around the world have been reported over time, including health injuries by contact direct or not with contaminated water, as well livestock and humans deaths [4]. Due to its sanitary importance, the water that the population consumes needs regular monitoring in order to avoid contact and possible adverse effects. The objective of this work is to propose a computational approach for the identification of Microcystis aeruginosa from images, in raw water reservoirs, automatically, using artificial intelligence techniques. The identification was developed using convolutional neural networks (CNN), where two distinct model were compared. The obtained results demonstrated the effectiveness of the proposed approach to solve the problem of recognizing cyanobacterias.

关键词： cyanobacteria convolutional neural networks image processing

来源：评论

学校读者我要写书评

暂无评论

AN EFFICIENT MEO SAR IMAGING ALGORITHM BASED ON OPTIMAL IMAGING COORDINATE SYSTEM

AN EFFICIENT MEO SAR IMAGING ALGORITHM BASED ON OPTIMAL IMAG...

引用

ieee International Geoscience and Remote Sensing symposium (IGARSS)

作者： Liu, Wenkang Sun, Guang-Cai Xing, Mengdao Pascazio, Vito Xidian Univ Natl Lab Radar Signal Proc Xian 710071 Peoples R China Xidian Univ Collaborat Innovat Ctr Informat Sensing & Underst Xian 710071 Peoples R China Univ Napoli Parthenope Dipartimento Tecnol I-80143 Naples Italy

ISBN: (纸本)9781728163741

The curved trajectory and long synthetic aperture time of medium-earth-orbit (MEO) synthetic aperture radar (SAR) lead to a two-dimensional spatial variation in the signals Traditional methods treat the range and azimuth variations separately, and usually suffer from high computational complexities. We investigate the Doppler rate distribution across a large scene, and exploit an optimal imaging coordinate system, in which the MEO SAR signals satisfy the azimuth-shift-invariant property. The additional processing of the azimuth spatial variation in MEO SAR imaging algorithms can be avoided, and the efficiency of the image formation processor can be improved. Finally, processing of simulated stripmap-mode data with 2-m resolution can validate the proposed algorithm.

关键词： MEO SAR effective focusing algorithm imaging coordinate system

来源：评论

学校读者我要写书评

暂无评论

Harvesting End-effector Design and Picking Control

Harvesting End-effector Design and Picking Control

引用

ieee symposium Series on computational intelligence (SSCI)

作者： Jonah Mapes Andong Dai Yunjun Xu Shinsuke Agehara Mechanical and Aerospace Engineering University of Central Florida Orlando USA Gulf Coast Research and Education Center University of Florida Wimauma USA

ISBN: (纸本)9781728190495

As an enabling technology in robotic harvesting, an end-effector requires precise positioning and picking without bruising strawberries or sabotaging strawberry beds. Here, an end-effector design with a spiral-curve tooth is presented, which can contain strawberries in a spherical shell before pulling them away from their stems. Visual feedback with a camera-in-hand configuration is used to identify strawberries of sufficient size and ripeness. A two-step estimation and control approach is proposed including aligning the end-effector with the target strawberries in the $X$ and $Y$ directions and then reaching them along the $Z$ direction. Specifically, the control logic is designed to correct the depth error based on the changing rate of a strawberry diameter in the pixel frame. The unique feature of this approach is that it can precisely move the end-effector to the desired position regardless of the initial depth and size of the target. The end-effector, XYZ table, and control algorithm were extensively tested in an artificial strawberry bed and also partially experimented in a commercial farm.

关键词： Visualization Robot kinematics image processing Laboratories Force Estimation End effectors

来源：评论

学校读者我要写书评

暂无评论

Kidney segmentation for quantitative analysis applying MaskRCNN architecture

Kidney segmentation for quantitative analysis applying MaskR...

引用

ieee symposium Series on computational intelligence (SSCI)

作者： Mathilde Overgaard Lauersen Büsra Köylü Bryan Haddock John Aa. Sorensen CIMT The Capital Region of Denmark Copenhagen Denmark DTU Engineering Technology Technical University of Denmark Ballerup Denmark Clinical Physiology Nuclear Medicine and PET Rigshospitalet Glostrup Glostrup Denmark

ISBN: (纸本)9781728190495

Results of applying the deep learning Mask Regional Convolutional Neural Network (MaskRCNN) to the problem of segmenting human kidney and delineating the kidney tissue into the 12 classes left/right kidney, cortex, medulla 1, 2, 3, 4 are presented, using a 3T MRI scanner. This problem originates from advances in magnetic resonance imaging MRI techniques to monitor blood perfusion and blood oxygenation in kidneys of persons suffering from reduced kidney function. This segmentation allows the processing of the numerous MRI arterial spin labelling (ASL) images and T2* images with respiration movement to calculate and map renal perfusion and evaluate oxygenation respectively. The quality of the segmentation is assessed visually and by the segment label classifier. In total, a selection of 37 segmentation experiments, combining MRI sequences and kidney tissue types are presented. To the best of the author's knowledge, applying this segmentation method on kidney tissue is new.

关键词： image segmentation Histograms Annotations Statistical analysis Magnetic resonance imaging Computer architecture Kidney

来源：评论

学校读者我要写书评

暂无评论

Part-level Action Parsing via a Pose-guided Coarse-to-Fine Framework

Part-level Action Parsing via a Pose-guided Coarse-to-Fine F...

引用

ieee International symposium on Circuits and Systems

作者： Xiaodong Chen Xinchen Liu Wu Liu Kun Liu Dong Wu Yongdong Zhang Tao Mei Universiy of Science nd Technoogy of Chin Hefei ChinJD.co Beijing ChinJD.co Beijing ChinJD.co Beijing ChinJD.co Beijing ChinUniversiy of Science nd Technoogy of Chin Hefei ChinJD.co Beijing Chin

ISBN: (数字)9781665484855

ISBN: (纸本)9781665484862

Action recognition from videos, i.e., classifying a video into one of the pre-defined action types, has been a popular topic in the communities of artificial intelligence, multimedia, and signal processing. However, existing methods usually consider an input video as a whole and learn models, e.g., Convolutional Neural Networks (CNNs), with coarse video-level class labels. These methods can only output an action class for the video, but cannot provide fine-grained and explainable cues to answer why the video shows a specific action. Therefore, researchers start to focus on a new task, Part-level Action Parsing (PAP), which aims to not only predict the video-level action but also recognize the frame-level fine-grained actions or interactions of body parts for each person in the video. To this end, we propose a coarse-to-fine framework for this challenging task. In particular, our framework first predicts the video-level class of the input video, then localizes the body parts and predicts the part-level action. Moreover, to balance the accuracy and computation in part-level action parsing, we propose to recognize the part-level actions by segment-level features. Furthermore, to overcome the ambiguity of body parts, we propose a pose-guided positional embedding method to accurately localize body parts. Through comprehensive experiments on a large-scale dataset, i.e., Kinetics-TPS, our framework achieves state-of-the-art performance and outperforms existing methods over 31.10% ROC score.

关键词： Circuits and systems Detectors Streaming media signal processing computational efficiency Convolutional neural networks Task analysis

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：