检索结果-内蒙古大学图书馆

Robust 3D reconstruction of rib cage bones in computed tomography images, by combining knowledge from radiological, machine learning, and innovative graph guidance

引用

Research on Biomedical Engineering 2019年第1期35卷 45-56页

作者： Sais, Barbara Teixeira Vital, Daniel Aparecido Moraes, Matheus Cardoso Laboratory of Image and Signal Processing of the Institute of Science and Technology Federal University of São Paulo – UNIFESP 330 Talim St. room 108 Jardim Aeroporto CEP São José dos CamposSP12231-280 Brazil

Purpose: More than 70 million computed tomography scans are made per year. A great number of them aim at the thoraxic region, due to the number of organs and structures within it. The 3D visualization of these structures, including the bone, can lead to a more precise medical diagnosis. There are a number of works regarding 3D bone reconstruction, but most fail to present a quantitative evaluation of their assessment or have not achieved an assessment close to 100%. We present an automatic method of bone segmentation followed by 3D reconstruction that approaches these current limitations. methods: The proposed methodology has three blocks: (1) Preprocessing, whereby a median filter was applied to images that presented a high level of noise;(2) feature extraction procedure, in which (i) the images intensity levels were converted to attenuation coefficients and (ii) a (MLP) neural network was used to populate the Space of Attributes with the corresponding feature vectors;and (3) 3D structural construction, whereby a red-and-black tree with graph guidance combined the regarding clustered feature vectors with their spatial neighbors. To evaluate the results, the accuracy between the 2D-segmented images and their corresponding gold standards was calculated. Results: The material is composed of a set of 90 CT chest volumes. We obtained high accuracy parameters, such as Overlap Dice (%) = 98.77 ± 0.58 and False Negative (%) = 0.13 ± 0.026. Conclusion: The proposed methodology presented significant accuracy values and processing time, having a large database to test the methodology, besides being a fully automatic method of 3D bone reconstruction. © 2019, Sociedade Brasileira de Engenharia Biomedica.

关键词： image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Euclid preparation - XXiiI. Derivation of galaxy physical properties with deep machine learning using mock fluxes and H-band images

引用

MONTHLY NOTICES OF THE ROYAL ASTRONOMICAL SOCIETY 2023年第3期520卷 3529-3548页

作者： Bisigello, L. Conselice, C. J. Baes, M. Bolzonella, M. Brescia, M. Cavuoti, S. Cucciati, O. Humphrey, A. Hunt, L. K. Maraston, C. Pozzetti, L. Tortora, C. van Mierlo, S. E. Aghanim, N. Auricchio, N. Baldi, M. Bender, R. Bodendorf, C. Bonino, D. Branchini, E. Brinchmann, J. Camera, S. Capobianco, V. Carbone, C. Carretero, J. Castander, F. J. Castellano, M. Cimatti, A. Congedo, G. Conversi, L. Copin, Y. Corcione, L. Courbin, F. Cropper, M. Da Silva, A. Degaudenzi, H. Douspis, M. Dubath, F. Duncan, C. A. J. Dupac, X. Dusini, S. Farrens, S. Ferriol, S. Frailis, M. Franceschi, E. Franzetti, P. Fumana, M. Garilli, B. Gillard, W. Gillis, B. Giocoli, C. Grazian, A. Grupp, F. Guzzo, L. Haugan, S. V. H. Holmes, W. Hormuth, F. Hornstrup, A. Jahnke, K. Kuemmel, M. Kermiche, S. Kiessling, A. Kilbinger, M. Kohley, R. Kunz, M. Kurki-Suonio, H. Ligori, S. Lilje, P. B. Lloro, I. Maiorano, E. Mansutti, O. Marggraf, O. Markovic, K. Marulli, F. Massey, R. Maurogordato, S. Medinaceli, E. Meneghetti, M. Merlin, E. Meylan, G. Moresco, M. Moscardini, L. Munari, E. Niemi, S. M. Padilla, C. Paltani, S. Pasian, F. Pedersen, K. Pettorino, V. Polenta, G. Poncet, M. Popa, L. Raison, F. Renzi, A. Rhodes, J. Riccio, G. Rix, H-W Romelli, E. Roncarelli, M. Rosset, C. Rossetti, E. Saglia, R. Sapone, D. Sartoris, B. Schneider, P. Scodeggio, M. Secroun, A. Seidel, G. Sirignano, C. Sirri, G. Stanco, L. Tallada-Crespi, P. Tavagnacco, D. Taylor, A. N. Tereno, I. Toledo-Moreo, R. Torradeflot, F. Tutusaus, I. Valentijn, E. A. Valenziano, L. Vassallo, T. Wang, Y. Zacchei, A. Zamorani, G. Zoubian, J. Andreon, S. Bardelli, S. Boucaud, A. Colodro-Conde, C. Ferdinando, D. Di Gracia-Carpio, J. Lindholm, V. Maino, D. Mei, S. Scottez, V. Sureau, F. Tenti, M. Zucca, E. Borlaff, A. S. Ballardini, M. Biviano, A. Bozzo, E. Burigana, C. Cabanac, R. Cappi, A. Carvalho, C. S. Casas, S. Castignani, G. Cooray, A. Coupon, J. Courtois, H. M. Cuby, J. Davini, S. De Lucia, G. Desprez, G. Dole, H. Escartin, J. A. Escoffier, S. Farina, M. Fotopoulou, S. Ganga, K. Garcia-Bellid Univ Padua Dipartimento Fis & Astron G Galilei Via Marzolo 8 I-35131 Padua Italy INAF Osservatorio Astrofis & Sci Spazio Bologna Via Piero Gobetti 93-3 I-40129 Bologna Italy Univ Nottingham Sch Phys & Astron Univ Pk Nottingham NG7 2RD England Univ Manchester Dept Phys & Astron Jodrell Bank Ctr Astrophys Oxford Rd Manchester M13 9PL Lancs England Univ Ghent Sterrenkundig Observ Krijgslaan 281 S9 B-9000 Ghent Belgium INFN Sect Naples Via Cinthia 6 I-80126 Naples Italy INAF Osservatorio Astron Capodimonte Via Moiariello 16 I-80131 Naples Italy Univ Federico II Dept Phys E Pancini Via Cinthia 6 I-80126 Naples Italy Univ Porto Inst Astrofis & Ciencias Espaco CAUP Rua Estrelas P-4150762 Porto Portugal INAF Osservatorio Astrofis Arcetri Largo E Fermi 5 I-50125 Florence Italy Univ Portsmouth Inst Cosmol & Gravitat Portsmouth PO1 3FX Hants England Ist Nazl Astrofis INAF Osservatorio Astrofis & Sci Spazio OAS Via Gobetti 93-3 I-40127 Bologna Italy Univ Groningen Kapteyn Astron Inst POB 800 NL-9700 AV Groningen Netherlands Univ Paris Saclay CNRS Inst Astrophys Spatiale F-91405 Orsay France Univ Bologna Dipartimento Fis & Astron Via Gobetti 93-2 I-40129 Bologna Italy INFN Sez Bologna Viale Berti Pichat 6-2 I-40127 Bologna Italy Ludwig Maximilians Univ Munchen Fak Phys Univ Sternwarte Munchen Scheinerstr 1 D-81679 Munich Germany Max Planck Inst Extraterr Phys Giessenbachstr 1 D-85748 Garching Germany INAF Osservatorio Astrofis Torino Via Osservatorio 20 I-10025 Pino Torinese TO Italy Univ Genoa Dipartimento Fis INFN Sez Genova Via Dodecaneso 33 I-16146 Genoa Italy INFN Sez Roma Tre Via Vasca Navale 84 I-00146 Rome Italy Univ Torino Dipartimento Fis Via P Giuria 1 I-10125 Turin Italy INFN Sez Torino Via P Giuria 1 I-10125 Turin Italy INAF IASF Milano Via Alfonso Corti 12 I-20133 Milan Italy Barcelona Inst Sci & Technol Inst Fis Altes Energies IFAE Campus UAB Bellaterra 08193 Barcelona Spain Port Informacio Cient Campus UABC Albar

Next-generation telescopes, like Euclid, Rubin/LSST, and Roman, will open new windows on the Universe, allowing us to infer physical properties for tens of millions of galaxies. Machine-learning methods are increasingly becoming the most efficient tools to handle this enormous amount of data, because they are often faster and more accurate than traditional methods. We investigate how well redshifts, stellar masses, and star-formation rates (SFRs) can be measured with deep-learning algorithms for observed galaxies within data mimicking the Euclid and Rubin/LSST surveys. We find that deep-learning neural networks and convolutional neural networks (CNNs), which are dependent on the parameter space of the training sample, perform well in measuring the properties of these galaxies and have a better accuracy than methods based on spectral energy distribution fitting. CNNs allow the processing of multiband magnitudes together with H-E-band images. We find that the estimates of stellar masses improve with the use of an image, but those of redshift and SFR do not. Our best results are deriving (i) the redshift within a normalized error of <0.15 for 99.9 per cent of the galaxies with signal-to-noise ratio >3 in the H-E band;(ii) the stellar mass within a factor of two (similar to 0.3 dex) for 99.5 per cent of the considered galaxies;and (iii) the SFR within a factor of two (similar to 0.3 dex) for similar to 70 per cent of the sample. We discuss the implications of our work for application to surveys as well as how measurements of these galaxy parameters can be improved with deep learning.

关键词： galaxies: evolution galaxies: general galaxies: photometry galaxies: star formation

来源：评论

学校读者我要写书评

暂无评论

The Pitfalls of Simplicity Bias in neural Networks 20

The Pitfalls of Simplicity Bias in Neural Networks

引用

Proceedings of the 34th International Conference on neural Information processing Systems

作者： Harshay Shah Kaustav Tamuly Aditi Raghunathan Prateek Jain Praneeth Netrapalli Microsoft Research Stanford University

ISBN: (纸本)9781713829546

Several works have proposed Simplicity Bias (SB)—the tendency of standard training procedures such as stochastic Gradient Descent (SGD) to find simple models—to justify why neural networks generalize well [1, 49, 74]. However, the precise notion of simplicity remains vague. Furthermore, previous settings [67, 24] that use SB to justify why neural networks generalize well do not simultaneously capture the non-robustness of neural networks—a widely observed phenomenon in practice [71, 36]. We attempt to reconcile SB and the superior standard generalization of neural networks with the non-robustness observed in practice by introducing piecewise-linear and image-based datasets, which (a) incorporate a precise notion of simplicity, (b) comprise multiple predictive features with varying levels of simplicity, and (c) capture the non-robustness of neural networks trained on real data. Through theoretical analysis and targeted experiments on these datasets, we make four observations:(i) SB of SGD and variants can be extreme: neural networks can exclusively rely on the simplest feature and remain invariant to all predictive complex features.(ii) The extreme aspect of SB could explain why seemingly benign distribution shifts and small adversarial perturbations significantly degrade model performance.(iii) Contrary to conventional wisdom, SB can also hurt generalization on the same data distribution, as SB persists even when the simplest feature has less predictive power than the more complex features.(iv) Common approaches to improve generalization and robustness—ensembles and adversarial training—can fail in mitigating SB and its *** the role of SB in training neural networks, we hope that the proposed datasets and methods serve as an effective testbed to evaluate novel algorithmic approaches aimed at avoiding the pitfalls of SB.

关键词：

来源：评论

学校读者我要写书评

暂无评论

State-of-the-Arts Person Re-Identification using Deep Learning 6

State-of-the-Arts Person Re-Identification using Deep Learni...

引用

6th International Conference on signal processing and Integrated Networks (SPIN)

作者： Jaiswal, Shradha Vishwakarma, Dinesh Kumar Delhi Technol Univ Dept Informat Technol Delhi India

ISBN: (纸本)9781728113807

Person Re-Identification has become prominent because of various reasons majorly due to its high-performance methods based on deep-learning. It is the process of person recognition from various images captured by different cameras. Provided two set of images the purpose is to find that the given set of images are identical or not. Person Re-Id is often a challenging task due to the similarity in nature like people with identical features, color or clothes. images are taken from various angles and distances of a given subject in order to achieve high accuracy, so it identifies correctly. Re-Identification has broadly two major categories: i) image Re-ID and ii) Video Re-ID. Based on the category Re-ID has numerous applications like robotics, automated video surveillance, forensics and multimedia that are deployed using various public datasets like Market1501, VIPeR, MARS, CUHK01, CUHK02, CUHK03, DukeMTMC-reID, MSMT17 etc. In this paper we aim to briefly discuss the process, datasets, recent work on Re-ID, challenges, its approaches and techniques that has been implemented using deep learning systems.

关键词： Person Re-Identification deeply learned systems Convolutional neural Networks Deep Learning

来源：评论

学校读者我要写书评

暂无评论

Improving Alzheimer's stage categorization with Convolutional neural Network using transfer learning and different magnetic resonance imaging modalities

引用

HELIYON 2020年第12期6卷 e05652页

作者： Aderghal, Karim Afdel, Karim Benois-Pineau, Jenny Catheline, Gwenaelle Univ Bordeaux CNRS Bordeaux INP LaBRIUMR 5800 F-33400 Talence France Ibn Zohr Univ Dept Comp Sci Fac Sci LabSIV Agadir Morocco Univ Bordeaux CNRS UMR 5287 Inst Neurosci Cognit & Integrat Aquitaine INCIA Bordeaux France

Background: Alzheimer's Disease (AD) is a neurodegenerative disease characterized by progressive loss of memory and general decline in cognitive functions. Multi-modal imaging such as structural MRI and DTI provide useful information for the classification of patients on the basis of brain biomarkers. Recently, CNN methods have emerged as powerful tools to improve classification using images. New Method: In this paper, we propose a transfer learning scheme using Convolutional neural Networks (CNNs) to automatically classify brain scans focusing only on a small ROI: e.g. a few slices of the hippocampal region. The network's architecture is similar to a LeNet-like CNN upon which models are built and fused for AD stage classification diagnosis. We evaluated various types of transfer learning through the following mechanisms: (i) cross-modal (sMRI and DTI) and (ii) cross-domain transfer learning (using MNIST) (iii) a hybrid transfer learning of both types. Results: Our method shows good performances even on small datasets and with a limited number of slices of small brain region. It increases accuracy with more than 5 points for the most difficult classification tasks, i.e., AD/MCI and MCI/NC. Comparison with Existing Method(s): Our methodology provides good accuracy scores for classification over a shallow convolutional network. Besides, we focused only on a small region;i.e., the hippocampal region, where few slices are selected to feed the network. Also, we used cross-modal transfer learning. Conclusions: Our proposed method is suitable for working with a shallow CNN network for low-resolution MRI and DTI scans. It yields to significant results even if the model is trained on small datasets, which is often the case in medical image analysis.

关键词： Alzheimer's Disease Magnetic Resonance Imaging (MRI) Diffusion Tensor Imaging (DTI) Multi-modality image classification Convolutional neural Network (CNN) Transfer learning Applied computing Applied computing in medical science Computing methodology Artificial intelligence signal processing image processing Medical imaging

来源：评论

学校读者我要写书评

暂无评论

A Generalized stochastic Implementation of the Disparity Energy Model for Depth Perception

引用

JOURNAL OF signal processing SYSTEMS FOR signal image AND VIDEO TECHNOLOGY 2018年第5期90卷 709-725页

作者： Boga, Kaushik Leduc-Primeau, Francois Onizawa, Naoya Matsumiya, Kazumichi Hanyu, Takahiro Gross, Warren J. McGill Univ Dept Elect & Comp Engn Montreal PQ Canada Tohuku Univ Sendai Miyagi Japan

Implementing neuromorphic algorithms is increasingly interesting as the error resilience and low-area, low-energy nature of biological systems becomes the potential solution for problems in robotics and artificial intelligence. While conventional digital methods are inefficient in implementing massively parallel systems, analog solutions are hard to design and program. stochastic Computing (SC) is a natural bridge that allows pseudo-analog computations in the digital domain using low complexity hardware. However, large scale SC systems traditionally suffered from long latencies, hence higher energy consumption. This work develops a VLSI architecture for an SC based binocular vision system based on a disparity-energy model that emulates the hierarchical multi-layered neural structure in the primary visual cortex. The 3-layer neural network architecture is biologically plausible and is tuned to detecting 5 different disparities. The architecture is compact, adder-free, and achieves better disparity detection compared to a floating-point version by using a modified disparity-energy model. A generalized 1x100 pixel processing system is synthesized using TSMC 65nm CMOS technology and it achieves 71 % reduction in area-delay product and 48 % in energy savings compared to a fixed-point implementation at equivalent precision.

关键词： stochastic computing Neuromorphic computing Approximate computing Gabor filters Disparity-energy model Computer vision Biomedical electronics neural networks

来源：评论

学校读者我要写书评

暂无评论

On the vulnerability of deep learning to adversarial attacks for camera model identification

引用

signal processing-image COMMUNICATION 2018年 65卷 240-247页

作者： Marra, F. Gragnaniello, D. Verdoliva, L. Univ Federico II Naples Naples Italy

Camera model identification is a fundamental task for many investigative activities, and is drawing great attention in the research community. In this context, convolutional neural networks (CNN) are expected to provide a significant performance gain over the current state of the art, as already happened for a wide range of image processing applications. However, recent studies enlightened the vulnerability of CNNs to adversarial attacks, casting shadows on their reliability for critical applications. In this paper, we investigate the robustness to adversarial attacks of CNN-based methods for camera model identification. Several networks and attack methods are considered, both when the attacker has complete knowledge of the network and when only the training set is available. In addition, the analysis concerns both original and JPEG compressed images, to simulate a social network environment. The experiments, carried out on a publicly available dataset with images coming from 29 different camera models, shed some light on the suitability of CNN-based approaches for this task.

关键词： Camera model identification Convolutional neural networks Adversarial attacks

来源：评论

学校读者我要写书评

暂无评论

Comparison of Facial Emotion Recognition Based on image Visual Features and EEG Features 4th

Comparison of Facial Emotion Recognition Based on Image Visu...

引用

4th International Conference on Cognitive Systems and Information processing (ICCSIP)

作者： Long, Yanfang Wanzeng, Kong B. Ling, Wenfen Yang, Can Zhu, Jieyong Hangzhou Dianzi Univ Hangzhou 310018 Zhejiang Peoples R China Hong Kong Univ Sci & Technol Hong Kong Peoples R China

ISBN: (纸本)9789811379864;9789811379857

Automatic facial emotion recognition plays an important role in human-computer interaction. Although humans can recognize emotions with little or no effort, reliable emotion recognition by machines is always a challenge. To explore the doubt that whether machines can discriminate better than humans or not, we proposed two different ideas about facial emotion recognition. One is based on image visual features, the other is based on EEG signals which were recorded when the subject is watching facial emotion pictures. Correspondingly, the Deep Convolutional neural Network (DCNN) model is adopted to enable the machine to learn visual features from facial emotion pictures automatically. The Gated Recurrent Unit (GRU) model is used to extract specific emotional EEG features from EEG signals. These two methods were verified on the Chinese Facial Affective Picture System (CFAPS) and our developed Emotion EEG data (EMOT), and the recognition performance based on EEG features was found to be significantly better than the image visual.

关键词： Deep learning EEG Facial emotion recognition

来源：评论

学校读者我要写书评

暂无评论

stochastic Fusion for Multi-stream neural Network in Video Classification

Stochastic Fusion for Multi-stream Neural Network in Video C...

引用

Asia-Pacific signal and Information processing Association Annual Summit and Conference (APSIPA)

作者： Yu-Min Huang Huan-Hsin Tseng Jen-Tzung Chien National Chiao Tung University Hsinchu Taiwan Department of Radiation Oncology University of Michigan Ann Arbor MI USA

ISBN: (数字)9781728132488

ISBN: (纸本)9781728132495

Spatial image and optical flow provide complementary information for video representation and classification. Traditional methods separately encode two stream signals and then fuse them at the end of streams. This paper presents a new multi-stream recurrent neural network where streams are tightly coupled at each time step. Importantly, we propose a stochastic fusion mechanism for multiple streams of video data based on the Gumbel samples to increase the prediction power. A stochastic backpropagation algorithm is implemented to carry out a multi-stream neural network with stochastic fusion based on a joint optimization of convolutional encoder and recurrent decoder. Experiments on UCF101 dataset illustrate the merits of the proposed stochastic fusion in recurrent neural network in terms of interpretation and classification performance.

关键词： Streaming media Optical imaging stochastic processes Decoding Feature extraction Convolution Optical feedback

来源：评论

学校读者我要写书评

暂无评论

Improvement of Edge-Tracking methods using Genetic Algorithm and neural Network

Improvement of Edge-Tracking Methods using Genetic Algorithm...

引用

International Conference of signal processing and Intelligent Systems (ICSPIS)

作者： Sajjad Ghazanfari Shabankareh Saeid Ghazanfari Shabankareh Faculty of Electrical Engineering Islamic Azad University Shiraz Iran Faculty of Science and Technology Aix-Marseille University Marseille France

ISBN: (数字)9781728153506

ISBN: (纸本)9781728153513

One of the most basic and important operations in the field of image processing is image extraction and detection. Edge recognition is very important for image clarity and image segmentation. The importance of edge detection is that the human eye can recognizes the existence of different objects by observing the edges. So it makes sense to do edge detection before interpreting images in automated systems. Edges are points in the image where the two pixels are two different values, or two values very large in numerical value. It can be said that one of the goals of edge reduction in data size in images is to preserve the original structure and shape of the images. Editing has various applications, including object recognition, segmentation and image coding, in a variety of medical images. One of the problems we encounter when editing images has noises. In this paper, a combination of several standard edge- matching algorithms, neural network and multi-objective genetic algorithm NonDominated Sorting Genetic Algorithm (NSGA-ii) is used to edge detection. In the proposed method, we give the standard multiedge finder to a forward propagating (FP) neural network as input. A multi-objective genetic algorithm with Non- Dominated Sorting Genetic Algorithm (NSGA-ii was used to select the number of edge detectors to access the neural network and to select the most accurate ones. The genetic algorithm selects the least accurate number of edge detectors to enter the neural network. To evaluate the proposed method, TP, TN, FP, FN criteria were used to compare with other methods. Finally, by comparing the results obtained on different images using existing methods and the proposed method it is observed that the proposed method has better accuracy in detecting the edge of noise images.

关键词： image edge detection Detectors signal processing algorithms Genetic algorithms Biological neural networks Sociology

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：