检索结果-内蒙古大学图书馆

Demonstrating Advanced Machine Learning and Neuromorphic computing Using IBM’s NS16e 1

Science and Information Conference, SAI 2020

作者： Barnell, Mark Raymond, Courtney Wilson, Matthew Isereau, Darrek Cote, Eric Brown, Dan Cicotta, Chris Air Force Research Laboratory Information Directorate RomeNY13441 United States SRC Inc. 6225 Running Ridge Road North SyracuseNY13212 United States

ISBN: (数字)9783030522490

ISBN: (纸本)9783030522483

The human brain can be viewed as an extremely power-efficient biological computer. As such, there have been many efforts to create brain-inspired processing systems to enable advances in low-power data processing. An example of brain-inspired processing architecture is the IBM TrueNorth Neurosynaptic System, a Spiking neural network architecture for deploying ultra-low power machine learning (ML) models and algorithms. For the first time ever, an advanced scalable computing architecture was demonstrated using 16 TrueNorth neuromorphic processors containing in aggregate over 16 million neurons. This system, called the NS16e, was used to demonstrate new ML techniques including the exploitation of optical and radar sensor data simultaneously, while consuming a fraction of the power compared to traditional Von Neumann computing architectures. The number of applications that have requirements for computing architectures that can operate in size, weight and power-constrained environments continues to grow at an outstanding pace. These applications include processors for vehicles, homes, and real-time data exploitation needs for intelligence, surveillance, and reconnaissance missions. This research included the successful exploitation of optical and radar data using the NS16e system. Processing performance was assessed, and the power utilization was analyzed. The NS16e system never used more than 15 W, with the contribution from the 16 TrueNorth processors utilizing less than 5 W. The image processing throughput was 16,000 image chips per second, corresponding to 1,066 image chips per second for each watt of power consumed. © 2020, Springer Nature Switzerland AG.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

A Deep Learning Approach to Short-Term Quantitative Precipitation Forecasting 10

A Deep Learning Approach to Short-Term Quantitative Precipit...

引用

10th International Conference on Climate Informatics (CI)

作者： Yadav, Nishant Ganguly, Auroop R. Northeastern Univ Sustainabil & Data Sci SDS Lab Boston MA 02115 USA

ISBN: (纸本)9781450388481

Short-term, spatially distributed quantitative precipitation forecasting (SD-QPF), or 'precipitation nowcasting', is important for hydrological and water resources applications, such as flash flood warning systems and operations of dams and reservoirs. The stateof-the-art methods in SD-QPF include radar extrapolations, numerical weather prediction (NWP) models, and hybrid methods that combine the two. Despite the diversity of methods that have been used, SD-QPF remains difficult: even sophisticated methods may not be able to consistently outperform relatively simple baselines such as Persistence. Methods in Deep Learning (DL) have demonstrated significant and often unexpected improvements across a wide variety of domains ranging from image and video processing to machine translation and speech recognition. Emerging research has suggested that that DL may improve point predictions in the context of very short-term - 0-2 hours - distributed QPF (VSD-QPF) by taking advantage of growing data from in-situ weather sensors as well as remote sensors such as radar and satellites, along with advances in computing. Here we examine the hypothesis that DL can improve VSD-QPF, specifically point predictions, based on observed hourly precipitation data over the contiguous United States, by leveraging a Convolutional Long Short-Term Memory (LSTM) recurrent neural network for 1-hour precipitation nowcasting. We find the DL approach performs better than the baseline method of Persistence and a state-of-the-art method using optical Flow.

关键词： deep learning optical flow precipitation forecasting

来源：评论

学校读者我要写书评

暂无评论

基于边缘节点的深度神经网络任务分配方法

引用

计算机工程与设计 2021年第1期42卷 113-121页

作者：陈明浩陈庆奎上海理工大学光电信息与计算机工程学院上海200093

随着边缘计算的发展,边缘节点的计算规模不断增加,现有的边缘设备难以搭载深度神经网络模型,网络通信与云端服务器承受着巨大压力。为解决上述问题,通过对Roofline模型进行改进,借助新模型对边缘设备的性能与网络环境进行动态评估。根... 详细信息

随着边缘计算的发展,边缘节点的计算规模不断增加,现有的边缘设备难以搭载深度神经网络模型,网络通信与云端服务器承受着巨大压力。为解决上述问题,通过对Roofline模型进行改进,借助新模型对边缘设备的性能与网络环境进行动态评估。根据评估指标,对神经网络模型进行分离式拆分,部分计算任务分配给边缘节点完成,云端服务器结合节点返回数据完成其它任务。该方法基于节点自身性能与网络环境,进行动态任务分配,具有一定兼容性与鲁棒性。实验结果表明,基于边缘节点的深度神经网络任务分配方法可在不同环境中利用设备的闲置性能,大幅度降低中心服务器的计算负载。

关键词：边缘计算深度神经网络分层模型网络通信任务分配方法

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Based on Ensemble to Diagnose of Retinal Disease using optical Coherence Tomography

Deep Learning Based on Ensemble to Diagnose of Retinal Disea...

引用

International Conference on Computational Science and Computational Intelligence (CSCI)

作者： Kuntha Pin Yunyoung Nam Sangho Ha Jung Woo Han Department of ICT Convergence Soonchunhyang University Asan Korea Department of Computer Science and Engineering Soonchunhyang University Asan Korea Department of Ophthalmology Bucheon Hospital Soonchunhyang University College of Medicine Bucheon Korea

ISBN: (纸本)9781665458429

optical coherence tomography (OCT) image plays an essential modality role in medical image analysis to diagnose various retinal diseases. Through the retinal cross-sectional OCT images, ophthalmologists need much time and effort in manual analysis to extract diagnostic features. In this study, we propose a method for the automatic diagnosis of five retinal diseases based on an ensemble of two transfer learning models to analyze OCT images. The individual predicted class probabilities of MobileNetV3Large are fused with the predicted class probabilities of ResNet50 to ensure robustness in the prediction. An image processing technique such as contrast limited adaptive histogram equalization (CLAHE) is applied to enhance the quality of OCT images before feeding them to neural network models. In addition, 5-fold cross-validation and EarlyStopping techniques are also used to prevent the overfitting of the training model. To conduct the experiment, a dataset containing 1999 OCT images is constructed.

关键词： Training Adaptation models Scientific computing Veins optical coherence tomography Computational modeling Transfer learning

来源：评论

学校读者我要写书评

暂无评论

High-Performance Object Detection for optical Remote Sensing Images with Lightweight Convolutional neural networks 22

High-Performance Object Detection for Optical Remote Sensing...

引用

22nd IEEE International Conference on High Performance computing and Communications, 18th IEEE International Conference on Smart City and 6th IEEE International Conference on Data Science and Systems, HPCC-SmartCity-DSS 2020

作者： Liu, Jia Liu, Renhua Ren, Kaijun Li, Xiaoyong Xiang, Jianjian Qiu, Shaohua China University of Geosciences School of Computer Science Wuhan China National University of Defense Technology College of Meteorology and Oceanography Changsha China National Key Laboratory of Science and Technology on Vessel Integrated Power System Naval University of Engineering Wuhan China

ISBN: (纸本)9781728176499

Convolutional neural network (CNN)-based object detection for optical remote sensing images has achieved higher accuracy compared with traditional detection methods with handcrafted features. However, the deep and large CNNs make it hard to be deployed in real-Time scenarios with limited computation, storage, power and bandwidth resources, for example, data processing onboard airborne, satellites and unmanned aerial vehicles for search and rescue. Therefore, in this paper we present a high-performance object detection approach for optical remote sensing images. Based on the widely used Faster R-CNN framework, we integrate state-of-The-Art lightweight CNNs as backbone to extract features, slim the heavy-head architecture of two-stage detector by reducing dimensions of features, and fine-Tune our models with NWPU VHR-10 optical remote sensing dataset. Besides, the multi-Threading for CPU and detection in batches for GPU are deployed to enhance the throughput of detectors and utilization of multi-core CPU and many-core GPU. Experiments show that our presented detection approach can significantly reduce the model size, computation complexity and detection time, while maintaining competitive accuracy. Specifically, the one with ShuffleNet-v2 and slimmed features has achieved a highest mean average precision of 94.39%, a lowest computational complexity of 18.97 Giga floating point operations, a highest detection speed of 90.10 frames per second (fps) for GPU and 3.07 fps for CPU, corresponding to speedups of 6.94X and 13.26X compared with the baseline on the benchmarked system, with a model size of 13.6 MB. Moreover, it further improves the efficiency and achieves 4.98 fps on CPU with 4 threads, and 200.24 fps on GPU with a batch size of 32. © 2020 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

SceneGATE: Scene-Graph Based Co-Attention networks for Text Visual Question Answering

arXiv

引用

arXiv 2022年

作者： Cao, Feiqi Luo, Siwen Nunez, Felipe Wen, Zean Poon, Josiah Han, Soyeon Caren School of Computer Science Faculty of Engineering University of Sydney CamperdownNSW2006 Australia Department of Computer Science and Software Engineering School of Physics Maths and Computing University of Western Australia CrawleyWA6009 Australia

Visual Question Answering (VQA) models fail catastrophically on questions related to the reading of text-carrying images. However, TextVQA aims to answer questions by understanding the scene texts in an image-question context, such as the brand name of a product or the time on a clock from an image. Most TextVQA approaches focus on objects and scene text detection, which are then integrated with the words in a question by a simple transformer encoder. The focus of these approaches is to use shared weights during the training of a multi-modal dataset, but it fails to capture the semantic relations between an image and a question. In this paper, we proposed a Scene Graph-Based Co-Attention network (SceneGATE) for TextVQA, which reveals the semantic relations among the objects, the optical Character Recognition (OCR) tokens and the question words. It is achieved by a TextVQA-based scene graph that discovers the underlying semantics of an image. We create a guided-attention module to capture the intra-modal interplay between the language and the vision as a guidance for inter-modal interactions. To permit explicit teaching of the relations between the two modalities, we propose and integrate two attention modules, namely a scene graph-based semantic relation-aware attention and a positional relation-aware attention. We conduct extensive experiments on two widely used benchmark datasets, Text-VQA and ST-VQA. It is shown that our SceneGATE method outperforms existing ones because of the scene graph and its attention modules. Copyright © 2022, The Authors. All rights reserved.

关键词： neural networks

来源：评论

学校读者我要写书评

暂无评论

Cascading Convolutional neural network for Steel Surface Defect Detection

Cascading Convolutional Neural Network for Steel Surface Def...

引用

10th Int Conf on Appl Human Factors and Ergon (AHFE) / AHFE Int Conf Human Factors in Artificial Intelligence and Social Comp / AHFE Int Conf on Human Factors, Software, Serv and Syst Engn / AHFE Int Conf of Human Factors in Energy

作者： Lin, Chih-Yang Chen, Cheng-Hsun Yang, Ching-Yuan Akhyar, Fityanul Hsu, Chao-Yung Ng, Hui-Fuang Yuan Ze Univ Dept Elect Engn Taoyuan Taiwan China Steel Corp Automat & Instrumentat Syst Dev Sect Kaohsiung Taiwan Univ Tunku Abdul Rahman Dept Comp Sci FICT Petaling Jaya Malaysia

ISBN: (纸本)9783030204549;9783030204532

Steel is the most important material in the world of engineering and construction. Modern steelmaking relies on computer vision technologies, like optical cameras to monitor the production and manufacturing processes, which helps companies improve product quality. In this paper, we propose a deep learning method to automatically detect defects on the steel surface. The architecture of our proposed system is separated into two parts. The first part uses a revised version of single shot multibox detector (SSD) model to learn possible defects. Then, deep residual network (ResNet) is used to classify three types of defects: Rust, Scar, and Sponge. The combination of these two models is investigated and discussed thoroughly in this paper. This work additionally employs a real industry dataset to confirm the feasibility of the proposed method and make sure it is applicable to real-world scenarios. The experimental results show that the proposed method can achieve higher precision and recall scores in steel surface defect detection.

关键词： Fully convolutional networks Defect detection SSD ResNet

来源：评论

学校读者我要写书评

暂无评论

Flow-based video segmentation for human head and shoulders

arXiv

引用

arXiv 2021年

作者： Kuang, Zijian Tie, Xinran Department of Computing Science University of Alberta Edmonton Canada

Video segmentation for the human head and shoulders is essential in creating elegant media for videoconferencing and virtual reality applications. The main challenge is to process high-quality background subtraction in a real-time manner and address the segmentation issues under motion blurs, e.g., shaking the head or waving hands during conference video. To overcome the motion blur problem in video segmentation, we propose a novel flow-based encoder-decoder network (FUNet) that combines both traditional Horn-Schunck optical-flow estimation technique and convolutional neural networks to perform robust real-time video segmentation. We also introduce a video and image segmentation dataset: ConferenceVideoSegmentationDataset. Code and pre-trained models are available on our GitHub repository: https://***/kuangzijian/Flow-Based-Video-Matting. Copyright © 2021, The Authors. All rights reserved.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Temporal Interlacing network 34

Temporal Interlacing Network

引用

34th AAAI Conference on Artificial Intelligence / 32nd Innovative Applications of Artificial Intelligence Conference / 10th AAAI Symposium on Educational Advances in Artificial Intelligence

作者： Shao, Hao Qian, Shengju Liu, Yu Tsinghua Univ Beijing Peoples R China SenseTime X Lab Hong Kong Peoples R China Chinese Univ Hong Kong Hong Kong Peoples R China

ISBN: (纸本)9781577358350

For a long time, the vision community tries to learn the spatio-temporal representation by combining convolutional neural network together with various temporal models, such as the families of Markov chain, optical flow, RNN and temporal convolution. However, these pipelines consume enormous computing resources due to the alternately learning process for spatial and temporal information. One natural question is whether we can embed the temporal information into the spatial one so the information in the two domains can be jointly learned once-only. In this work, we answer this question by presenting a simple yet powerful operator - temporal interlacing network (TIN). Instead of learning the temporal features, TIN fuses the two kinds of information by interlacing spatial representations from the past to the future, and vice versa. A differentiable interlacing target can be learned to control the interlacing process. In this way, a heavy temporal model is replaced by a simple interlacing operator. We theoretically prove that with a learnable interlacing target, TIN performs equivalently to the regularized temporal convolution network (r-TCN), but gains 4% more accuracy with 6x less latency on 6 challenging benchmarks. These results push the state-of-the-art performances of video understanding by a considerable margin. Not surprising, the ensemble model of the proposed TIN won the 1st place in the ICCV19 - Multi Moments in Time challenge. Code is made available to facilitate further research.(1)

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Predicting PON networking Traffic Flow Based on LSTM neural network with Periodic Characteristic Data

Predicting PON Networking Traffic Flow Based on LSTM Neural ...

引用

Optoelectronics Global Conference (OGC)

作者： ZiYao Yang Dezhi Zhang Jian Tang China Telecom Research Institute Shanghai China

ISBN: (数字)9781728198606

ISBN: (纸本)9781728198613

PON(Passive optical network)traffic prediction can provide data base for port expansion and bandwidth dynamic adjustment, so as to simplify PON traffic operation and improve bandwidth utilization. In this paper, based on the LSTM(Long Short-Term Memory)neural network, the characteristic data is redesigned based on the periodic characteristics of the PON port traffic. Compared with ARIMA(Autoregressive Integrated Moving Average model)and the basic LSTM neural network, the prediction accuracy is significantly improved and the calculation time is reduced.

关键词： Predictive models Data models Passive optical networks Telecommunications Correlation optical computing Image motion 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：