检索结果-内蒙古大学图书馆

您好，读者！请登录

内蒙古大学图书馆

首页
概况
党建
资源
服务
科研支持
- 论文收录引用证明
- 科技查新
知识产权
档案馆
帮助

咨询与建议

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

您的常用邮箱：*

您的手机号码：*

问题描述：

当前已输入0个字，您还可以输入200个字

全部搜索
期刊论文
图书
学位论文
标准
纸本馆藏
外文资源发现
数据库导航
超星发现

高级检索

时间限定

出版年份：

文献类型

图书期刊文献学位论文多媒体

馆藏选择

电子馆藏纸本馆藏

核心期刊

全部期刊 SCI 收录期刊 SSCI 收录期刊 EI 收录期刊 CSCD 收录期刊 CSSCI 收录期刊

语言

中文英文

文献类型

期刊文献图书学位论文标准纸本馆藏

帮助

文字说明：

T=题名（书名、题名），A=作者（责任者），K=主题词，P=出版物名称，PU=出版社名称，O=机构（作者单位、学位授予单位、专利申请人），L=中图分类号，C=学科分类号，U=全部字段，Y=年（出版发行年、学位年度、标准发布年）

检索规则说明：

AND代表“并且”；OR代表“或者”；NOT代表“不包含”；(注意必须大写,运算符两边需空一格)

检索范例：

范例一：(K=图书馆学 OR K=情报学) AND A=范并思 AND Y=1982-2016
范例二：P=计算机应用与软件 AND (U=C++ OR U=Basic) NOT K=Visual AND Y=2011-2016

分类表

所选分类

>> <<

限定检索结果

文献类型

4,743 篇 会议
55 册 图书
13 篇 期刊文献

馆藏范围

4,811 篇 电子文献
0 种 纸本馆藏

日期分布

学科分类号

3,133 篇 工学
- 2,794 篇 计算机科学与技术...
- 2,208 篇 软件工程
- 1,245 篇 信息与通信工程
- 740 篇 控制科学与工程
- 691 篇 生物工程
- 666 篇 光学工程
- 511 篇 电气工程
- 454 篇 生物医学工程（可授...
- 400 篇 电子科学与技术（可...
- 299 篇 网络空间安全
- 182 篇 安全科学与工程
- 163 篇 机械工程
- 160 篇 仪器科学与技术
- 159 篇 交通运输工程
- 152 篇 动力工程及工程热...
- 152 篇 化学工程与技术
- 142 篇 石油与天然气工程
1,643 篇 理学
- 709 篇 生物学
- 637 篇 物理学
- 604 篇 数学
- 242 篇 化学
- 180 篇 统计学（可授理学、...
827 篇 管理学
- 463 篇 图书情报与档案管...
- 443 篇 管理科学与工程(可...
- 119 篇 工商管理
433 篇 医学
- 363 篇 临床医学
- 246 篇 基础医学(可授医学...
- 214 篇 药学(可授医学、理...
- 105 篇 公共卫生与预防医...
125 篇 法学
- 109 篇 社会学
95 篇 农学
45 篇 经济学
44 篇 文学
33 篇 教育学
12 篇 军事学

主题

320 篇 image processing
312 篇 deep learning
230 篇 image segmentati...
191 篇 signal processin...
172 篇 image classifica...
167 篇 image enhancemen...
162 篇 computer vision
140 篇 image recognitio...
138 篇 convolutional ne...
136 篇 feature extracti...
125 篇 training
115 篇 machine learning
96 篇 computer applica...
83 篇 neural networks
83 篇 computational mo...
79 篇 visualization
78 篇 object detection
71 篇 natural language...
69 篇 semantics
68 篇 artificial intel...

机构

24 篇 chitkara univers...
17 篇 maharishi school...
15 篇 lovely professio...
15 篇 tianjin universi...
13 篇 university of ch...
12 篇 the islamic univ...
12 篇 beijing institut...
11 篇 school of comput...
11 篇 centre of interd...
10 篇 peng cheng labor...
10 篇 graphic era hill...
9 篇 national univers...
9 篇 sun yat-sen univ...
9 篇 school of comput...
9 篇 vivekananda glob...
9 篇 college of compu...
8 篇 harbin engineeri...
8 篇 school of comput...
8 篇 maharishi school...
8 篇 gla university d...

作者

13 篇 chauhan rahul
11 篇 yang yang
10 篇 anand vatsala
8 篇 kukreja vinay
8 篇 gupta rupesh
7 篇 luo xiaonan
7 篇 zhao yang
6 篇 bin wang
6 篇 pei jihong
6 篇 kumar ankit
6 篇 choudhury tanupr...
6 篇 zhu ce
6 篇 ce zhu
6 篇 kumar gaurav
6 篇 bing li
6 篇 li wei
5 篇 gill kanwarparta...
5 篇 zhiguo cao
5 篇 nagaraj p.
5 篇 singh kamred udh...

语言

4,762 篇 英文
48 篇 其他
17 篇 中文
1 篇 土耳其文

检索条件"任意字段=2023 IEEE International Conference on Image Processing and Computer Applications, ICIPCA 2023"

共 4811 条记录，以下是4781-4790 订阅

全选清除本页清除全部题录导出标记到"检索档案"

详细简洁

排序：

image Encryption Algorithm Based on Quantum Chaotic Mapping and Wavelet Transform 8th

Image Encryption Algorithm Based on Quantum Chaotic Mapping ...

引用

8th international conference on Cognitive Systems and Information processing, ICCSIP 2023

作者： Jinyu, Zheng Fei, Xiang Henan University of Science and Technology Henan Luoyang471023 China Henan International Joint Laboratory of Cyberspace Security Applications Henan Luoyang471023 China

ISBN: (纸本)9789819980178

Aiming at the problems of low security and slow encryption speed of traditional chaotic system, this paper designs a wavelet domain image encryption algorithm which combines quantum chaotic mapping and Lorenz hyper chaotic mapping. Firstly, the SHA-256 hash algorithm is used to generate the initial key, and the quantum chaotic sequence and Lorenz hyper chaotic sequence are generated iteratively. Then, the plaintext image matrix is decomposed by discrete wavelet, and the sub-band coefficient matrix of each component after decomposition is scrambled by quantum chaotic sequence, and then it is reconstructed by discrete wavelet to complete the scrambling. Finally, Lorenz hyper chaotic sequence is used to diffuse the scrambled images, and the cipher text images after secondary encryption are obtained and analyzed for security. The simulation results show that the chaotic sequence combined with quantum chaos and used for encryption can be associated with the plaintext image through SHA-256, and the way of wavelet domain scrambling can get a better encryption effect, good performance against all kinds of attacks, and has the characteristics of high security and fast computing speed. © 2024, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Discrete wavelet transforms

来源：评论

学校读者我要写书评

暂无评论

Wheel-Terrain Contact Geometry Estimation using the Aside-Wheel RGB-D Sensor over Soft Terrain

Wheel-Terrain Contact Geometry Estimation using the Aside-Wh...

引用

Robotics, Automation and Artificial Intelligence (RAAI), ieee international conference on

作者： Pengxuan Zhana Chen Yao Longteng Hu Feng Xue Peichen Wang Zheng Zhu Zhenzhong Jia Department of Mechanical and Energy Engineering Southern University of Science and Technology Shenzhen China Department of Mechanical and Energy Engineering Shenzhen Key Laboratory of Biomimetic Robotics and Intelligent Systems Southern University of Science and Technology (SUSTech) Shenzhen China SUSTech Guang-dong Provincial Key Laboratory of Human-Augmentation and Rehabilitation Robotics in Universities Shenzhen China Harbin of Institute of Technology (HIT) Harbin China College of Engineering University of Michigan Ann Arbor Michigan USA School of Electronics Harbin Institute of Technology Harbin China

ISBN: (数字)9798350307566

ISBN: (纸本)9798350307573

Wheeled mobile robots (WMRs) have exhibited notable competitiveness in various applications, including geo-graphic and planetary exploration. Nonetheless, when traversing on soft terrains, their wheels might encounter sinkage and slippage, leading to mission failure. Consequently, real-time estimation of contact parameters such as sinkage and slip ratio holds significant importance for rover operations. In this paper, we introduce a novel algorithm that employs only one aside-wheel RGB-D camera for the simultaneous detection of wheel sinkage and slip ratio on deformable terrains in real-time. We estimate sinkage through the depth module by processing point cloud data, and we estimate slip ratio through the color module with a sequence of image processing steps. Utilizing only one single camera reduces errors arising from sensor signal alignment and enhances detection accuracy. With a single-wheel testbed, we simulate wheel motion under different slip ratios and carry out numerous experiments to validate the effectiveness of the proposed algorithm. Our algorithm attains a detection frequency of up to 20Hz and exhibits outstanding precision. The relative sinkage error remains below 4%, while the relative slippage error is confined within 0.8%.

关键词： Accuracy image color analysis Robot vision systems Wheels Estimation Cameras Real-time systems

来源：评论

学校读者我要写书评

暂无评论

Group-Mix SAM: Lightweight Segmentation Solution for Industrial Implementations

Group-Mix SAM: Lightweight Segmentation Solution for Industr...

引用

Cognitive Computing and Complex Data (ICCD), international conference on the

作者： Liang Wu Puhong Duan Xiaoguang Ma Faculty of Robot Science and Engineering Northeastern University Shenyang China School of Robotics Hunan University Changsha China

ISBN: (数字)9798350352894

ISBN: (纸本)9798350352900

Since its emergence in early 2023, Segment Anything Model(SAM) has garnered significant academic interests and spawned many investigations from various perspectives due to its zero-shot generalization ability. However, the deployment of SAM in scenes that required real-time performance, such as industrial assembly liners, had severe challenges due to its heavy image encoder, whose parameters were as large as 632M. In this study, we proposed a lightweight segmentation solution for industrial assembly line applications, wherein the heavyweight image encoder was replaced with a lightweight one. Specifically, we employed decoupled distillation to train the encoder of MobileSAM in a resource-limited setting. The entire knowledge distillation experiment could be completed in a single day on a single RTX 4090. The resulting lightweight SAM, called Group- Mix SAM, had 37.63% (2.16M) fewer parameters and 42.5% (15614.7M) fewer floating-point operations compared to MobileSAM. Additionally, on our homemade industrial dataset: MALD, its mIoU was only marginally lower than that of MobileSAM, at 0.615. Finally, we conducted comprehensive experiments to demonstrate the superiority of Group-Mix SAM in real industrial scenes.

关键词： image segmentation Computational modeling computer architecture Real-time systems Data models Cognitive systems Assembly

来源：评论

学校读者我要写书评

暂无评论

MRI image based Ensemble Voting Classifier for Alzheimer's Disease Classification with Explainable AI Technique

MRI image based Ensemble Voting Classifier for Alzheimer's D...

引用

international conference for Convergence of Technology (I2CT)

作者： Uppin Rashmi Tripty Singh Sateesh Ambesange CSE Dept. Amrita School of Computing Bangalore Amrita Vishwa Vidyapeetham India Department of Computer Science & Engineering Amrita School of Computing Bangalore Amrita VishwaVidyapeetham India Pragyan AI Pragyan AI School Bangalore Pragyan SmartAI Technology LLP

Alzheimer's is one of the causes of dementia, which causes memory loss, problem-solving disability, speaking, and a lot more difficulties in day-to-day life. Generally, dementia is a loss of memory, problem-solving ability, language fluency, and other thinking abilities that severely affect day-to-day life. Alzheimer's creates a huge impact on family life, the economy, and finally, the country as a whole is affected. According to statistics every 3 seconds, one person develops dementia in the world and the estimates say that by 2030, 78 million people will be affected, and by 2050 139 million people will have dementia. Estimates say that the economic impact due to dementia by 2030 in the US will be $2.8 Trillion which causes a huge loss and needs to be ***'s can be diagnosed at various stages, with different datasets like Magnetic Resonance Imaging (MRI) Test images, Speech Tests, Symptoms, genes, and other data. Several models are developed to diagnose, but doctors expect proper insights about results apart from diagnosis, so the paper explains the results using various explainable methods like SHapley Additive exPlanations (SHAP) and Local Interpretable Model-agnostic Explanations (LIME).Data Sets used are MRI Features data extracted with generic information, Cross-sectional MRI data, and Longitudinal MRI Data. The step-by-step data processing includes data balancing using SMOTEENN, and then data transferred, using Quantile Transformer and PCA dimension reduction technique for 6 features, and Meta machine learning model, first level six key machine learning methods and finally voting classifier with hyperparameter tuning to get performance, 97.6 %, Precision 95.8%, recall 97.9% and finally F1 Score 96.8%.

关键词： Magnetic resonance imaging Biological system modeling Medical services Feature extraction Transformers Data models Problem-solving

来源：评论

学校读者我要写书评

暂无评论

Advancements in Video-Based Insect Tracking: A Bibliometric Analysis to A Short Survey 23

Advancements in Video-Based Insect Tracking: A Bibliometric ...

引用

Proceedings of the 2023 7th international conference on Advances in image processing

作者： Erick A. Adje Arnaud S. R. M. Ahouandjinou Gilles Delmaire Gilles Roussel Ratheil V. Houndji Unité de Recherche en Sciences Informatiques et Applications Université d'Abomey-Calavi Benin and Laboratoire d'Informatique Signal et Image de la Côte d'Opale Université du Littoral Cote d'Opale France Unité de Recherche en Sciences Informatiques et Applications Université d'Abomey-Calavi Benin Laboratoire d'Informatique Signal et Image de la Côte d'Opale Université du Littoral Cote d'Opale France

ISBN: (纸本)9798400708275

Video-based insect tracking provides vital insights into insect behavior and ecology, enhancing our understanding of their movements and interactions. Therefore, examining trends in this field over the last few years is essential. This study aims to conduct a bibliometric analysis to unveil the growing interest in video-based insect tracking with a short review based on documents used for bibliometric analysis. To achieve this, 453 documents were extracted from Scopus on 12 June 2023. Only documents in English published between 2010 and 2023, resulting in a dataset of 318 documents, were analyzed. The findings illustrate a consistent growth in video-based insect research over the last years, with a significant peak in 2021, comprising 32 documents. The journal PLOS ONE stands out as the most productive source. The USA exhibited the most significant interest in video-based insect tracking over the last years. Keyword analysis reflects the multidisciplinary nature of insect tracking research. The review demonstrated that video-based insect tracking serves two primary objectives: pose estimation and trajectory information. However, the main challenge in video-based insect tracking is to preserve the identity of multiple individuals in situations involving occlusions or complex interactions.

关键词： bibliometric analysis computer vision insect movement behavior video tracking

来源：评论

学校读者我要写书评

暂无评论

Manipulation of Motion Parallax Gain Distorts Perceived Distance and Object Depth in Virtual Reality

Manipulation of Motion Parallax Gain Distorts Perceived Dist...

引用

ieee Annual international Symposium Virtual Reality

作者： Xue Teng Robert S. Allison Laurie M. Wilcox Department of Electrical Engineering & Computer Science Centre for Vision Research York University Department of Psychology Centre for Vision Research York University

Virtual reality (VR) is distinguished by the rich, multimodal, im-mersive sensory information and affordances provided to the user. However, when moving about an immersive virtual world the vi-sual display often conflicts with other sensory cues due to design, the nature of the simulation, or to system limitations (for example impoverished vestibular motion cues during acceleration in racing games). Given that conflicts between sensory cues have been as-sociated with disorientation or discomfort, and theoretically could distort spatial perception, it is important that we understand how and when they are manifested in the user experience. To this end, this set of experiments investigates the impact of mismatch between physical and virtual motion parallax on the per-ception of the depth of an apparently perpendicular dihedral angle (a fold) and its distance. We applied gain distortions between visual and kinesthetic head motion during lateral sway movements and measured the effect of gain on depth, distance and lateral space compression. We found that under monocular viewing, observers made smaller object depth and distance settings especially when the gain was greater than 1. Estimates of target distance declined with increasing gain under monocular viewing. Similarly, mean set depth decreased with increasing gain under monocular viewing, except at 6.0 m. The effect of gain was minimal when observers viewed the stimulus binocularly. Further, binocular viewing (stereopsis) improved the precision but not necessarily the accuracy of gain perception. Overall, the lateral compression of space was similar in the stereoscopic and monocular test conditions. Taken together, our results show that the use of large presentation distances (at 6 m) combined with binocular cues to depth and distance enhanced humans' tolerance to visual and kinesthetic mismatch.

关键词： Visualization Solid modeling Three-dimensional displays Sensitivity Shape Stereo image processing Virtual reality

来源：评论

学校读者我要写书评

暂无评论

Scalable and Reliable Cloud-Based UV Monitoring for Public Health applications

Scalable and Reliable Cloud-Based UV Monitoring for Public H...

引用

Global conference for Advancement in Technology (GCAT)

作者： Dankan Gowda V Shekhar R Kdv Prasad Pullela Svvsr Kumar Sreeti Gangadharan Srividya C N Department of Electronics and Communication Engineering BMS Institute of Technology and Management Bangalore Karnataka Department of Computer Science and Engineering Alliance University Bangalore Karnataka India Symbiosis Institute of Business Management Hyderabad Symbiosis International (Deemed University) Pune India Department of Computer Science & Engineering Aditya College of Engineering Surampalem Andhra Pradesh India Department of Science and Humanities Mahaguru Institute of Technology Kerala India Department of Electronics and Communication Engineering BGS Institute of Technology Adichunchanagiri University (ACU) B.G Nagara Karnataka India

The creation and assessment of a scalable and trustworthy cloud-based UV monitoring system designed for public health applications are the main topics of this research article. Monitoring UV radiation is important because it plays a crucial part in understanding and reducing a variety of health hazards, such as vitamin D insufficiency and skin cancer. Traditional monitoring techniques frequently have limitations due to their confined character and lack of access to real-time data. We suggest a cloud-based UV monitoring system that integrates cutting-edge UV sensors with cloud computing infrastructure to overcome these constraints. The process entails placing UV sensors in various geographical places in order to collect UV radiation data in real time. Due to the seamless integration of these sensors with a cloud platform, continuous data transmission, archival, and analysis are possible. The raw sensor readings used in our study are transformed into meaningful UV index values using data processing techniques, giving us valuable information for public health initiatives. This study introduces a system for tracking the ultraviolet (UV) index that is based on the Internet of Things (IoT). The goal is to help avoid ailments that can be brought on by prolonged exposure to sun UV light. In reaction to variations in the UV index level over time, the system accomplishes this by sending warning messages and suggesting preventive steps. Monitoring the UV index is crucial for avoiding a number of illnesses, including calcium shortage, cardiovascular disease, and skin cancer. This study introduces a system for tracking the ultraviolet (UV) index that is based on the Internet of Things (IoT). The goal is to help avoid ailments that can be brought on by prolonged exposure to sun UV light. In reaction to variations in the UV index level over time, the system accomplishes this by sending warning messages and suggesting preventive steps. Monitoring the UV index is crucial for avoidin

关键词：

来源：评论

学校读者我要写书评

暂无评论

Towards a standard for PET raw data

Towards a standard for PET raw data

引用

ieee Symposium on Nuclear Science (NSS/MIC)

作者： K. Thielemans A. Kesner E. Asma J. Cabello M. Cook M. Hansen J. Jones N. A. Karakatsanis E. K. S. Leung P. Markiewicz S. Prevrhal A. Rahmim B. Saboury J. Stairs S. Stute H. Tashima G. Wells Institute for Nuclear Medicine University College London London Centre for Medical Image Computing University College London London United Kingdom Memorial Sloan Kettering Cancer Center New York Washington United States of America Canon Medical Research USA Vernon Hills Washington United States of America Siemens Medical Solutions USA Inc. Knoxville Tennessee United States of America GE HealthCare Waukesha Wisconsin United States of America Microsoft Research Redmond Washington United States of America Weill Cornell Medical College Cornell University New York New York United States of America UIH America Inc. Houston Texas United States of America Dept Medical Physics and Bioengineering University College London London United Kingdom Philips Research Hamburg Germany University of British Columbia Vancouver British Columbia Canada National Institutes of Health (NIH) Clinical Center Bethesda Maryland United States of America CRCI2NA Nantes University Hospital Nantes France National Institutes for Quantum Science and Technology Chiba Japan University of Ottawa Heart Institute Ottawa Ontario Canada

As the capabilities of computer hardware and software progresses, the importance of standardized data is becoming more relevant. While DICOM is widely used as a standard for PET images, no agreed format exists for PET raw data. In addition, all current vendor formats differ, to accommodate different architectures and processing *** international working group comprised of PET imaging experts from academia and industry was formed in 2022 to propose a new standard format. Efforts include establishing the key informational elements to include in the standard, data container formats, and integration initiatives. An important feature of the proposal is the use of a meta-language (called Yardl) for defining data structure and protocols for accessing, transferring, and storing data, allowing logical separation of data elements and container formats. All portions of the format, tools for accessing the data in the standardized format, and example data will be made open source and publicly *** this work, we will provide a status report of the progress of the working group. The effort is ongoing, and we welcome interest, feedback, and support from the greater community. We hope that this standard will facilitate a new paradigm for PET innovation, including new opportunities for inter-scanner and inter-vendor harmonization and AI applications, and aide in the advancement of novel PET applications and analysis tools.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Edge AI-Based Closed-Loop Peripheral Nerve Stimulation System for Gait Rehabilitation after Spinal Cord Injury

Edge AI-Based Closed-Loop Peripheral Nerve Stimulation Syste...

引用

international ieee/EMBS conference on Neural Engineering, CNE

作者： Ahnsei Shon Alex Stefanov Michelle Hook Hangue Park Department of Multidisciplinary Engineering Texas A&M University College Station TX USA Department of Neuroscience and Experimental Therapeutics Texas A&M University Bryan TX USA Department of Biomedical Engineering and Department of Intelligent Precision Healthcare Convergence Sungkyunkwan University Suwon South Korea Department of Electrical and Computer Engineering and the Department of Multidisciplinary Engineering Texas A&M University College Station TX USA

Recently, it has been revealed that Vsx2 neurons in the lumbar of the spinal cord are essential to restore locomotor function after spinal cord injury (SCI). However, Vsx2 neurons still need afferent feedback from peripheral nerves for locomotor rehabilitation. Also, movement-dependent electrical stimulation after SCI can solidify synaptic connections in a directed way. Based on these two facts, we hypothesized that providing movement-dependent electrical stimulation to hindlimb nerves may facilitate the rehabilitation process for restoring locomotion function after SCI while preventing aberrant remodeling of denervated spinal circuits. To evaluate our hypothesis, we developed an edge artificial intelligence (AI)-based closed-loop peripheral nerve stimulation system which can timely generate stimulation pulses for distal-tibial and peroneal nerves based on stance phase detection. The main parts of the system consist of a multi-site EMG electrode, neural amplifiers, an edge AI processing circuit, and neural stimulators. Medial gastrocnemius (MG) electromyography (EMG) was used as a input source of the AI model to detect the stance phase. The AI model was deployed on a dual-core 32-bit microprocessor. The whole system was evaluated with two SCI rats walking bipedally on a motorized treadmill. The accuracy of the presented AI model was calculated as 97.19%. In the animal experiments with SCI rats, stimulation pulses for the distal-tibial and peroneal nerves were timely generated for 200 ms and 100 ms, respectively based on the stance phase detected by the AI model. The experimental results suggest that the presented system can be a powerful neural interface tool to investigate the efficacy of edge AI-based closed-loop peripheral nerve stimulation on restoring locomotion function after SCI.

关键词： Electrodes image edge detection Neurons Rats Electrical stimulation Electromyography Recording

来源：评论

学校读者我要写书评

暂无评论

Prototyping of Low-Cost Configurable Sparse Neural processing Unit with Buffer and Mixed-Precision Reshapeable MAC Array

Prototyping of Low-Cost Configurable Sparse Neural Processin...

引用

international conference on Parallel and Distributed Systems (ICPADS)

作者： Binyi Wu Wolfgang Furtner Bernd Waschneck Christian Mayr Faculty of Electrical and Computer Engineering Technische Universit¨at Dresden Dresden Germany Division of Power and Sensor Systems Infineon Technologies Munich Germany Centre for Tactile Internet with Human-in-the-loop (CeTI) Technische Universität Dresden Dresden Germany

More recently, it has become possible to run deep learning algorithms on edge devices such as microcontrollers due to continuous improvements in neural network optimization algorithms such as quantization and neural architecture search. Nonetheless, most of the embedded hardware available today still falls short of the requirements of running deep neural networks. As a result, specialized processors have emerged to improve the inference efficiency of deep learning algorithms. However, most are not for edge applications that require efficient and low-cost hardware. Therefore, we design and prototype a low-cost configurable sparse Neural processing Unit (NPU). The NPU has a built-in buffer and a reshapable mixed-precision multiply-accumulator (MAC) array. The computing and memory resources of the NPU are parameterized, and different NPUs can be derived. Besides, users can also conFigure the NPU at runtime to fully utilize the resources. In our experiments, the 200MHz NPU with only 32 MACs is more than 32 times faster than the 400MHzSTM32H7 when inferring MobileNet-Vl. Besides, the yielded NPUs can achieve roofline or even beyond roofline performance. The buffer and reshapeable MAC array push the NPU’s attainable performance to the roofline, while the feature of supporting sparsity allows the NPU to obtain performance beyond the roofline.

关键词： Deep learning Runtime Quantization (signal) Program processors Neural networks Random access memory Prototypes

来源：评论

学校读者我要写书评

暂无评论

没有更多数据了...

全选清除本页清除全部题录导出标记到“检索档案”

共482页 << < 473 474 475 476 477 478 479 480 481 482 > >>

检索报告对象比较合并检索0

隐藏清空

合并搜索

回到顶部

执行限定条件

内容：

评分：

请选择保存的检索档案：

请选择收藏分类：

订阅名称：

通借通还

温馨提示：

图书名称：

借书校区：

取书校区：

手机号码：

邮箱地址：

一卡通帐号：

电话和邮箱必须正确填写，我们会与您联系确认。

联系人：

所在院系：

联系邮箱：

联系电话：

内蒙古自治区呼和浩特市赛罕区大学西街235号邮编: 010021

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：