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

4,753 篇 会议
55 册 图书
11 篇 期刊文献

馆藏范围

4,819 篇 电子文献
0 种 纸本馆藏

日期分布

学科分类号

3,161 篇 工学
- 2,819 篇 计算机科学与技术...
- 2,231 篇 软件工程
- 1,266 篇 信息与通信工程
- 762 篇 控制科学与工程
- 691 篇 生物工程
- 670 篇 光学工程
- 530 篇 电气工程
- 456 篇 生物医学工程（可授...
- 419 篇 电子科学与技术（可...
- 301 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
- 153 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
- 636 篇 物理学
- 605 篇 数学
- 242 篇 化学
- 181 篇 统计学（可授理学、...
830 篇 管理学
- 466 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
436 篇 医学
- 365 篇 临床医学
- 248 篇 基础医学(可授医学...
- 214 篇 药学(可授医学、理...
- 108 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
232 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
6 篇 shin seokjoo
5 篇 gill kanwarparta...
5 篇 zhiguo cao
5 篇 nagaraj p.

语言

4,709 篇 英文
109 篇 其他
21 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 4819 条记录，以下是4321-4330 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：

Denoising Diffusion Algorithm for Single image Inplaine Super-Resolution in CBCT Scans of the Mandible

Denoising Diffusion Algorithm for Single Image Inplaine Supe...

引用

international Scientific conference on Information Technology and Management Science of Riga Technical University (ITMS)

作者： Ivars Namatevs Kaspars Sudars Arturs Nikulins Anda Slaidina Laura Neimane Oskars Radzins Edgars Edelmers Riga Technical University & Institute of Electronics and Computer Science Riga Latvia Institute of Electronics and Computer Science Riga Latvia Department of Conservative Dentistry and Oral Health Institute of Stomatology Rīga Stradiņš University Riga Latvia Department of Prosthetic Dentistry Institute of Stomatology Rīga Stradiņš University Riga Latvia Faculty of Medicine Institute of Electronics and Computer Science Rīga Stradiņš University Riga Latvia

Deep Learning models are currently the cornerstone of artificial intelligence in medical imaging. The performance of Deep Learning in medical imaging is significantly influenced by the amount and quality of training data. Diffusion models have recently attracted the attention of the computer vision community as they enable photorealistic synthetic image-to-image translation. Previous attempts to use diffusion models for super-resolution imaging have produced satisfactory high-resolution images from low-resolution inputs. However, the drawback is the slow speed of inference, which severely hinders practical applications in medicine. To speed up inference and further improve performance, we propose an accelerated algorithm based on denoising diffusion probability modelling approach for medical image super-resolution. Instead of sampling from pure Gaussian noise, the intermediate distributions of noisy low- and high-resolution images are compared and used to generate super-resolution images. Our proposed algorithm is used to convert low-resolution panoramic X-ray images from Cone-beam Computed Tomography scans of the mandible into high-resolution images for the identification of osteoporosis.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Recent Developments in the Quality Evaluation of Rice Disease Detection System

Recent Developments in the Quality Evaluation of Rice Diseas...

引用

Advances in Computing, Communication Control and Networking (ICAC3N), international conference on

作者： Pardeep Seelwal Tilak Raj Rohilla Department of Computer Science and Applications Baba Mastnath University Rohtak India

ISBN: (数字)9798350330861

ISBN: (纸本)9798350330878

The recent progress in the field of computer vision has facilitated the implementation of diverse applications across several domains of human existence. One notable domain of application pertains to the classification of plant diseases, with a particular focus on rice plants. However, the classification of rice plants has shown itself as a multifaceted challenge, necessitating additional advancements in this field. The classification of rice diseases poses considerable difficulties as a result of similarities between different classes and the irregular characteristics within each class. The careful selection of data acquisition sensors and the adoption of an appropriate feature representation approach are of utmost importance in light of the extensive range of applications in this field. In recent years, there has been a growing utilisation of computer vision techniques for the purpose of evaluating food quality. This paper provides an overview of the current advancements in techniques and their application in conjunction with diverse sensing devices for the purpose of detecting plant problems. The identification and resolution of significant challenges related to defect detection in rice plants entail different technical obstacles. This study provides an overview of the latest developments in computer vision methodologies used for assessing the quality of rice plants. This study showcases the potential of image processing techniques in the assessment of rice plant quality. Additionally, it highlights several unresolved difficulties that require additional investigation to facilitate the widespread implementation of image processing technology for evaluating the quality of rice plants.

关键词： computer vision Plant diseases image resolution Plants (biology) Data acquisition Process control Machine learning

来源：评论

学校读者我要写书评

暂无评论

An Approach for Detecting and Restoring Tampering in Digital image Watermarking 5th

An Approach for Detecting and Restoring Tampering in Digital...

引用

5th international conference on Cognitive Computing and Information processing, CCIP 2023

作者： A. C., Pavan Somashekar, M.T. Department of Computer Science and Applications Bangalore University Bengaluru India

ISBN: (纸本)9783031607240

In addition to ensuring image authenticity, watermarking aids in the recovery of digital images that have been altered or tampered with during transmission. Although many techniques have been developed for embedding the watermark that results in recovering itself of photos, they only effectively work in the uncompressed multiple domains, leaving images vulnerable to attacks on image compression, such as noise addition. In order to effectively detect tampered areas and assure self-recovery of JPEG images, we offer a solution to address these issues in the JPEG domain. Source code compresses the original image, while channel code protects it from tampering. As a result, the image will be compressed and then used as the watermark which is embedded. The approach proposed is successfully combined with these JPEG images, producing increased image quality as well as strong performance against noise attacks. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

关键词： image compression

来源：评论

学校读者我要写书评

暂无评论

Evaluation of Xilinx Versal Architecture for Next-Gen Edge Computing in Space

Evaluation of Xilinx Versal Architecture for Next-Gen Edge C...

引用

ieee conference on Aerospace

作者： Noah Perryman Christopher Wilson Alan George NSF SHREC Center University of Pittsburgh Pittsburgh PA USA Science Data Processing Branch Code 587 NASA Goddard Space Flight Center Greenbelt MD United States NSF SHREC Center University of Pittsburgh Pittsburgh PA United States

Space edge computing has unique considerations (e.g., size, power, space radiation, etc.) that limit the performance capabilities of achievable onboard processing. Due to these restrictions, current state-of-the-art devices for space edge computing are unable to meet the resource and performance requirements for next-gen communication, navigation, and artificial intelligence (AI) applications planned for future science and defense missions. To address these issues, space designers are considering domain-specific architectures (DSAs) with specialized acceleration hardware, such as the Xilinx Versal Adaptive Compute Acceleration Platform (ACAP) architecture. This platform is heterogeneous and provides developers with scalar/vector processing units and programmable logic for mission-specific customization. In this paper, the performance and power-efficiency tradeoffs of different Versal ACAP processing subsystems (i.e., AI Engines, ARM Cortex-A72, ARM Cortex-R5F, and programmable logic) were evaluated by comparing execution of representative space applications. These experiments included three convolutional neural network (CNN)-based image-classification applications (i.e., MobileNetV1, ResNet-50, and GoogLeNet) for AI and a multitaper spectral estimation application for communication. Notably, the vector-based Versal AI Engines showed a promising performance improvement over the ARM Cortex-A72 and Cortex-R5F while usually being outperformed by the programmable logic. However, the AI Engines typically consumed significantly less power, with some AI engine applications being 1.48× more energy efficient compared to the programmable logic while not losing static power to excessive resource utilization. Overall, this evaluation demonstrates the versatility and heterogeneity of the Versal architecture and the tradeoffs between the on-chip subsystems. Therefore, the Versal will be considered for a future single-board computer to address challenges set by demanding next-gen s

关键词： Performance evaluation Program processors Space missions Estimation computer architecture Energy efficiency Resource management

来源：评论

学校读者我要写书评

暂无评论

Hyperspectral image Classification Using Multi-feature Fusion Residual Hypergraph Convolution Network

Hyperspectral Image Classification Using Multi-feature Fusio...

引用

ieee international conference on Signal and image processing (ICSIP)

作者： Tianxing Zhu Qin Liu Lixiang Zhang School of Software Engineering Tongji University Shanghai China School of Computer Science Northwestern Polytechnical University Xi’an China

Convolutional neural network (CNN) has been broadly adopted on hyperspectral image (HSI) processing due to its impressive feature extraction capabilities. Nevertheless, it is still a challenge for CNN-based hyperspectral image classification methods to extract more effective spectral-spatial features considering all spectral bands. The recent advances of graph convolutional neural networks (GCN), a combination of CNN and graph representation learning, have shown great potential in the HSI. However, the existing GCN-based methods have lower accuracy than the mature CNN methods due problems such as overdependence on the adjacency matrix, usage of a single modal feature. In this paper, we propose the Multi-Feature Fusion Residual Hypergraph Convolution Network (MF-RHCN) for HSI classification. We present a novel Hypergraph Convolution Network that fuses multimodal features effectively, and automatically updates the hyperedge weights, which solves the difficulties encountered by the previous GCN-based hyperspectral classification methods. MF-RHCN first generates hyperedges from features of different modalities to construct a hypergraph representing multimodal features in HSI. Then, the HSI and the extracted hypergraph are input into the hypergraph convolutional neural network for learning. We enhance the hypergraph convolutions with residual connection and input mapping, which helps reduce the over-smoothing issue associated with deep hypergraph models. Sufficient experiments on three datasets have proved the effectiveness of MF-RHCN.

关键词：

来源：评论

学校读者我要写书评

暂无评论

SCNN-Assisted Fast image Reconstruction using Small Dataset: Gains and Limitations

SCNN-Assisted Fast Image Reconstruction using Small Dataset:...

引用

ieee Symposium on Nuclear Science (NSS/MIC)

作者： A. Hashemi Y. Feng H. Sabet Radiology Department Harvard Medical School/Massachusetts General Hospital Boston Massachusetts United States of America

The emergence of equivariant Convolutional Neural Networks (CNNs) has shown great promise for image processing in non-Euclidean spaces and symmetrical domains. Equivariant networks shows significant enhancement in image quality and computational efficiency, making them well-suited for medical imaging applications. However, there are several challenges associated with image reconstruction from Monte Carlo simulated data, including computational complexity, statistical noise, and. This study investigates the potential of utilizing equivariant CNNs in augmenting traditional reconstruction algorithms for medical imaging applications. This study aims to explore the use of equivariant spherical CNN (SCNN) in conjunction with maximum likelihood expectation maximization (MLEM) reconstruction to lower computational costs and dependency on the completeness of input data. The approach will be compared with a conventional CNN-based augmentation framework to assess the relative benefits and limitations. Our results show that the computational cost of SCNN is drastically decreased from 2000+ seconds for each iteration of MLEM and 820 seconds for CNN to only 15 seconds for SCNN. The findings of this study have the potential to contribute to the development of more efficient and accurate reconstruction algorithms for medical imaging, ultimately leading to better outcomes and advancements in the field.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Deep Learning and image Recognition

Deep Learning and Image Recognition

引用

ieee international conference on Electronic Information and Communication Technology (ICEICT)

作者： Chaoyang Li Xiaohan Li Manni Chen Xinyao Sun School of Computer Science and Engineering Huizhou University Guangdong China Department of News Northeastern University Liaoning China

Over the next decade, we can expect artificial intelligence (AI) to have a profound impact on our society. As AI technologies continue to advance and become more integrated into various industries, we will witness significant changes that are shaping the human being's future for the better. This paper reviews the major deep learning algorithms that have accomplished latest achievements on various fields such as manufacturing robots, smart assistants, automated financial investing, social media monitoring, marketing chat-bots and self-driving car. These algorithms are trained on large datasets of labeled images, where each image is associated with a set of labels or categories. This paper provides an overview of popular techniques for image recognition. It discusses the convolutional neural network (CNN), which is made up of multiple convolutional layers followed by pooling layers, dropout layers, batch normalization lays and fully connected layers. The paper addresses other techniques used in deep learning for image recognition include transfer learning, where a pre-trained model is refined on new datasets, and generative adversarial networks (GANs), which can generate realistic images by training two neural networks against each other. This paper highlights that deep learning has revolutionized the field of image recognition and has opened up new possibilities for many applications.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Enhancing Tone Recognition in Large-Scale Social Media Data with Deep Learning and Big Data processing

Enhancing Tone Recognition in Large-Scale Social Media Data ...

引用

ieee international conference on Artificial Intelligence and computer applications (ICAICA)

作者： Xingyu Huo School of Theater Film and Television Communication University of China Beijing China

Huge processing power, big data, and deep learning algorithms have made major technical improvements possible. Technology is developing to analyse, predict, and meet our unmet requirements. Speech is the primary human communication method, and when a microphone sensor is employed, it may be utilised to support human-computer interaction (HCI). A new field of study in human-computer interaction (HCI) is quantifying emotion recognition from speech signals using these sensors. This research has applications in virtual reality, healthcare, emergency call centres, behaviour assessment, and human-reboot interaction. Our main contributions in this study are to: (i) further develop discourse feeling acknowledgment (SER) exactness over the cutting edge; what's more (ii) bring down the figuring cost of the proposed SER model. The method involved with recognizing a speaker's feelings from their discourse signal is known as discourse feeling acknowledgment. The three essential strides in the feeling acknowledgment process are highlight extraction, include determination, and classifier. Improving a framework's capacity to perceive discourse feelings using different element extraction techniques is the essential objective of this work. The work centers around the pre-handling of the sound examples that are gotten, where channels are utilized to wipe out commotion from discourse tests. Worldwide discriminative elements are learned in completely associated layers, though neighborhood stowed away examples are learned in convolutional layers that bring additional consideration to down-example the component maps rather than the pooling layer. Discourse feeling classification is finished utilizing a SoftMax classifier. It shows the worth and viability of the proposed SER approach and exhibits that it is so relevant to functional purposes.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Dynamic TF-TDNN: Dynamic Time Delay Neural Network Based on Temporal-Frequency Attention for Dialect Recognition

Dynamic TF-TDNN: Dynamic Time Delay Neural Network Based on ...

引用

international conference on Acoustics, Speech, and Signal processing (ICASSP)

作者： Chao Liao Jinwen Huang Huan Yuan Peng Yao Jianchao Tan Dawei Zhang Feng Deng Xiaorui Wang Chengru Song Kuaishou Technology Beijing China

Dialect recognition aims to recognize dialect categories in utterances, which has been applied in many audio applications. Recently, various Time Delayed Neural Network (TDNN) based AI models are proposed to solve dialect recognition problems, such as D-TDNN, DMC-TDNN, and ECAPA-TDNN, however, most of them only perform temporal attention in the last statistical pooling layer of the TDNN network, which ignores the importance of simultaneously capturing both frequency and temporal key information in utterances under different receptive fields. In contrast, we introduce a hybrid attention mechanism in both the temporal and frequency domain, called the TF-attention module, which adaptively pays more attention to the indeed important frames and the frame-level important information under different receptive fields for dialect recognition. Moreover, we are the first to introduce a dynamic architecture mechanism in the field of dialect recognition to dynamically reduce the computational cost and the number of parameters of models. We evaluate the proposed dynamic TF-TDNN on the OLR challenge AP20-OLR-dialect task and achieve State-Of-The-Art (SOTA) performance with fewer model parameters.

关键词： Time-frequency analysis Computational modeling Delay effects Neural networks Speech recognition computer architecture Signal processing

来源：评论

学校读者我要写书评

暂无评论

Design of Risc-V processing Unit Using Posit Number System

Design of Risc-V Processing Unit Using Posit Number System

引用

international conference on Signal processing and Communication (ICSPC)

作者： D Malathi R Sneha M Shanmugapriya S Sethurajan Department of Electronics and Communication Kongu Engineering College Perundurai Tamilnadu India

RISC-V is an Open Instruction Set (ISA) architecture which allows the implementation of the custom instruction set. The RISC-V processor was introduced to reduce the instruction set and increase register resource investment. Unlike other ISA designs, RISC-V is offered under an open-source licence. The chip and device manufacturing industries have provided the CPU with substantial support. RISC-V is therefore primarily made to be flexibly expandable and configurable for use in a variety of applications. In RISC-V, the number system is represented in several formats of real number arithmetic. The representation of real numbers in computer are standard ieee 754 Floating point. It has many impacts like rounding, excess of Not a Number (NaN), signed Zero. By combining these aspects using posit number format, posit processing unit is developed. The Posit number system is alternative to ieee 754 Floating point representation. Modification to the standard RISC-V ISA which enables conversion of 8 or 16 bit posits to 16-bit ieee Floating point number to obtain more dynamic range within the given number of bits. The proposed model is to make the chip smaller and lower the power. Low-precision concepts present an improved solution than approximate computing strategies that attempt to accept lower response quality. When compared to floats of the same size, high-precision positions offer more accurate decimals. Posit avoid the illogical error of assigning a value to the NaN (not-a-number).

关键词： Reduced instruction set computing Power demand computer architecture Dynamic range Signal processing Software Registers

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 << < 429 430 431 432 433 434 435 436 437 438 > >>

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：