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

14,191 篇 会议
294 篇 期刊文献
199 册 图书
6 篇 学位论文

馆藏范围

14,690 篇 电子文献
1 种 纸本馆藏

日期分布

学科分类号

7,903 篇 工学
- 5,308 篇 计算机科学与技术...
- 2,565 篇 软件工程
- 2,534 篇 电气工程
- 1,454 篇 信息与通信工程
- 939 篇 控制科学与工程
- 753 篇 光学工程
- 548 篇 生物医学工程（可授...
- 536 篇 电子科学与技术（可...
- 513 篇 仪器科学与技术
- 488 篇 机械工程
- 358 篇 生物工程
- 246 篇 测绘科学与技术
- 185 篇 材料科学与工程（可...
- 173 篇 航空宇航科学与技...
- 150 篇 交通运输工程
- 111 篇 安全科学与工程
- 96 篇 化学工程与技术
3,834 篇 理学
- 2,639 篇 物理学
- 1,009 篇 数学
- 411 篇 生物学
- 276 篇 系统科学
- 256 篇 统计学（可授理学、...
- 133 篇 化学
1,610 篇 医学
- 1,422 篇 临床医学
- 250 篇 特种医学
- 185 篇 基础医学(可授医学...
- 113 篇 药学(可授医学、理...
782 篇 管理学
- 469 篇 管理科学与工程(可...
- 339 篇 图书情报与档案管...
- 98 篇 工商管理
67 篇 农学
60 篇 法学
37 篇 教育学
30 篇 经济学
27 篇 文学
20 篇 军事学
19 篇 艺术学
1 篇 哲学

主题

2,879 篇 image processing
1,507 篇 image segmentati...
1,202 篇 signal processin...
660 篇 feature extracti...
626 篇 cameras
592 篇 image edge detec...
590 篇 signal processin...
530 篇 image coding
524 篇 image reconstruc...
475 篇 real-time system...
416 篇 algorithm design...
354 篇 image analysis
352 篇 image color anal...
333 篇 deep learning
332 篇 image enhancemen...
312 篇 object detection
305 篇 pixel
300 篇 hardware
299 篇 real time system...
294 篇 computer archite...

机构

22 篇 signal and image...
19 篇 samara national ...
14 篇 school of electr...
12 篇 department of el...
11 篇 samara national ...
11 篇 univ texas san a...
10 篇 school of electr...
10 篇 center for signa...
9 篇 information syst...
9 篇 institute of ima...
9 篇 tongji universit...
9 篇 university of ch...
8 篇 school of electr...
8 篇 univ chinese aca...
8 篇 university of to...
8 篇 purdue univ sch ...
8 篇 electrical compu...
8 篇 image processing...
8 篇 information syst...
8 篇 samara national ...

作者

19 篇 anon
13 篇 de-shuang huang
12 篇 jagdish chand ba...
11 篇 p. pirsch
10 篇 jie yang
10 篇 panetta karen
10 篇 j.m. mendel
9 篇 c.-c.j. kuo
9 篇 mahalanobis a
9 篇 r. chellappa
8 篇 miller jwv
8 篇 semenishchev e.
8 篇 frank hannig
8 篇 hannig frank
8 篇 ilias maglogiann...
7 篇 busch christoph
7 篇 zelensky a.
7 篇 bovik alan c.
7 篇 jerry m. mendel
7 篇 kupriyanov a.v.

语言

14,370 篇 英文
220 篇 其他
82 篇 中文
20 篇 土耳其文
6 篇 德文
3 篇 葡萄牙文
2 篇 西班牙文
2 篇 俄文
1 篇 法文
1 篇 波兰文

检索条件"任意字段=Conference on Image Processing - Algorithms and Systems III"

共 14690 条记录，以下是921-930 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：

Hierarchical DNN-Based image Segmentation Algorithm Using Texton, Superpixels, and Layer-Adaptive Loss Functions 16

Hierarchical DNN-Based Image Segmentation Algorithm Using Te...

引用

16th International conference on Signal-image Technology and Internet-Based systems (SITIS)

作者： Yu, Cheng-Hsuan Ding, Jian-Jiun Natl Taiwan Univ Grad Inst Commun Engn Taipei Taiwan

ISBN: (纸本)9781665464956

image segmentation is critical to object-oriented image processing. Many conventional segmentation algorithms are based on the superpixel, since it integrates the pixels with similar colors and locations in prior and is beneficial for segmentation. Recently, several segmentation algorithms based on deep learning were developed. However, due to the irregular shape and size of superpixels, it is hard to apply the superpixel directly in a leaning-based segmentation algorithm. In this paper, we propose a novel segmentation method that well integrates the techniques of the deep neural network (DNN), the superpixel, adaptive loss functions, and multi-layer feature extraction. First, different from other learning-based algorithm, which applies an image or its bounding boxes as the input, we adopt the mean and the histogram differences of the features of two superpixels as the input of the DNN to determine whether they should be merged. Moreover, to well consider both largescaled and small-scaled features, a hierarchical architecture is adopted. For different layers, the DNN models with different loss functions are applied. A larger penalty for over-merging is applied in the first layer and a larger penalty for oversegmentation is applied in the following layer. Moreover, according to human perception, the features of colors, areas, the gradient at the boundary, and the texton, which is highly related to the texture, are applied. Experiments show that the proposed method outperforms other state-of-the-art image segmentation methods and produces highly accurate segmentation results.

关键词： image segmentation superpixel merging deep learning

来源：评论

学校读者我要写书评

暂无评论

Towards Video Prediction

Towards Video Prediction

引用

IEEE conference on Industrial Electronics and Applications (ICIEA)

作者： Bradley Zylstra Weibo Gong University of Massachusetts Amherst MA USA

ISBN: (数字)9798350360868

ISBN: (纸本)9798350360875

This paper introduces a novel approach to video prediction and object recognition based on treating image signals as dynamic system operators. We develop algorithms that extract invariant features from pixel patches to construct numerical matrices for image frame transitions. Our method diverges from conventional 2D signal processing by viewing images as operators rather than initial conditions. This perspective aligns more closely with biological visual systems and offers potential for efficient electronic implementations. We demonstrate the efficacy of our approach through experiments in mental rotation, affine transformations, and rotated MNIST digit recognition. Our results show that deformation invariance can be obtained without prior knowledge of the transformation. This work contributes to the broader goal of developing biologically plausible computer vision systems, with implications for video prediction, object recognition, and abstract concept formation for AI.

关键词： Visualization Computer vision Deformation Heuristic algorithms Signal processing algorithms Prediction algorithms Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Remote measurement of the clinical prescription of spectacle lenses 25

Remote measurement of the clinical prescription of spectacle...

引用

conference on Novel Optical systems, Methods, and Applications XXV

作者： Guan, Yuanxin Miller, Joseph M. Harvey, Erin M. Ashley, Sean Schwiegerling, Jim Univ Arizona James C Wyant Coll Opt Sci 1630 E Univ Blvd Tucson AZ 85721 USA Univ Arizona Dept Ophthalmol & Vis Sci 655 N Alvernon Way 204 Tucson AZ 85711 USA

ISBN: (纸本)9781510654174;9781510654167

The SpectRx system has been developed to measure sphero-cylindrical spectacle lens power as an alternative to clinical lensmeters. This work was inspired by the ongoing global pandemic, which limited physical access to eye care facilities for regular eye exams The SpectRx system aims to bypass this limitation by providing at-home prescription measurements. The power and orientation of the spectacle lenses are obtained by the use of readily available objects such as a cell phone camera, a displayed or printed target, and a fixed-dimension magnetic stripe card. The magnification of the lenses can be calculated by examining the image captured through the lens of the target at a fixed distance. The magnification may be spatially varying due to the cylinder component of the lens. processing the pictures captured with a cell phone camera is done automatically with standard image processing algorithms. The processed images, in turn, are used to calculate a clinical prescription, i.e., Sph/CylxAxis. The SpectRx may expand access to quality eye care in not only the current pandemic situation but also in locations where eye care may not be easily accessible, such as some rural or remote areas. The image processing and clinical prescription calculation are discussed here.

关键词： Spectacle lenses power measurement telemedicine image processing

来源：评论

学校读者我要写书评

暂无评论

The Estimation of Inertial Measurement Units Accuracy Using Digital image processing algorithms

The Estimation of Inertial Measurement Units Accuracy Using ...

引用

2022 Wave Electronics and its Application in Information and Telecommunication systems, WECONF 2022

作者： Shmatko, E.V. Poroykov, A. Yu. National Research University Moscow Power Engineering Institute Department of Physics Named after V.A. Fabrikant Moscow Russia

ISBN: (纸本)9781665462402

The paper describes an approach for estimation of inertial measurement unit using image processing algorithm to determine the position of an object in space. The results of measurements of the angular velocities of a freely rotating platform using the inertial and photogrammetry methods are presented. The inertial method data contains the measurement results of a three-axis solid gyroscope. The photogrammetry method determines the position of the platform using digital video cameras based on images of fiducial markers installed on it. A comparison of the measurement results is presented, conclusions about the accuracy of the tested gyroscope are drawn. © 2022 IEEE.

关键词： Angular velocity

来源：评论

学校读者我要写书评

暂无评论

Wavefront coded image restoration algorithm based on compressed sensing for off-axis mirror system 7

Wavefront coded image restoration algorithm based on compres...

引用

7th Asia Pacific conference on Optics Manufacture (APCOM)

作者： Yi, Huang Jun, Chang Li Dongmei Cao Jiajing Shan, Du Beijing Inst Technol Beijing 100081 Peoples R China Univ Chinese Acad Sci Beijing 100049 Peoples R China Chinese Acad Sci Inst Semicond Beijing 100083 Peoples R China

ISBN: (数字)9781510652095

ISBN: (纸本)9781510652095;9781510652088

By combining optical systems and image processing, wavefront coding can greatly expand the depth of focus and depth of field of optical systems. It has been widely used in iris detection, high-power microscopic objective lens, infrared optical system athermalized, and so on. At present, the image restoration algorithms commonly used in wavefront coding are based on deconvolution, Wiener filtering, and so on. Although these algorithms can achieve an excellent image restoration effect, they will also bring boundary ringing effects and artifacts to the image. When the image is disturbed by strong noise, the restoration effect will also be seriously affected. To solve these problems, a wavefront coded image restoration algorithm based on compressed sensing is proposed in this paper. The strong data reconstruction ability of the compressed sensing restoration algorithm is used to restore the encoded image obtained by the wavefront coding system. This method can effectively suppress noise and reconstruct the image without artifact and boundary ringing effect. Through the comparison of simulation results, the effectiveness of the proposed method is verified.

关键词： wavefront coding compressed sensing artifact

来源：评论

学校读者我要写书评

暂无评论

ResRnnNet: Learning to Reconstruct the 3D Room Layout from a Single RGB Panorama 17th

ResRnnNet: Learning to Reconstruct the 3D Room Layout from a...

引用

17th International conference on Intelligent Information Hiding and Multimedia Signal processing (IIH-MSP) / 14th International conference on Frontiers of Information Technology, Applications and Tools (FITAT)

作者： Zhao, Shaonan Li, Wei Univ New South Wales Sch Civil Engn & Environm Sydney NSW 2052 Australia Harbin Engn Univ Coll Comp Sci & Technol Harbin 150001 Peoples R China

ISBN: (纸本)9789811910579;9789811910562

3D room layout reconstruction from a single RGB panoramic image has been an emerging research topic in recent years. To achieve better prediction accuracy, in this paper, we propose a new approach to predict 3D room layout from a single panoramic image. Our reconstruction flow follows a common framework which is same as LayoutNet [9] and HorizonNet [4];however, we redesign a new deep learning architecture with recurrent neural networks (RNNs) encoder-decoder as an extension for keypoints refinement and use a gradient ascent optimization algorithm to minimize the similar loss. Experiments on both cuboid-shaped and general Manhattan layouts show that the proposed work outperforms recent algorithms in prediction accuracy.

关键词： 3D room layout Deep learning Manhattan world Panorama image

来源：评论

学校读者我要写书评

暂无评论

6G XL-MIMO Channel Estimation with Sub-Nyquist Tensor Completion

6G XL-MIMO Channel Estimation with Sub-Nyquist Tensor Comple...

引用

Digital Signal processing and its Applications (DSPA), International conference on

作者： Semyon Dorokhin Vladimir Lyashev Mikhail Makurin Moscow Institute of Physics and Technology Moscow Russia Radio Transmission Technology Lab Moscow Russia

ISBN: (数字)9798331518752

ISBN: (纸本)9798331518769

Channel estimation overhead reduction is one of the main problems for 6G XL-MIMO systems. As the number of antennas and subcarriers grows, so does the overhead of traditional channel estimation methods. High overhead limits user mobility and affects the latency. Most of the popular tensor-based channel estimation algorithms combine reference signals transmission and channel tensor estimation into one problem. Such approach typically imposes limitations on reference signals, making backward-compatibility with existing 5G standard challenging. We propose to separate tensor completion and channel tensor elements estimation into different tasks. We show that tensor completion algorithms from image processing and fMRI scanning can be reused for sub-Nyquist completion of OFDM MIMO tensors to reduce the overhead. Furthermore, we extend one of these algorithms from CPD to Tensor Train (TT) and demonstrate that TT-based algorithm reduces channel estimation error up to 2 times. With the proposed approach other algorithms can be further developed based on tensor completion theory.

关键词： 6G mobile communication Tensors OFDM image processing Channel estimation Signal processing algorithms Estimation Massive MIMO Functional magnetic resonance imaging Standards

来源：评论

学校读者我要写书评

暂无评论

A Survey of Transformer-Based Few-Shot image Classification Techniques

A Survey of Transformer-Based Few-Shot Image Classification ...

引用

Natural Language processing (ICNLP), International conference on

作者： Chaoqi Song Ying Liu Jinglu He Center for Image and Information Processing Xi’an University of Posts and Telecommunications Xi’an China

ISBN: (数字)9798350349115

ISBN: (纸本)9798350349122

Few-shot image classification is a critical issue in the field of computer vision, facing challenges related to data scarcity and model generalization. Transformer models, representing self-attention mechanisms, have made significant strides in recent years in the domain of few-shot classification. This paper commences with an introduction to the background and challenges of few-shot classification, along with a description of the principles and structure of the Transformer model. Subsequently, the paper categorizes Transformer-based few-shot image classification methods into meta-learning-based, metric-learning-based, fine-tuning-based, and feature-enhancement-based approaches, whose theoretical foundations of each method are expounded and the comparative analysis of representative algorithms are also provided. Furthermore, the paper delves into prospective research directions in this field.

关键词： Surveys Representation learning Learning systems Computational modeling Neural networks Computer architecture Transformers Natural language processing Classification algorithms image classification

来源：评论

学校读者我要写书评

暂无评论

Abnormal image Classification Using Vision Transformer for Smart Agriculture 8

Abnormal Image Classification Using Vision Transformer for S...

引用

8th IEEE International conference on Engineering Technologies and Applied Sciences, ICETAS 2023

作者： Hafeez, Talha Shahid, Muhammad Farrukh Khanzada, Tariq J.S. Fast School of Computing FAST-NUCES Karachi Pakistan King Abdulaziz University Faculty of Computing and Information Technology Department of Information System Jeddah21589 Saudi Arabia Mehran Uet Computer Systems Engineering Departments Jamshoro Pakistan

ISBN: (纸本)9798350327090

Smart agriculture is an emerging innovative sector that harnesses technological advancements like sensor monitoring, image processing, soil quality evaluation, and automated water systems, all of which contribute towards enhanced crop yields and superior quality of produce. With the progression towards automated systems in smart agriculture, the need for a robust data infrastructure that assures data integrity is critical. This demands an integrated approach with anomaly detection algorithms and a robust classification technique to ensure an error-free data pipeline. This study employs two distinct datasets to achieve abnormal image classification using a vision transformer (ViT) and a convolutional neural network (CNN). The primary dataset, obtained through direct collection from cotton fields in Karachi and Balochistan, Pakistan, comprises high-resolution agricultural images captured using specialised cameras and Unmanned Aerial Vehicles (UAVs). The secondary dataset was sourced from a publicly available database. The two classification models, ViT and CNN, were applied to both datasets. These supervised models are primarily used to classify images as normal or abnormal. The results of the two classification models were also compared, with the ViT model surpassing the CNN model in performance and achieving the highest accuracy on both datasets. The Vision Transformer model proved to be robust for supervised data, showing remarkable accuracy of 96.00% on the public dataset, 99.18% on both camera based images and 97.56% on UAV images of collected dataset as compared to the CNN model's accuracy that are 82.00% on the public dataset and 98.36% on camera and 95.12% on UAV images of collected dataset. © 2023 IEEE.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

image processing for Future Machine Learning Algorithm Applications on Infrared Thermography of Building Envelope systems 5th

Image Processing for Future Machine Learning Algorithm Appli...

引用

5th International conference on Building Energy and Environment, COBEE 2022

作者： Gertsvolf, David Horvat, Miljana Khademi, April Aslam, Danesh Berardi, Umberto Department of Architectural Science Toronto Metropolitan University TorontoON Canada Department of Electrical Computer and Biomedical Engineering Toronto Metropolitan University TorontoON Canada

ISBN: (纸本)9789811998218

This paper presents and discusses the development of a procedure required for the future application of machine learning (ML) in the review and investigation of infrared (IR) thermography images for building envelopes. A background on ML and existing applications within the building science industry is presented to capture an understanding of the possibilities available within this innovative method of evaluating deficiencies of building envelopes. Then, a breakdown of the groundwork and image processing required for the application of ML algorithms is reported. A procedure developed using a python coding environment is presented, demonstrating the process of taking an RGB IR image, grey scaling it and normalizing for the application of unsupervised ML algorithm. Similarly, for the purpose of supervised ML, the same approach is presented, followed by a procedure for the creation of segmentation masks using two types of approaches: temperature threshold and manually identified segments. A discussion of possible development of these new techniques concludes this paper. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Anomaly detection

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500页 << < 89 90 91 92 93 94 95 96 97 98 > >>

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：