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

15,795 篇 会议
362 篇 期刊文献
192 册 图书

馆藏范围

16,349 篇 电子文献
0 种 纸本馆藏

日期分布

学科分类号

8,318 篇 工学
- 5,418 篇 计算机科学与技术...
- 2,987 篇 软件工程
- 2,164 篇 电气工程
- 1,865 篇 信息与通信工程
- 1,127 篇 控制科学与工程
- 878 篇 光学工程
- 647 篇 机械工程
- 639 篇 电子科学与技术（可...
- 538 篇 生物医学工程（可授...
- 537 篇 仪器科学与技术
- 477 篇 生物工程
- 224 篇 测绘科学与技术
- 217 篇 交通运输工程
- 203 篇 材料科学与工程（可...
- 199 篇 航空宇航科学与技...
- 164 篇 化学工程与技术
- 163 篇 网络空间安全
- 141 篇 安全科学与工程
4,006 篇 理学
- 2,389 篇 物理学
- 1,359 篇 数学
- 527 篇 生物学
- 368 篇 统计学（可授理学、...
- 341 篇 系统科学
- 205 篇 化学
1,522 篇 医学
- 1,348 篇 临床医学
- 200 篇 特种医学
- 197 篇 基础医学(可授医学...
1,055 篇 管理学
- 613 篇 管理科学与工程(可...
- 481 篇 图书情报与档案管...
- 140 篇 工商管理
84 篇 法学
66 篇 农学
45 篇 经济学
43 篇 教育学
37 篇 军事学
32 篇 文学
26 篇 艺术学
1 篇 哲学

主题

2,780 篇 image processing
1,550 篇 image segmentati...
1,513 篇 signal processin...
865 篇 feature extracti...
652 篇 image edge detec...
634 篇 cameras
633 篇 signal processin...
604 篇 real-time system...
575 篇 image reconstruc...
517 篇 image coding
498 篇 algorithm design...
399 篇 clustering algor...
397 篇 image color anal...
396 篇 deep learning
379 篇 object detection
367 篇 image enhancemen...
352 篇 accuracy
332 篇 hardware
323 篇 image analysis
312 篇 computer vision

机构

38 篇 xi'an jiaotong u...
32 篇 northwestern pol...
32 篇 school of electr...
31 篇 xi an jiao tong ...
31 篇 school of electr...
26 篇 university of ch...
24 篇 xi an jiao tong ...
23 篇 national laborat...
22 篇 xi an jiao tong ...
21 篇 signal and image...
19 篇 samara national ...
17 篇 xidian universit...
16 篇 school of inform...
16 篇 institute of art...
15 篇 department of el...
15 篇 school of automa...
15 篇 national key lab...
15 篇 school of comput...
14 篇 school of electr...
14 篇 chang'an univers...

作者

39 篇 chen xi
27 篇 xi chen
24 篇 licheng jiao
22 篇 nanning zheng
20 篇 li xi
20 篇 liu ying
18 篇 xuelong li
17 篇 anon
15 篇 xi ning
15 篇 xi li
14 篇 yuan yuan
13 篇 de-shuang huang
13 篇 wei li
13 篇 ying liu
12 篇 jie yang
12 篇 yang xi
12 篇 yanning zhang
11 篇 p. pirsch
11 篇 liu yang
11 篇 mingyi he

语言

16,039 篇 英文
165 篇 其他
133 篇 中文
20 篇 土耳其文
3 篇 葡萄牙文
2 篇 法文
2 篇 俄文
1 篇 西班牙文

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

共 16349 条记录，以下是4811-4820 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：

CLIPCEIL: Domain Generalization through CLIP via Channel rEfinement and image-text aLignment 38

CLIPCEIL: Domain Generalization through CLIP via Channel rEf...

引用

38th conference on Neural Information processing systems, NeurIPS 2024

作者： Yu, xi Yoo, Shinjae Lin, Yuewei Artificial Intelligence Department Computing and Data Science United States Directorate Brookhaven National Laboratory UptonNY11973 United States

Domain generalization (DG) is a fundamental yet challenging topic in machine learning. Recently, the remarkable zero-shot capabilities of the large pre-trained vision-language model (e.g., CLIP) have made it popular for various downstream tasks. However, the effectiveness of this capacity often degrades when there are shifts in data distribution during testing compared to the training data. In this paper, we propose a novel method, known as CLIPCEIL, a model that utilizes Channel rEfinement and image-text aLignment to facilitate the CLIP to the inaccessible out-of-distribution test datasets that exhibit domain shifts. Specifically, we refine the feature channels in the visual domain to ensure they contain domain-invariant and class-relevant features by using a lightweight adapter. This is achieved by minimizing the inter-domain variance while maximizing the inter-class variance. In the meantime, we ensure the image-text alignment by aligning text embeddings of the class descriptions and their corresponding image embedding while further removing the domain-specific features. Moreover, our model integrates multi-scale CLIP features by utilizing a self-attention fusion module, technically implemented through one Transformer layer. Extensive experiments on five widely used benchmark datasets demonstrate that CLIPCEIL outperforms the existing state-of-the-art methods. The source code is available at https://***/yuxi120407/CLIPCEIL. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

EXTRACTION OF EDGE-BASED AND REGION-BASED FEATURES FOR OBJECT RECOGNITION 12

EXTRACTION OF EDGE-BASED AND REGION-BASED FEATURES FOR OBJEC...

引用

12th conference on Intelligent Robots and Computer Vision: algorithms and Techniques

作者： COUTTS, B RAVI, S HU, GZ SHRIKHANDE, N CENT MICHIGAN UNIV DEPT COMP SCIMT PLEASANTMI 48858

ISBN: (纸本)0819413208

One of the central problems of computer vision is object recognition. A catalogue of model objects is described as a set of features such as edges and surfaces. The same features are extracted from the scene and matched against the models for object recognition. Edges and surfaces extracted from the scenes are often noisy and imperfect. In this paper algorithms are described for improving low level edge and surface features. Existing edge extraction algorithms are applied to the intensity image to obtain edge features. Initial edges are traced by following directions of the current contour. These are improved by using corresponding depth and intensity information for decision making at branch points. Surface fitting routines are applied to the range image to obtain planar surface patches. An algorithm of region growing is developed that starts with a coarse segmentation and uses quadric surface fitting to iteratively merge adjacent regions into quadric surfaces based on approximate orthogonal distance regression. Surface information obtained is returned to the edge extraction routine to detect and remove fake edges. This process repeats until no more merging or edge improvement can take place. Both synthetic (with Gaussian noise) and real images containing multiple object scenes have been tested using the merging criteria. Results appeared quite encouraging.

关键词： Edge detection Object recognition image segmentation Algorithm development Feature extraction Detection and tracking algorithms image processing algorithms and systems Computer vision technology Machine vision Visual process modeling

来源：评论

学校读者我要写书评

暂无评论

Automatic generation of multipath algorithms in the cellular nonlinear network

Automatic generation of multipath algorithms in the cellular...

引用

conference on Applications of Artificial Neural Networks in image processing VI

作者： Preciado, VM Guinea, D Montúfar, R Spanish Council Sci Res Inst Automat Ind Madrid 28500 Spain

ISBN: (纸本)0819439835

The traditional image processing techniques require a lot of computational effort due to data on each pixel are computed in a sequential way and the path of information is an A/D converter. The delay accumulation create in this process is unacceptable in real time image processing because of the high information flow managed in the usual vision tasks (e.g. automatic industrial inspection, vision problems in robotics, pattern analysis, etc.). Thus, the use of a massive parallel architecture working with analog signals avoids the previous problems. This is just the basis idea of Cellular Neural Network (CNN's): an array of analogic dynamic processors which cells interact directly within a finite local neighborhood. The local CNN connectivity allow its realization as VLSI chips that can operate at a very high speed and complexity. Nowadays CNN architectures implemented as VLSI chips shows the aptitude of extremely high speed compared with traditional digital image processing tools. The proliferation of more and more sophisticated CNN architectures, and the increasing effort to implant practical system based in CNN chips, make important the development of analog algorithm to perform complex image processing tasks dedicated to many different fields, i.e. industrial applications, robotic systems and pattern recognition. The objective of this work is to generate a learning machine capable of find solutions for complex image processing task by CNN's. First a general machine for automatic analog algorithm design independent of the problem to solve is created, this is accomplished through an evolutionary strategy that is an extension of genetic programming. Second, this work introduces a suite of sub-mechanisms that increase the power of genetic programming and contribute to reduce the enormous space search for producing a plentiful search. Some concepts in this section are related with AI theory, in such a way that in this work we are in the intersection field of AI and Imag

关键词： cellular neural network visual processing parallel computation peal time processing automatic templates generation genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Implementation of Dubin Curves-Based RRT* Using an Aerial image for the Determination of Obstacles and Path Planning to Avoid Them During Displacement of the Mobile Robot

Implementation of Dubin Curves-Based RRT* Using an Aerial Im...

引用

Multidisciplinary International conference of Research Applied to Defense and Security (MICRADS)

作者： Daniel Tenezaca, B. Canchignia, Christian Aguilar, Wilbert Mendoza, Dario Univ Fuerzas Armadas ESPE Sangolqui Ecuador

ISBN: (纸本)9789811391552;9789811391545

The application of mobile robots in autonomous navigation has contributed to the development of exploration tasks for the recognition of unknown environments. There are different methodologies for obstacles avoidance implemented in mobile robots;however, this research introduces a novel approach for a path planning of an unmanned ground vehicle (UGV) using the camera of a drone to get an aerial view that allows to recognize ground features through image processing algorithms for detecting obstacles and target them in a determined environment. After aerial recognition, a global planner with Rapidly-exploring Random Tree Star (RRT*) algorithm is executed, Dubins curves are themethod used in this case for non-holonomic robots. The study also focuses on determining the compute time which is affected by a growing number of iterations in the RRT*, the value of step size between the tree's nodes and finally the impact of a number of obstacles placed in the environment. This project is the initial part of a larger research about a Collaborative Aerial-Ground Robotic System.

关键词： RRT Path planning image processing

来源：评论

学校读者我要写书评

暂无评论

image recognition using convolutional neural network combined with ensemble learning algorithm 4

Image recognition using convolutional neural network combine...

引用

2019 4th International conference on Intelligent Computing and Signal processing, ICSP 2019

作者： Mo, Weilong Luo, xiaoshu Zhong, Yexiu Jiang, Wenjie College of Electronic Engineering Guangxi Normal University Guilin541004 China

An image recognition algorithm based on ensemble learning algorithm and convolution neural network structure (ELA-CNN) is proposed to solve the problem that a single convolution neural network (CNN) classifier may be more prone to error or unreliable prediction. In order to improve the effect of ensemble learning, enhance the transfer of features, extract deeper features and multi-scale features, the network structure uses various model structure of the mainstream algorithms. Bagging training method is used in the training process, that is, different learners use different data sets to ensure the learning differences. Finally, the prediction result of all classifiers is used to get the final image target recognition according to the decision algorithm. The algorithm is simulated with of the open data set of cifar-10. The experimental results show that the proposed algorithm has a high recognition accuracy. The recognition rate of the test set reaches 98.89% and the recognition result is securer and reliable. © 2019 IOP Publishing Ltd. All rights reserved.

关键词： image recognition

来源：评论

学校读者我要写书评

暂无评论

Training Transformers with 4-bit Integers 37

Training Transformers with 4-bit Integers

引用

37th conference on Neural Information processing systems (NeurIPS)

作者： xi, Haocheng Li, Changhao Chen, Jianfei Zhu, Jun Tsinghua Univ Dept Comp Sci & Tech Inst AI BNRist CtrTHBI LabTsinghua Bosch Joint ML Ctr Beijing Peoples R China Tsinghua Univ Inst Interdisciplinary Informat Sci Beijing Peoples R China

ISBN: (纸本)9781713899921

Quantizing the activation, weight, and gradient to 4-bit is promising to accelerate neural network training. However, existing 4-bit training methods require custom numerical formats which are not supported by contemporary hardware. In this work, we propose a training method for transformers with all matrix multiplications implemented with the INT4 arithmetic. Training with an ultra-low INT4 precision is challenging. To achieve this, we carefully analyze the specific structures of activation and gradients in transformers to propose dedicated quantizers for them. For forward propagation, we identify the challenge of outliers and propose a Hadamard quantizer to suppress the outliers. For backpropagation, we leverage the structural sparsity of gradients by proposing bit splitting and leverage score sampling techniques to quantize gradients accurately. Our algorithm achieves competitive accuracy on a wide range of tasks including natural language understanding, machine translation, and image classification. Unlike previous 4-bit training methods, our algorithm can be implemented on the current generation of GPUs. Our prototypical linear operator implementation is up to 2.2 times faster than the FP16 counterparts and speeds up the training by 17.8% on average for sufficiently large models. Our code is available at https://***/xijiu9/Train_Transformers_with_INT4.

关键词： Distribution transformers

来源：评论

学校读者我要写书评

暂无评论

Research on Location of Pulse-Diagnosis Point Based on image processing

Research on Location of Pulse-Diagnosis Point Based on Image...

引用

Chinese Intelligent systems conference, CISC 2019

作者： Liu, Qunpo xi, xiulei Liu, Guanghui Yang, Lingxiao Wang, Hongqi Henan Polytechnic University JiaozuoHenan454003 China

ISBN: (纸本)9789813296978

Intelligent pulse diagnosis robot is a research direction of intelligent medical treatment, many researchers in China and abroad have devoted themselves to promoting the modernization of Pulse-diagnosis with the development of modern science and technology. Before the robotic arm reaches the position of pulse diagnosis, the robot needs to detect the initial position of pulse diagnosis point accurately if it wants to complete autonomous localization. In order to detect the position of Pulse-diagnosis point, a method of Pulse-diagnosis point location based on image processing is proposed in this paper. The corresponding position and posture transformation relation is obtained by establishing coordinate system among image acquisition device, robotic arm and finger. The positioning method of "inch", "close" and "ruler" is proposed, and the recognition of pulse diagnosis point is realized by using gray scale transformation, middle finger positioning algorithm, palm-arm pattern detection algorithm and image coordinate transformation. The results showed that above methods could be used to locate the Pulse-diagnosis point and prepare for the intelligent pulse diagnosis process. © 2020, Springer Nature Singapore Pte Ltd.

关键词： Location

来源：评论

学校读者我要写书评

暂无评论

Distributed framework for delivery of medical image processing algorithms

Distributed framework for delivery of medical image processi...

引用

Medical Imaging 2001 conference

作者： Petrick, N Chiao, PC Clinthorne, NH Univ Michigan Dept Radiol Ann Arbor MI 48109 USA

ISBN: (纸本)0819440094

The availability of new digital detector technologies and high speed computer processing has led to the development of CAD (computer-aided diagnostic) tools that assist radiologists in detecting and characterizing mammographic lesions. To meet the challenge of developing and implementing algorithms that are computationally intensive, it is desirable to develop reusable components that can execute in a distributed environment. It is well know that the Common Object Request Broker Architecture (CORBA) provides an open solution in distributed computing. We have implemented a hybrid component model consisting of a CORBA server and a Contract Net Protocol (CNP) algorithm for distributing tasks to multiple computers for enhanced processing. Support classes were developed to wrap algorithms developed in C to operate within the distributed framework. CORBA provides communication between agents on different computers and computer platforms and the CNP algorithm is used to select the "optimal" computer for performing a task. We have evaluated this framework with CAD processing applied to digitized mammograms by transparently scheduling and distributing multiple tasks on three server computers. We achieved a significant reduction in processing times compared to processing on a single computer.

关键词： computer-aided diagnosis (CAD) medical imaging distributed computing distributed CAD

来源：评论

学校读者我要写书评

暂无评论

A model of natural computation based on recurrent neural networks and reciprocal images

A model of natural computation based on recurrent neural net...

引用

2nd IASTED International conference on Computational Intelligence

作者： Greer, Douglas S. Gen Manifolds Carlsbad CA 92009 USA

ISBN: (纸本)9780889866027

A new design that connects together many small feed-forward neural networks to form a large, recurrent, image association processor is presented. The resulting recurrent neural network uses local image information in order to form global associations. Just as a single bit of information and its complement hold each other in place in an SR Rip-flop, two arbitrary reciprocal images can hold each other in place in an image processor. The new, T-MAP design consists of a pair of two-dimensional processor grids. Each processing element has only locally connections, but in unison, they form an orderly interconnection pattern analogous to the recurrent bit connections in an SR flip-flop. The central nervous system contains many topological maps that mathematically resemble images. Consequently, algorithms that can store and recall image associations are interesting from both a scientific and an engineering point of view. An array of these association processors is similar to the collection of Brodmann areas in the neocortex. We also present a new, physiologically realistic mechanism for controlling the processor array.

关键词： natural neural systems connectionism and image understanding

来源：评论

学校读者我要写书评

暂无评论

A flexible system for gesture based human-robot interaction 51

A flexible system for gesture based human-robot interaction

引用

51st CIRP conference on Manufacturing systems (CIRP CMS)

作者： Tellaeche, Alberto Kildal, Johan Maurtua, Inaki IK4 Tekniker Calle Inaki Goenaga 5 Eibar Gipuzkoa Spain

In industrial collaborative robotics, operators and robots perform complex tasks working together without physical barriers. Under this premise, the availability of a flexible, robust and fast interaction system between the robot and the workers is a necessity. Human beings use voice and gestures to achieve a natural interaction. Taking into account the environmental conditions usually present in workshops with noise and poor lighting conditions, combining both communication channels can contribute to make the interaction more robust. This research work presents a solution to define, setup and run a flexible and robust gesture interaction system to integrate in collaborative robotics applications. (C) 2018 The Authors. Published by Elsevier B.V. Peer-review under responsibility of the scientific committee of the 51st CIRP conference on Manufacturing systems.

关键词： Robust gesture detection Human-Robot collaboration image processing machine learning algorithms

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 << < 478 479 480 481 482 483 484 485 486 487 > >>

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：