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

4,604 篇 会议
499 册 图书
65 篇 期刊文献

馆藏范围

5,167 篇 电子文献
1 种 纸本馆藏

日期分布

学科分类号

3,375 篇 工学
- 2,739 篇 计算机科学与技术...
- 1,816 篇 软件工程
- 723 篇 信息与通信工程
- 720 篇 光学工程
- 662 篇 电气工程
- 415 篇 生物工程
- 359 篇 控制科学与工程
- 348 篇 生物医学工程（可授...
- 234 篇 机械工程
- 223 篇 电子科学与技术（可...
- 125 篇 仪器科学与技术
- 122 篇 化学工程与技术
- 116 篇 安全科学与工程
- 95 篇 交通运输工程
- 90 篇 材料科学与工程（可...
- 67 篇 建筑学
- 57 篇 土木工程
1,679 篇 理学
- 925 篇 物理学
- 721 篇 数学
- 452 篇 生物学
- 203 篇 统计学（可授理学、...
- 119 篇 化学
- 57 篇 系统科学
488 篇 管理学
- 268 篇 图书情报与档案管...
- 240 篇 管理科学与工程(可...
- 78 篇 工商管理
439 篇 医学
- 418 篇 临床医学
- 127 篇 基础医学(可授医学...
- 89 篇 药学(可授医学、理...
60 篇 法学
- 59 篇 社会学
54 篇 农学
35 篇 教育学
31 篇 艺术学
29 篇 经济学
11 篇 文学
10 篇 军事学

主题

816 篇 computer vision
624 篇 pattern recognit...
534 篇 image processing
355 篇 image processing...
347 篇 artificial intel...
336 篇 image segmentati...
276 篇 feature extracti...
203 篇 image recognitio...
174 篇 deep learning
155 篇 computer graphic...
150 篇 computer imaging...
139 篇 face recognition
124 篇 machine learning
122 篇 object recogniti...
120 篇 convolutional ne...
115 篇 image analysis
113 篇 object detection
108 篇 information syst...
105 篇 cameras
105 篇 shape

机构

25 篇 istituto italian...
24 篇 google deepmind ...
24 篇 meta ai barcelon...
23 篇 politecnico di t...
14 篇 national center ...
14 篇 institute of ima...
10 篇 shanghai jiao to...
10 篇 tsinghua univers...
10 篇 university of ch...
10 篇 institute of aut...
9 篇 institute of ima...
9 篇 computer vision ...
8 篇 indian statistic...
8 篇 inria nancy gran...
8 篇 dalian universit...
8 篇 icube université...
8 篇 erasmus mc - uni...
8 篇 tencent jarvis l...
8 篇 university of ba...
7 篇 department of co...

作者

24 篇 cristian canton
24 篇 alessio del bue
24 篇 tatiana tommasi
24 篇 jordi pont-tuset
17 篇 de-shuang huang
14 篇 stefanie speidel
12 篇 wilfried philips
9 篇 paul scheunders
9 篇 h. bischof
9 篇 jie yang
9 篇 stéphane cotin
9 篇 dan popescu
9 篇 mohamed kamel
8 篇 umapada pal
8 篇 aurélio campilho
8 篇 yefeng zheng
8 篇 kok wai wong
8 篇 christian barill...
8 篇 ilias maglogiann...
8 篇 tieniu tan

语言

4,851 篇 英文
296 篇 其他
65 篇 中文

检索条件"任意字段=4th International Conference on Computer Vision, Pattern Recognition and Image Processing"

共 5168 条记录，以下是771-780 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：

Natural and Imperceptible Backdoor Attack against Deep Neural Networks 4

Natural and Imperceptible Backdoor Attack against Deep Neura...

引用

4th international conference on Electronic Communication and Artificial Intelligence, ICECAI 2023

作者： Ni, Shifeng Wang, Xin Shang, Yifan Zhang, Linji Nanjing University of Aeronautics and Astronautics College of Computer Science and Technology Nanjing China

ISBN: (纸本)9798350314496

Over the past few years, deep learning has demonstrated impressive performance across a wide range of applications. As the same time, researchers have become increasingly concerned with ensuring the security of deep learning models. Specifically, backdoor attacks have emerged as a significant threat to Deep Learning Model, whereby attackers control the predictions of model by implanting a concealed backdoor into the model. Under this scenario, the backdoored model will appear to make normal predictions on clean images, but will exhibit abnormal behaviors when the trigger is presented. Many existing backdoor attacks utilize a fixed pattern as the trigger, which can be easily detected by defense methods or even humans. Furthermore, existing backdoor attack methods are rarely targeted specifically at the vision Transformer (ViT) model. therefore, in this paper, we propose a novel natural backdoor attack method. We exploit natural phenomena to carry out a backdoor attack called the fog backdoor attack, which can utilize fog in the natural world as a trigger to be seamlessly integrated into clean images without being perceived by humans. the generated fog will diffuse over the image, creating a natural-looking effect. In contrast to the fixed and limited triggers produced by other methods, our triggers are more natural and imperceptible. Experimental results demonstrate the effectiveness and robustness of the proposed backdoor attack on different models. Specifically, the attack success rate of the proposed backdoor attack is 98.85% on VGG-16 model, 99.44% on ResNet-18 model and 9S.56% on ViT model, respectively. Furthermore, the proposed attack does not compromise the clean accuracy of the model. © 2023 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Integrating computer vision and Crowd Sourcing to Infer Drug Use on Streets: A Case Study with 311 Data in San Francisco 3rd

Integrating Computer Vision and Crowd Sourcing to Infer Dr...

引用

3rd international conference on Intelligent Edge processing in the IoT Era, Edge-IoT 2022, and the 4th international conference on Smart Governance for Sustainable Smart Cities, SmartGov 2022

作者： Yi, Hye Seon Bhuiyan, Tanvir Chellappan, Sriram University of South Florida TampaFL33620 United States

ISBN: (纸本)9783031359811

Use of illegal drugs has been a bane of society for decades, and the problem is only increasing globally. Billions of dollars, time, effort and expertise among various stakeholders are being invested to combat this problem, but novel approaches are still needed. In this paper, we investigate the exciting possibility of leveraging big-data provided by the crowd to detect illegal drug use on streets. Our discussions with stakeholders revealed that drug use can be inferred via detection of any of the following components - syringes (including barrels and plungers), needles, needle caps, vials and tourniquets. We also learnt that in localities, where drug is common, these items are casually discarded on streets. Motivated by these, we conducted an investigation centered around San Francisco. We collected images from their SF311 platform where the public can actually upload their observations in the city for service. From more than 5 million images uploaded, we extracted 1, 759 images that contained one or more of drug use components mentioned above. We designed and validated state of the art computer vision techniques to both detect and localize components of interest with an uploaded image. Results demonstrate high utility, as we report later in the paper. To the best of our knowledge, our study is amongst the first to tackle the problem of detecting drug use automatically via AI techniques from crowd uploaded images. We believe that our study has practical impact. © 2023, ICST Institute for computer Sciences, Social Informatics and Telecommunications Engineering.

关键词： Needles

来源：评论

学校读者我要写书评

暂无评论

Vehicle Detection and Tracking in Real-time using YOLOv4-tiny 7

Vehicle Detection and Tracking in Real-time using YOLOv4-tin...

引用

7th international conference on image and Signal processing and their Applications, ISPA 2022

作者： Amrouche, Aissa Bentrcia, Youssouf Abed, Ahcene Hezil, Nabil Scientific and Technical Research Centre for the Development of Arabic Language Crstdla Algiers Algeria Usdb Signal and Communication Laboratory Blida Algeria

ISBN: (纸本)9781665480420

Vehicle detection and tracking is a popular research topic in Intelligent Transportation Systems. the goal of this paper is to detect, identify, and track vehicles in surveillance camera footage in order for them to be extracted efficiently and accurately. You Only Look Once version 4 (YOLOv4) algorithm is used in this paper to propose a real-time vehicle detection and tracking system. the suggested system has been evaluated using a variety of measures, including accuracy, precision, and recognition recall. For the experimental data, the system attained an accuracy of 96.30 percent and an overall accuracy of 94.17 percent. the results reveal that the suggested system successfully tracks vehicles in the scene. © 2022 IEEE.

关键词： computer vision

来源：评论

学校读者我要写书评

暂无评论

image Defogging Algorithm Based on Inception Mechanism 4th

Image Defogging Algorithm Based on Inception Mechanism

引用

4th EAI international conference on 6G for Future Wireless Networks, 6GN 2021

作者： Geng, Jiahao Miao, Zhuang Lin, Kezheng School of Computer Science and Technology Harbin University of Science and Technology Harbin China

ISBN: (纸本)9783031042447

In order to solve the problem that the existing defogging algorithms can’t differentiate according to the characteristics of different regions of the fogged image, an image Defogging Algorithm Based on Inception Mechanism is proposed(I-defog algorithm). the attention mechanism is added to the algorithm to adaptively assign weights to the features of different regions;It is more accurate and effective to use the module with perception mechanism to predict the global value. the predicted value, transmittance and foggy image are input into the atmospheric scattering model to get the defogging image, and the defogging image is input into the Markov discriminant (PatchGAN) to judge whether it is true or not. the results show that the algorithm achieves good defogging effect on both indoor and outdoor images, and improves the brightness and saturation of defogging images. © 2022, ICST Institute for computer Sciences, Social Informatics and Telecommunications Engineering.

关键词： pattern recognition

来源：评论

学校读者我要写书评

暂无评论

Research on Automatic processing and recognition of Complex images Combined with Deep Learning

Research on Automatic Processing and Recognition of Complex ...

引用

Data Science and computer Application (ICDSCA), IEEE international conference on

作者： Mingrui Zhu Qingchen Yang Zhizhong Duan South China University of Technology Guangzhou China University of Electronic Science and Technology of China Chengdu China National University of Defense Technology Changsha China

ISBN: (数字)9798350368239

ISBN: (纸本)9798350368246

With the rapid development of information technology, image processing technology plays an increasingly important role in many fields. However, traditional methods face many limitations when dealing with complex and fuzzy images, and it isn't easy to achieve accurate recognition and segmentation. Aiming at this challenge, this study developed an improved YOLOv5 complex image segmentation model with efficient complex image processing and segmentation using advanced image processing and deep learning technologies. According to the image characteristics, this study carried out efficient image preprocessing, including several steps such as grayscale processing, noise removal, edge enhancement, binarization, and morphological processing, to significantly improve the image quality and highlight key features to optimize the effective recognition area. then, this study introduced the YOLOv5 object detection model of the SE attention mechanism with the processed image as the input end. the experimental results show that the model has made remarkable achievements in key indicators such as precision and recall rate, which not only shows the strong advantages of deep learning in processing complex images but also provides strong support for further research.

关键词： Deep learning YOLO image segmentation Attention mechanisms Accuracy Text recognition Target recognition image edge detection Face recognition Noise

来源：评论

学校读者我要写书评

暂无评论

MRCI: Multi-range Context Interaction for Boundary Refinement in image Segmentation 27th

MRCI: Multi-range Context Interaction for Boundary Refineme...

引用

27th international conference on pattern recognition, ICPR 2024

作者： Wu, Yaqiang Lyu, Wanjun Liang, Xianchen Zheng, Qinghua Wei, Jin Jin, Lianwen Xi’an Jiaotong University Xi’an China Lenovo Research Beijing China Beijing University of Posts and Telecommunications Beijing China South China University of Technology Guangzhou China

ISBN: (纸本)9783031801358

In the era of foundational image segmentation models, there is a pressing need to leverage the outputs of these models and enhance the boundary accuracy of domain-specific segmentation results using lightweight post-processing techniques. Numerous existing boundary refinement approaches neglect the significance of incorporating diverse contextual scopes and global knowledge, resulting in restricted adaptability to different coarse segmentation errors. Moreover, the prevailing models are often lacking in lightweight design. To address these challenges, we propose a novel framework called Multi-Range Context Interaction (MRCI) that aims to refine the boundaries of predicted masks by incorporating comprehensive context knowledge while maintaining computational efficiency. Our approach utilizes a multi-range context-aware strategy to extract more informative local features and incorporates global knowledge prompts to guide the boundary refinement process. Experimental results on the widely used Cityscapes, ADE20K and satellite remote sensing dataset SpaceNet demonstrate the effectiveness of our approach, achieving top-tier Average Precision (AP) and mean IoU among the current state-of-the-art boundary refinement models while utilizing only 4M parameters. the source code will be available. © the Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Privacy and Data Control on Social Networks using Deep Learning 4

Privacy and Data Control on Social Networks using Deep Learn...

引用

4th international conference on Cybernetics, Cognition and Machine Learning Applications, ICCCMLA 2022

作者： Parimala, A. Vijayalata, Y. Gokaraju Rangaraju Institute of Engineering and Technology Dept. of Computer Science and Engineering Telangana Hyderabad India

ISBN: (数字)9781665462464

ISBN: (纸本)9781665462464

Online social networks are huge data exchange platforms that help to promote and share a lot of good information about products, news, education, tourism, health care, etc., also there is a great risk involved to individual's privacy and security. Online posted photos can be shared, tagged, and reposted again without having any consent from the people present in the images. In this project, we are proposing a facial recognition-based system that detects every single individual and checks their online account relationship between the individual. the post or image that is reposted, tagged, or copied by known or unknown individuals will be checked. If it is an unknown individual the images convolved through a low-pass filter kernel which removes the high-frequency content like noise and edges to smooth the images that make the image blur. For this mechanism, we proposed to use deep learning based VGG19 architecture for image classification and Haar cascade viola jones algorithm for detection, and computer vision-based blurring techniques used. algorithm performance as the area under the ROC curve (AUC) accuracy of 88% obtained. this proposed model achieved 91.53% classification accuracy and further we evaluated the model using Accuracy, Precision, Recall and F1-Score metrices and achieved 70.83%, 82%, 71% and 0.68%. © 2022 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Object Detection and Segmentation for Scene Understanding via Random Forest

Object Detection and Segmentation for Scene Understanding vi...

引用

international conference on Advancements in Computational Sciences (ICACS)

作者： Bisma Riaz Chughtai Ahmad Jalal Dept. of Computer Science Air University Islamabad Pakistan

In recent days, object detection become a vast topic in computer vision. Accurate object detection and scene understanding is not an easy task due to illumination, viewpoints, and color intensities. Visual features like color, texture, boundaries, and shape, make an image different from another image. the main goal of scene understanding is to machine work like a human and understand the visual information of an image. Currently, researchers working on novel approaches in this field to make a better understanding of the scene. computer vision portrays a major role in different applications such as health, safety, security surveillance, traffic monitoring, autonomous driving car, object recognition, and tracking. In this paper, we work on a meaningful understanding of an image in the scene. To understand the scene we have done region-based segmentation, and for object detection and labeling, we use the tensor flow algorithm, for geometric features mean of each pixel, harry corner edge detection, and scale-invariant feature transform descriptor. And then object recognition by using random forest. We have performed this experiment on UIUC Sports dataset. the presented model achieved 89.45% recognition accuracy.

关键词： image segmentation Visualization computer vision Tensors image color analysis Object detection Transforms

来源：评论

学校读者我要写书评

暂无评论

Optic Disc Segmentation Based on Mask R-CNN in Retinal Fundus images 4

Optic Disc Segmentation Based on Mask R-CNN in Retinal Fundu...

引用

4th international conference on Biomedical Engineering, IBIOMED 2022

作者： Pande Darma Suardika, I Gede Dendi Maysanjaya, I Md. Antara Kesiman, Made Windu Universitas Pendidikan Ganesha Virtual Vision Image and Pattern Research Group Faculty of Engineering and Vocational Bali Indonesia

ISBN: (数字)9781665460798

ISBN: (纸本)9781665460798

An optic disc is an object on the retina of the eye that has the characteristics of being brightly colored and round. Optical disc segmentation is the most commonstep taken before processing a retinal fundus image. the bright characteristics of the optic disc often interfere withthe detection of other objects in the retinal fundus image. therefore, the optic disc is the first step before processingthe fundus image of the retina. With the help of digital image processing will help in the removal of the optic discon the fundus image of the retina. Many methods can be used in optical disc segmentation, one of which is the deep learning method. the deep learning method chosen is Mask R-CNN to produce a mask from the results of object detection on the retinal fundus image. there are 3 stages in the segmentation process using the Mask R-CNN. First, the data used in the training process will be labeled. thereis 1 label given, namely optic disc. then the model is trained using the restnet50 backbone architecture and finally, the model will be evaluated. To evaluate the results obtained from the two methods, it uses Intersection over Union (IoU) by comparing directly the results of prediction and ground truth. the data used is an IDRiD dataset containing retinal fundus images taken from eye clinics across India. As the result, Mask R-CNN can segment the optical disc with an IoU value of 0.843. it is hoped that the results of this research can help the process in processing retinal fundus images in the future. © 2022 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Research on Oracle Bone Inscription Segmentation and recognition Model Based on Deep Learning

Research on Oracle Bone Inscription Segmentation and Recogni...

引用

Electronic Technology, Communication and Information (ICETCI), IEEE international conference on

作者： Donghao Li Binfang Du Shandong University Qingdao China

ISBN: (数字)9798350361643

ISBN: (纸本)9798350361650

In this paper, the automatic segmentation and recognition of oracle bone inscription images are studied. the Oracle bone inscription image segmentation model and Oracle bone inscription text recognition model are established by using computer vision technology and deep learning framework, and the application potential and scientific value of deep learning in the field of complex text image recognition are demonstrated. Firstly, we use a variety of digital image processing methods such as bilateral filtering technology and contrast-limited adaptive histogram equalization technology to optimize the recognition conditions of oracle bone inscription images. Secondly, we use the advanced Yolov8 to build a detection model to improve the text detection accuracy in oracle bone inscription images. the effect of the model was evaluated from different dimensions, such as accuracy rate, recall rate, F1 value, etc., to verify the performance of the model. Finally, we trained a model based on Resnet18 architecture to recognize the single word of the oracle bone inscription original topology image. the recognition results were displayed in the form of statistics and graphics, and the prediction accuracy of each character and the overall performance evaluation of the model were recorded in detail, thus verifying the effectiveness and practicability of the proposed method.

关键词： Deep learning image segmentation Adaptation models Accuracy Text recognition image edge detection Noise

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 << < 74 75 76 77 78 79 80 81 82 83 > >>

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：