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

4,146 篇 会议
73 篇 期刊文献
10 册 图书

馆藏范围

4,229 篇 电子文献
0 种 纸本馆藏

日期分布

学科分类号

2,938 篇 工学
- 2,152 篇 计算机科学与技术...
- 1,311 篇 软件工程
- 717 篇 信息与通信工程
- 481 篇 电气工程
- 452 篇 光学工程
- 398 篇 生物医学工程（可授...
- 350 篇 生物工程
- 287 篇 控制科学与工程
- 203 篇 电子科学与技术（可...
- 188 篇 机械工程
- 188 篇 仪器科学与技术
- 128 篇 测绘科学与技术
- 120 篇 化学工程与技术
- 111 篇 材料科学与工程（可...
- 99 篇 土木工程
- 89 篇 建筑学
- 82 篇 航空宇航科学与技...
- 81 篇 环境科学与工程（可...
- 71 篇 安全科学与工程
- 68 篇 网络空间安全
1,312 篇 理学
- 521 篇 数学
- 512 篇 物理学
- 384 篇 生物学
- 173 篇 统计学（可授理学、...
- 139 篇 化学
488 篇 管理学
- 285 篇 图书情报与档案管...
- 223 篇 管理科学与工程(可...
445 篇 医学
- 364 篇 临床医学
- 187 篇 基础医学(可授医学...
- 134 篇 药学(可授医学、理...
71 篇 农学
64 篇 法学
25 篇 经济学
23 篇 教育学
12 篇 艺术学
10 篇 军事学
7 篇 文学
1 篇 哲学

主题

281 篇 image reconstruc...
238 篇 image segmentati...
214 篇 feature extracti...
188 篇 deep learning
180 篇 data mining
170 篇 image processing
133 篇 image classifica...
126 篇 training
118 篇 convolutional ne...
110 篇 accuracy
106 篇 data models
92 篇 image recognitio...
88 篇 image enhancemen...
81 篇 remote sensing
73 篇 image analysis
72 篇 computer vision
68 篇 image color anal...
65 篇 machine learning
63 篇 computational mo...
61 篇 visualization

机构

13 篇 chitkara univers...
12 篇 chitkara univ in...
6 篇 school of comput...
6 篇 chitkara univers...
6 篇 school of data a...
5 篇 vishwakarma inst...
5 篇 graphic era deem...
5 篇 uttaranchal univ...
4 篇 beijing institut...
4 篇 china academy of...
4 篇 chennai
4 篇 wuhan univ sch r...
4 篇 center for earth...
4 篇 univ chinese aca...
4 篇 computer science...
4 篇 geomatics engine...
4 篇 cccc big data te...
4 篇 university of ch...
4 篇 sun yat sen univ...
4 篇 waseda universit...

作者

8 篇 li jonathan
7 篇 wang cheng
7 篇 zhang wei
6 篇 sharma neha
6 篇 li li
6 篇 wang wei
6 篇 wang yan
6 篇 li yan
6 篇 chen yen-wei
5 篇 wang jing
5 篇 gupta rupesh
5 篇 b. solaiman
5 篇 m.s. bouhlel
5 篇 rueckert daniel
5 篇 zhang jun
4 篇 vibhute amol d.
4 篇 zhang lei
4 篇 gill kanwarparta...
4 篇 xie bin
4 篇 chen chen

语言

4,038 篇 英文
168 篇 其他
33 篇 中文
2 篇 法文
1 篇 泰文

检索条件"任意字段=2nd Conference on Image Reconstruction from Incomplete Data"

共 4229 条记录，以下是321-330 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：

Colon Polyp image Segmentation Method Based on CNN and Transformer 2

Colon Polyp Image Segmentation Method Based on CNN and Trans...

引用

2nd IEEE International conference on image Processing and Computer Applications, ICIPCA 2024

作者： Zhao, Hong Mi, Shan An, Ding Tian, Weiquan Lanzhou University of Technology Lanzhou China Cccc Big Data Technology Co. Ltd. Lanzhou China

ISBN: (纸本)9798350360240

To solve the problem of low accuracy in traditional pure convolutional neural network models for intestinal polyp segmentation images, we propose a dual-branch multi-scale intestinal polyp segmentation model based on CNN and BiFormer. The model primarily consists of two main components: an encoder and a decoder. During the encoding stage, BiFormer employs a dual-layer routing attention mechanism as the multi-scale feature extractor in the encoder branch to capture global information. Simultaneously, a convolutional block is designed as the convolutional encoder branch to extract local semantic information and detailed features of polyp images. Additionally, an Enhanced Spatial Attention (ESA) mechanism is introduced in the convolutional block of each encoder layer to effectively utilize local spatial information and enhance the spatial localization accuracy of polyps, thereby further improving the model's segmentation performance. In the decoding stage, a Skip Fusion Block (SFB) is employed to comprehensively integrate semantic features from both the encoder and decoder, addressing issues such as discontinuous boundary segmentation and further enhancing segmentation accuracy. Our method was evaluated on the open-source datasets Kvasir-SEG, CVC-ClinicDB, and CVC-ColondB, achieving Dice coefficients of 0.888, 0.927, and 0.911, respectively, and mean Intersection over Union (mIoU) scores of 0.876, 0.902, and 0.860, respectively. Compared to traditional convolutional neural network-based models for intestinal polyp segmentation, our method demonstrates improvements of 2.1% in Dice coefficient and 2.5% in mIoU, highlighting its higher practical value. © 2024 IEEE.

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

Damaged image Repair using Masks with Computer Vision Inpaint Method 2

Damaged Image Repair using Masks with Computer Vision Inpain...

引用

2nd International conference on Applied Artificial Intelligence and Computing, ICAAIC 2023

作者： Ramakrishna, Doddi Purna Gana Emmanuel, G. S. Michael Selvan, Mercy Paul Sathyabama Institute of Science and Technology Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9781665456302

image inpainting is the technique used to automatically fix damaged areas using data from sections that have been saved. With the development of deep learning in recent years, image drawing performance has substantially increased. This research study reviews the main methods used for automating image inpainting. This research study provides a brief overview of traditional techniques while concentrating on deep learning-based inpainting techniques, covering model categorization, strengths and drawbacks, scope of application, and performance comparison. Finally, the challenges and trends surrounding automated image inpainting are examined and foreseen. A tool called image inpainting uses the data from the remaining components to repair damaged areas. With the advancement of society, image inpainting has become a vital research area in the field of computer vision. It is extensively used in culture, daily life, and security, including object removal and the preservation of digital cultural assets. Conventional methods build geometric models based on geometric consistency and image content similarity, or they use texture generation to patch up small sections of damaged images. It partially solves the problem of loose coupling between high-level semantics and low-level image properties, enabling deep learning to gradually overtake traditional methods in computer vision. © 2023 IEEE.

关键词： Computer vision

来源：评论

学校读者我要写书评

暂无评论

A Hybrid Approach to Validate and Retrieve Fingerprint image for Authenticating database Sensitive Information Access and to Verify Some of Topological Multi-Granular Rough Sets Properties on Authenticated database Fields 2

A Hybrid Approach to Validate and Retrieve Fingerprint Image...

引用

2nd International conference on Emerging Trends in Information Technology and Engineering, ic-ETITE 2024

作者： Raghavan, R. John Singh, K. School of Computer Science Engineering and Information Systems Vellore Institute of Technology Vellore India

ISBN: (纸本)9798350328202

Content based image retrieval has already begun to turn into an intriguing and exceptionally needed field to meet with the client prerequisite. Fingerprint approvals likewise have become a fascinating field to principally uphold the confirmation framework. Notwithstanding foundations regulating guaranteed about information straightforwardly, affiliations are also obliging them to examine the fingerprint endorsement. Despite establishments overseeing ensured about data, directly associations are similarly obliging them to research the fingerprint check. Organizations and foundations both are typically utilizing the idea of fingerprint for the participation and presence reason. Nevertheless, establishments presently managing secured information as like institutes and additionally going along with them to investigate the fingerprint authentication. More than this usual action there are some particular applications where the authentication would be considerable to help the application than using a traditional pattern of password. We proposed a fingerprint authentication system for the scientific based industry database access system. We removed the noise if any found in the input fingerprint using anisotropic Gauss filtering. The noise free fingerprint image is not further passed to the binarization phase as the need of binarization and some form of the drawback of binarization is replaced with a thinning algorithm. The thinned fingerprint image is followed by the matching process and once it is matched, then only the sensitive and highly confidential information are allowed to access from the database. The testing is done using OpenCV and FVC 2004 dataset to assess the projected retrieval method. Further the matched fingerprints are used to prove some of the internal structure properties of multi-granular rough sets (MGRS) on authenticated database fields. The accuracy obtained in this hybrid approach is found to be more satisfactory and hence the security towards acce

关键词： database systems

来源：评论

学校读者我要写书评

暂无评论

Enhanced Medical image Segmentation using CNN based on Histogram Equalization 2

Enhanced Medical Image Segmentation using CNN based on Histo...

引用

2nd International conference on Applied Artificial Intelligence and Computing, ICAAIC 2023

作者： Saifullah, Shoffan Drezewski, Rafal Institute of Computer Science Agh University of Science and Technology Kraków Poland Upn Veteran Yogyakarta Department of Informatics Indonesia

ISBN: (纸本)9781665456302

Medical image segmentation is an essential process in disease analysis and diagnosis, in which medical images are divided into specific areas of anatomic or pathological structures. However, this process has challenges, such as the quality, variability, and complexity of medical images. Thus, we propose a medical image segmentation method using the Convolutional Neural Network (CNN) and the Histogram Equalization (HE) image preprocessing. We evaluated medical image segmentation using 2 secondary datasets, namely Lung CT-Scan and Chest X-Rays, with 267 and 3616 images, respectively. The HE method uses an averaging of the optimal cumulative distribution function (CDF) values with a range of 0 to 39. The experimental results show that adding the HE method can increase the accuracy of CNN on the Lung CT-Scan and Chest X-rays data sets, respectively, by 0.73 percentage points from 91.28% to 92.08 % and 1.23 percentage points from 95.56% to 96.85% for the training process. The results of the testing process (validation) have an increase in accuracy of 2.82 percentage points from 89.77% to 92.59% and 0.91 percentage points from 96.06% to 96.97%. Implementing the HE method increases accuracy in CNN medical image segmentation based on the value of dice similarity coefficient (DSC) and structural similarity index measurement (SSIM) close to 0.95. © 2023 IEEE.

关键词： Computerized tomography

来源：评论

学校读者我要写书评

暂无评论

Weed Recognition Using image Patches Based Global Hybrid Attention with Densenet-169 Model 2

Weed Recognition Using Image Patches Based Global Hybrid Att...

引用

2nd IEEE International conference on Networks, Multimedia and Information Technology, NMITCON 2024

作者： Valiveti, Hima Bindu Almusawi, Muntather Satish, E.G. Aluvala, Srinivas Challaraj Emmanuel, E.S. Gokaraju Rangaraju Institute of Engineering and Technology Department of Ece Hyderabad India The Islamic University College of Technical Engineering Department of Computers Techniques Engineering Najaf Iraq Nitte Meenakshi Institute of Technology Dept. of Computer Science and Engineering Bengaluru India Sr University School of Computer Science and Engineering Warangal India Kristu Jayanti College Department of Life Sciences Bengaluru India

ISBN: (纸本)9798350372892

Weeds are one of the main problem in agriculture that affect the crop yield. Accurate classification and recognition is major challenges in weeding because of same visual characteristics between plants and weeds. To solve this problem, in this research patch based image recognition is integrated with the Deep Learning (DL) approach to improving the classification and recognition of weeds. The weed image data are collected from the Deepweeds dataset. For pre-processing Enhanced Super Resolution Generative Adversarial Network (ESRGAN) is utilized for change the image size and then generated to small patches. Selection of relative important patches are performed by Fast Fourier Transform (FFT) and Laplacian Filter (LF) are trained to the DL model. The proposed Global Hybrid Attention mechanism with Densenet-169 model (GHA-Densenet-169) employed for classification of weeds with respective classes. The evaluation results of the proposed method using performance metrics are Accuracy, Recall, Precision and F1-score. The proposed approach attained high accuracy of 99.28%, recall of 99.26%, precision of 99.27% and f1-score of 99.25% which is greater than existing deep learning models such as Densenet-201, VGG-SVM, Adapting MobilenetV2, Resnet-SVM, and Swin Transformer and Two-stage Transfer Learning (ST-TSTL). © 2024 IEEE.

关键词： and deep learning fast fourier transform and laplacian filter global hybrid attention mechanism image patches weed recognition

来源：评论

学校读者我要写书评

暂无评论

FPGA Implementation of DNA Computing and Genetic Algorithm Based image Encryption Technique 2nd

FPGA Implementation of DNA Computing and Genetic Algorithm B...

引用

2nd International conference on Biomedical Engineering Science and Technology (ICBEST) - Roadway from Laboratory to Market

作者： Rajashree, R. Durai, S. Ananiah VIT SENSE Chennai India VIT CNVD Chennai India

ISBN: (纸本)9783031545467;9783031545474

In the digital age, an increasing dependency of humans on digital technologies is much prevalent. A high demand in securing the private and confidential data is essential, which would otherwise lead to mishandling and misuse of personal information for ulterior motives. To avoid such vulnerabilities many Cryptographic and stenographic techniques were developed over the past decades. Though such existing schemes provided satisfactory shielding, high computational cost, increased processing time, low embedding capacity, and low imperceptibility, renders it as unsuitable for applications requiring high security and high speed such as medical data security. A hybrid DNA computing technology combined with a genetic algorithm is developed in this work to cater for medical image encryption, which will address the aforementioned limitations. The suggested approach converts plaintext information into binary by first mapping it to the DNA sequence. The binary information is then embedded in the cover image using genetic algorithm to yield a stego image. Elliptic Curve Cryptographic hardware (ECC) encrypts the generated stego image, providing a more secure means of storage/communication. The suggested hardware system is implemented in the Zynq 7000 FPGA device, which occupied a comparatively low overall LUT and DSP slices of around 5796 and 19 respectively.

关键词： DNA computation Genetic Algorithm Steganography image encryption Binary data conversion Elliptic Curve Cryptography

来源：评论

学校读者我要写书评

暂无评论

A Novel Method of Ground Target Positioning based on Heterologous image Registration 2

A Novel Method of Ground Target Positioning based on Heterol...

引用

2nd IEEE International conference on Signal, Information and data Processing, ICSIDP 2024

作者： Shang, Shize Wu, Han Chen, Yawei Cheng, Qiang Xia, Linghao Yang, Yuhao Nanjing Research Institute of Electronics Technology Nanjing210039 China National Key Laboratory of Radar Detection and Sensing Nanjing210039 China Jiangsu Provincial Key Laboratory of Detection and Perception Technology Nanjing210039 China

ISBN: (纸本)9798331515669

In this paper, we propose a target high-precision positioning method based on heterogeneous image registration, including target longitude, latitude and altitude. The longitude and latitude are obtained by registering video SAR images with optical map, and the positioning information on the optical map is precisely known. Based on the registration of video SAR images and optical video images, the target height can be solved according to the projection direction difference of the height target on the two types of images, and the target altitude is calculated by adding the ground altitude. The video SAR and optical video images come from the integrated sensor, and the high frame rate imaging of video SAR can simultaneously satisfy the positioning of moving and stationary targets. Finally, the positioning results of this method is verified by measured data. © 2024 IEEE.

关键词： image registration

来源：评论

学校读者我要写书评

暂无评论

Enhanced Feature Extraction for image Dehazing: A Comparative Study between Deep Learning Architectures and FFA-NET 2

Enhanced Feature Extraction for Image Dehazing: A Comparativ...

引用

2nd International conference on Inventive Computing and Informatics (ICICI)

作者： Chaudhary, Sarthak Gupta, Samridh Iniyan, S. SRM Inst Sci & Technol Dept Comp Sci & Engn Chennai 603203 Tamil Nadu India SRM Inst Sci & Technol Dept Comp Technol Sch Comp Chennai 603203 Tamil Nadu India

ISBN: (纸本)9798350373301;9798350373295

The problem of poor visibility in foggy images has spurred various image de-hazing strategies. As the need for high-quality images grows, especially for autonomous systems, this research aims to leverage different Deep Learning (DL) architectures to draw out key details from images, localizing this retrieved data to mitigate the impact of haze. The work explores using DL methods, particularly contrasting the regression and classification models of Convolutional Neural Networks (CNN), to remove haze from foggy images. This work sets the stage for further developments in image processing, particularly in conditions with poor visibility. It opens opportunities for improving image quality in various applications, such as autonomous driving and outdoor robotics, where clarity of vision is crucial. The final stage of the proposed model involves three specific pre-processing methods: contextual regularization, air light estimation and boundary constraint for optimal results. The next stage sets out to determine the best DL model for producing clear images from de-hazed ones.

关键词： Machine Learning (ML) Convolutional Neural Networks (CNN) Feature Fusion Attention Network(FFA-NET) Deep Learning Dark Channel Prior(DCP) Artificial Intelligence (AI)

来源：评论

学校读者我要写书评

暂无评论

Metric3D: Towards Zero-shot Metric 3D Prediction from A Single image

Metric3D: Towards Zero-shot Metric 3D Prediction from A Sing...

引用

IEEE/CVF International conference on Computer Vision (ICCV)

作者： Yin, Wei Zhang, Chi Chen, Hao Cai, Zhipeng Yu, Gang Wang, Kaixuan Chen, Xiaozhi Shen, Chunhua DJI Technol Shenzhen Peoples R China Tencent Shenzhen Peoples R China Zhejiang Univ Hangzhou Peoples R China Intel Labs Hillsboro OR USA

ISBN: (纸本)9798350307184

Reconstructing accurate 3D scenes from images is a long-standing vision task. Due to the ill-posedness of the single-image reconstruction problem, most well-established methods are built upon multi-view geometry. State-of-the-art (SOTA) monocular metric depth estimation methods can only handle a single camera model and are unable to perform mixed-data training due to metric ambiguity. Meanwhile, SOTA monocular methods trained on large mixed datasets achieve zero-shot generalization by learning affine-invariant depths, which cannot recover real-world metrics. In this work, we show that the key to a zero-shot single-view metric depth model lies in the combination of large-scale data training and resolving the metric ambiguity from various camera models. We propose a canonical camera space transformation module, which explicitly addresses the ambiguity problems and can be effortlessly plugged into existing monocular models. Equipped with our module, monocular models can be stably trained over 8 millions of images with thousands of camera models, resulting in zero-shot generalization to in-the-wild images with unseen camera settings. Experiments demonstrate SOTA performance of our method on 7 zero-shot benchmarks. Notably, our method won the championship in the 2nd Monocular Depth Estimation Challenge. Our method enables the accurate recovery of metric 3D structures on randomly collected internet images, paving the way for plausible single-image metrology. The potential benefits extend to downstream tasks, which can be significantly improved by simply plugging in our model. For example, our model relieves the scale drift issues of monocular-SLAM (Fig. 1), leading to high-quality metric scale dense mapping. The code is available at https://***/YvanYin/Metric3D.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Digital Watermarking Techniques for Secure image Distribution 2

Digital Watermarking Techniques for Secure Image Distributio...

引用

2nd IEEE International conference on Intelligent and Innovative Technologies in Computing, Electrical and Electronics, ICIITCEE 2024

作者： Baskar, Radhika Dwibedi, Rajat Kumar Kumar, T.Rajesh Vijayalakshmi, N. Rajesh Kanna, R. Reddy, Gosu Nageswara Saveetha Institute of Medical and Technical Sciences Saveetha School of Engineering Chennai India Vmrf Aarupadai Veedu Institute of Technology Chennai India Saveetha Institute of Medical and Technical Sciences Saveetha School of Engineering Department of Cse Chennai India Sns College of Technology Department of Cse Coimbatore India Department of Cse Bangalore India Vel Tech Rangarajan Dr.Sagunthala R&d Institute of Science and Technology Department of Chemistry Chennai India

ISBN: (纸本)9798350306415

In the contemporary era of digital advancements, it is of utmost importance to prioritize the establishment of robust security measures and traceability protocols for photos. This necessity arises from the inherent risk associated with the effortless diffusion of unlicensed information. Digital watermarking, which implants hidden data into digital photographs to verify their validity, is frequently used. This level emphasizes the need of safe photo distribution, digital platform problems, and unauthorized reproductions. The purpose of this research is to explain digital watermarking fundamentals. It emphasizes verification, IP protection, and digital watermarking monitoring. This research compares spatial and frequency domain watermarking approaches. Direct pixel manipulation in spatial domain techniques is vulnerable to attacks. Integrating watermarks with transform domains like Discrete Cosine Transform improves robustness in frequency domain techniques. The study also studies adaptive watermarking, which adjusts the watermark to the image's content to balance visibility and durability. The purpose of this research is to explore watermark identification methods. These methods use blind and non-blind watermarking. We discuss the security risks that might compromise watermarked photographs and the ways to reduce their likelihood. © 2024 IEEE.

关键词： Digital image Digital Water Marking image Distributions Pixel values

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423页 << < 29 30 31 32 33 34 35 36 37 38 > >>

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：