检索结果-内蒙古大学图书馆

4th International Congress on Human-Computer Interaction, Optimization and Robotic applications, HORA 2022

作者： Salim, Alaa Sabeeh Abdulkareem, Mohammed Basil Fadhel, Yarub Essam Abdulkarem, Abdulkarem Basil Shantaf, Ahmed Muhi Abdulkareem, Ahmed B. Ministry of Education Dept. Computer Science Al-Diwaniyah-Al-Qadisiyah Iraq Al-Maarif University College Dept. Computer Engineering Techniques Ramadi Iraq CITIC University of Granada Dept. Computer Science Techniques Baghdad Iraq Continuous Learning Centre University of Anbar Ramadi Iraq

ISBN: (纸本)9781665468350

With advances in artificial intelligence field and computer vision, image captioning (IC) tool has progressively attracted researchers' attention. IC automatically generates natural text descriptions according to the image content. IC combines the knowledge of computer vision and natural language processing. In this article, a Novel image Captioning system was developed. The final system was validated on Flicker8K dataset. The novel designed system consists of Long Short Time Memory (LSTM) and VGG16 with Convolution neural Network (CNN). The main improvements of this system are in structure of designed system by adapting batch size. Also, studying deep learning parameters such as Regularization terms that can be added to loss function, CNN Optimizers and Dropout layer. The results showed the effectiveness of the designed system. Finally, this article highlighted some open challenges in the describing images task. © 2022 IEEE.

关键词： image annotation

来源：评论

学校读者我要写书评

暂无评论

Training a neural Network to Predict House Rents Using Artifical Intelligence and Deep Learning

Training a Neural Network to Predict House Rents Using Artif...

引用

image processing and Computer applications (ICIPCA), IEEE International Conference on

作者： Chung-Hsing Chao Department of Intelligent Vehicles and Energy Minth University of Science and Technology Hsinchu Taiwan China

We are training neural networks to predict house rents using artificial intelligence and deep learning, which is being used in the real estate and financial industries. Real estate agents, financial institutions, and real estate developers can benefit from rent forecasting, which is an important application scenario. Many real estate websites now use machine learning models to predict rents to help tenants and landlords find the right rental price. Machine learning models are also being used by some real estate companies and financial institutions to determine rents more accurately. Training neural networks to predict rent is still a relatively new field that needs more research and experimentation to improve its performance and accuracy.

关键词：

来源：评论

学校读者我要写书评

暂无评论

An Action Recognition Algorithm Based on Two-Stream Deep Learning for Metaverse applications 20

An Action Recognition Algorithm Based on Two-Stream Deep Lea...

引用

20th IEEE International Wireless Communications and Mobile Computing (IEEE IWCMC) - Green and Intelligent Communications

作者： Liu, Jiayue Mao, Tianqi Huang, Yicheng He, Dongxuan Beijing Inst Technol MIIT Key Lab Complex Field Intelligent Sensing Beijing 100081 Peoples R China Beijing Inst Technol Jiaxing Yangtze Delta Reg Acad Jiaxing 314019 Peoples R China Tsinghua Univ Dept Elect Engn Beijing 100084 Peoples R China Beijing Inst Technol Sch Informat & Elect Beijing 100081 Peoples R China

ISBN: (纸本)9798350361261;9798350361278

Action recognition algorithms have gained significant attention in recent years, which can be indispensable for a plethora of cutting-edge applications like extended reality or Metaverse. These services often pose stringent requirement on immediate sensing and cognition of the surroundings, which necessitates immediate classifications of the captured actions (e.g., video data) that classical signal processing methods can hardly attain. In this paper, we introduced a residual artificial neural network with two-stream structure to further improve the accuracy of action recognition algorithm. Specifically, two residual networks (ResNet101) are trained separately, one by spatial RGB image streams, and another by optical flow streams. The two-strem network outputs are then fed into a fusion classifier, in which information extracted by spatial network and temporal network jointly determines the classification result. Moreover, in the training process, hyper-parameters setting and optimizer selection are performed numerically to achieve optimal performance. Finally, the recognition accuracy of the proposed algorithm has been compared to other existing widely-employed counterparts, where UCF101 data set is utilized for training and testing. Simulations validates aiming that the network can achieve higher recognition accuracy than traditional algorithms, and the two-stream method shows its superiority over the single-network counterpart.

关键词： Action recognition deep learning residual network ResNet101 two-stream method

来源：评论

学校读者我要写书评

暂无评论

Efficient Multi-Scale Attention Residual Network for Breast Cancer Histopathology image Classification 23

Efficient Multi-Scale Attention Residual Network for Breast ...

引用

Proceedings of the 2023 International Conference on Advances in artificial Intelligence and applications

作者： Lu Cao Ke Pan School of Computer Science and Engineering Dalian Minzu University China

ISBN: (纸本)9798400708268

In recent years, convolutional neural networks have achieved significant success in the field of breast cancer histopathological image classification. This study reexamines the application of scale analysis and cross-spatial learning, introducing EMAResNet, a novel breast cancer classification approach based on multi-scale attention mechanisms. EMAResNet integrates feature grouping and cross-spatial learning, efficiently processing and distributing spatial semantic features while significantly reducing computational complexity. By parallel processing of multi-scale features and efficient feature fusion, the model effectively captures and utilizes crucial information in image data. This approach has been extensively evaluated on the BreakHis breast cancer pathology image dataset, demonstrating superior classification performance.

关键词： Breast Cancer Convolutional neural Network Efficient Multi-Scale Attention Histopathological image Classification

来源：评论

学校读者我要写书评

暂无评论

An Advanced Deep Learning Approach Combining image Analysis for Precise Retinal Disease Detection 1

An Advanced Deep Learning Approach Combining Image Analysis ...

引用

1st International Conference on artificial Intelligence for Innovations in Healthcare Industries, ICAIIHI 2023

作者： Kumar, Ankit Singh, Kamred Udham Singh, Teekam Kumar, Gaurav Pandey, Saroj Kumar Dwivedi, Pushpendra Dhar Gla University Department of Computer Engineering & Apllications UP Mathura India School of Computing Graphic Era Hill University Dehradun India Graphic Era Deemed to Be University Department of Computer Science and Engineering Dehradun India Raipur India

ISBN: (纸本)9798350330915

The purpose of this research is to showcase a unique hybrid image-centric deep learning system that was developed for the purpose of accurately identifying and diagnosing retinal diseases. Our method overcomes the limits that are connected with conventional imaging and computational analytics, specifically with regard to the accuracy of diagnostics and the level of processing speed. Our technique incorporates optimised models for retinal pictures in a seamless manner by combining the capabilities of convolutional neural networks (CNNs) with the flexibility of transfer learning. These models are then supplemented by advanced feature extraction algorithms. This new development makes it easier to accurately recognise small visual differences in retinal pictures, which are often missed by traditional approaches. Our study, which is based on a large dataset that includes a wide range of retinal anomalies, guarantees not only the correctness of our solution but also its flexibility to a variety of different situations. By demonstrating considerable gains in diagnostic accuracy, sensitivity, and specificity, as well as a notable decrease in the amount of manual feature engineering efforts, rigorous testing speeds up the diagnostic process without sacrificing its dependability. In addition to its use in retinal diagnosis, this model shows the potential of combining deep learning with conventional image analysis. It suggests that this combination might have a transformational influence on digital healthcare, with applications that could be extended to greater issues in medical imaging. In the rapidly developing field of digital healthcare, our results shed light on a potentially fruitful avenue. This avenue involves the use of advanced machine learning methods for the purpose of improving medical diagnosis. © 2023 IEEE.

关键词： Accuracy CNN Data Optimization Deep Learning Feature Extraction Hybrid image Ophthalmological Retinal Disease SVM

来源：评论

学校读者我要写书评

暂无评论

Evading Deepfake Detectors via High Quality Face Pre-processing Methods 26

Evading Deepfake Detectors via High Quality Face Pre-Process...

引用

26th International Conference on Pattern Recognition / 8th International Workshop on image Mining - Theory and applications (IMTA)

作者： Kim, Jeongho Kim, Taejune Kim, Jeonghyeon Woo, Simon S. Sungkyunkwan Univ Artificial Intelligence Suwon South Korea Sungkyunkwan Univ Comp Sci & Engn Suwon South Korea Sungkyunkwan Univ Appl Data Sci Suwon South Korea

ISBN: (数字)9781665490627

ISBN: (纸本)9781665490627

Today, various multimedia content can be accessed and shared from any location via the Internet. In addition to normal content, an extensive amount of manipulated multimedia can raise various social concerns. Among the various types of manipulated media, deepfakes can be abused in impersonation or spreading fake information. Therefore, numerous studies have been performed to detect deepfakes to alleviate these concerns, and studies such as FaceForensics++ (FF++) and DeepFake Detection Challenge (DFDC) have sparked these studies by providing deepfake datasets. The deepfake datasets were utilized for supervised learning in conjunction with developing sophisticated neural networks and showed a high detection performance. Since powerful neural networks can learn even subtle details about an image, they must be trained on realistic deepfakes created by advanced deepfake generation technologies to improve the robustness of existing detectors. In order to boost the performance of deepfake detection models, we propose an approach to creating more realistic deepfake images by removing "detectable" artifacts from existing deepfake datasets' images, including GAN-based images. By applying the proposed method to the original deepfake dataset, we demonstrate that our technique can significantly reduce the detection performance of existing deepfake detectors. Our experimental results show the vulnerability of deployed detectors and pave the way for further improvement.

关键词： Deepfakes Face recognition neural networks Supervised learning Pipelines Detectors Media

来源：评论

学校读者我要写书评

暂无评论

36th International Florida artificial Intelligence Research Society Conference, FLAIRS-36 2023

36th International Florida Artificial Intelligence Research ...

引用

36th International Florida artificial Intelligence Research Society Conference, FLAIRS-36 2023

The proceedings contain 132 papers. The special focus in this conference is on Florida artificial Intelligence Research Society. The topics include: Causanom: Anomaly Detection With Flexible Causal Graphs;answering Student Queries with a Supervised Memory Conversational Agent;topological Data Analysis in Natural Language processing – A Tutorial;k-medianoids Clustering Algorithm;automatic Generation of Paragraph Templates Using a Textual Energy Approach: applications in Literary Text Production;on Bounding the Behavior of a Neuron;semantics-based Framework for Incentivized Research Data Sharing;considerations in the Design of Realistic Agents for Serious Games;impact of artificial Intelligence Regulations on Organizational Risks;an Interactive Interpretation Viewer for Typed First-order Logic;security of Text to Icon/image conversions;Code Generation for Collectible Card Games with Complex APIs;a Study of Compressed Language Models in Social Media Domain;biogeography-based optimization for feature selection;malicious or Benign? Towards Effective Content Moderation for Children’s Videos;a Multi-Dictionary Approach to Abstractness/Concreteness-Based Authorship Attribution;bayesian Network Conflict Detection for Normative Monitoring of Black-Box Systems;relative Performance of Bilateral Multiattribute Negotiation Strategies in Open Markets;relative Effects of Positive and Negative Explanations on Satisfaction and Performance in Human-Agent Teams;ARCHIMEDES: Active Reasoning Conducted with Heterogeneous Information Monitors for Evading Dangers and Extended Surveillance;increasing Fairness in Predictions Using Bias Parity Score Based Loss Function Regularization;visual Episodic Memory-based Exploration;A Comparison of AutoML Hyperparameter Optimization Tools For Tabular Data;DynamicG2B: Dynamic Node Classification with Layered Graph neural networks and BiLSTM;unsupervised Keyword Extraction for Hashtag Recommendation in Social Media;deep Separable Hypercomplex networks.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multi-step implicit Adams predictor-corrector network for fire detection

引用

IET image processing 2022年第9期16卷 2338-2350页

作者： Deng, Zhen Hu, Shuhao Yin, Shibai Wang, Yibin Basu, Anup Cheng, Irene Ningxia Univ Dept Informat Engn Yinchuan Ningxia Peoples R China Southwestern Univ Finance & Econ Dept Econ Informat Engn Chengdu 611130 Sichuan Peoples R China Sichuan Normal Univ Dept Engn Chengdu Sichuan Peoples R China Univ Alberta Dept Comp Sci Edmonton AB T6G 2E8 Canada

Fire detection methods based on the Convolutional neural networks (CNN) have advantages of high accuracy, wide coverage and robustness, receiving significant attention from researchers. Among CNN-based methods, ResNet has achieved better performance than other CNN frameworks in fire detection system, since it uses stacked residual blocks to enlarge the receptive field to overcome the vanishing gradient problem with residual learning. The merits of ResNet can be attributed to the similarity between ResNet and the single-step explicit solver for Ordinary Differential Equations (ODEs), for example, the Euler method. Motivated by the theory of numerical ODE that a multi-step implicit solver has higher accuracy than a single-step explicit solver, the Multi-step Implicit Adams predictor-corrector (MIAPC) network for fire detection is proposed. The MIAPC method is first mapped to a corresponding predictor-corrector Adams block which achieves higher accuracy than a single-step explicit solver. Then, Adaptive Feature Fusion (AFF) and the Spatial Attention Layer (SAL) are utilized to extract hierarchical features from stacked predictor-corrector Adams blocks, forming the corresponding Adams module. Finally, the 4 Adams modules which are made of 4, 6, 8, 10 predictor-corrector Adams blocks and followed by AFF and SAL form the crucial ODE-based approximation part in the proposed network. By adding a simple feature extraction and detection in front of and after the ODE-based approximation part, the MIAPC network is built. Experiments demonstrate that the method achieves 87% accuracy in the challenging test dataset, outperforming existing methods by at least 6%. Besides, the 5.3M model size with inference speed of 4.7 frames/second in CPU and 65.7 frames/second in GPU enables the proposed method to be used in practical applications.

关键词： fire detection methods MIAPC network differential equations single-step explicit solver corresponding predictor-corrector Adams block residual blocks convolutional neural nets vanishing gradient problem simple feature extraction stacked predictor-corrector Adams blocks learning (artificial intelligence) crucial ODE-based approximation part fire detection system 10 predictor-corrector Adams blocks multistep implicit solver 4 predictor-corrector Adams blocks feature extraction ResNet deep learning (artificial intelligence) MIAPC method 8 predictor-corrector Adams blocks 6 predictor-corrector Adams blocks corresponding Adams module Euler method Spatial Attention Layer Convolutional neural networks Multistep Implicit Adams predictor-corrector network numerical ODE robustness CNN frameworks CNN-based methods Multistep implicit Adams predictor-corrector network residual learning 4 Adams modules

来源：评论

学校读者我要写书评

暂无评论

An Optimized Hybrid approach to Detect Cataract

An Optimized Hybrid approach to Detect Cataract

引用

2022 IEEE Global Conference on Computing, Power and Communication Technologies, GlobConPT 2022

作者： Pilania, Urmila Diwakar, Chetna Arora, Khushi Chaudhary, Shruti Manav Rachna University Computer Science Department Haryana Faridabad India

ISBN: (数字)9781665493659

ISBN: (纸本)9781665493659

Cataract, a disease in which the lens of the eye slowly turns opaque resulting in blurry vision, is one of the most occurring diseases among older people. According to the sources, more than 50% of people aged 80 or more have had eye cataract. While cataracts only used to affect elderly people, nowadays it is prevalent to see them in ages below 50 due to modern lifestyles, lack of exercise, continuous usage of mobile phones and computers. Over 12 million people are blind in India and 62.6% of blindness is a result of cataract. If the detection is done before the final stages of cataract, this disease can be cured in less time. artificial Intelligence (AI) is a significant tool for the development of computer-aided technology. The retinal image is an important medical reference that helps in diagnosing cataract. This paper proposes a hybrid method using the Fuzzy Technique, Retinal image Quality Assessment (RIQA), and Convolutional neural networks (CNN). The fuzzy technique is used for image quality improvement, RIQA is used for image quality assessment and CNN is used for processing of the retina to automatic cataract detection. © 2022 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Diversifying Inference Path Selection: Moving-Mobile-Network for Landmark Recognition

引用

IEEE TRANSACTIONS ON image processing 2021年 30卷 4894-4904页

作者： Qian, Biao Wang, Yang Hong, Richang Wang, Meng Shao, Ling Hefei Univ Technol Key Lab Knowledge Engn Big Data Minist Educ Hefei 230000 Peoples R China Hefei Univ Technol Dept Comp Sci & Informat Engn Hefei 230000 Peoples R China Incept Inst Artificial Intelligence Abu Dhabi U Arab Emirates

Deep convolutional neural networks have largely benefited computer vision tasks. However, the high computational complexity limits their real-world applications. To this end, many methods have been proposed for efficient network learning, and applications in portable mobile devices. In this paper, we propose a novel Moving-Mobile-Network, named M(2)Net, for landmark recognition, equipped each landmark image with located geographic information. We intuitively find that M(2)Net can essentially promote the diversity of the inference path (selected blocks subset) selection, so as to enhance the recognition accuracy. The above intuition is achieved by our proposed reward function with the input of geo-location and landmarks. We also find that the performance of other portable networks can be improved via our architecture. We construct two landmark image datasets, with each landmark associated with geographic information, over which we conduct extensive experiments to demonstrate that M(2)Net achieves improved recognition accuracy with comparable complexity.

关键词： image recognition Convolution Mobile handsets image coding Visualization Computational modeling Network architecture Deep learning network compression and acceleration landmark recognition mobile devices

来源：评论

学校读者我要写书评

暂无评论

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：

请选择收藏分类：

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

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：