检索结果-内蒙古大学图书馆

2nd IEEE International Conference on Advanced Technologies in Intelligent Control, Environment, Computing and Communication Engineering, ICATIECE 2022

作者： Kumar, Krishna Marimuthu, M. Das Gupta, Ayan Pant, Bhasker Shukla, Surendra Kumar Kapila, Dhiraj Bansal Institute of Engineering and Technology Department of Computer Science and Engineering Lucknow India Coimbatore Institute of Technology Department of Computing-Data Science Coimbatore India Chandernagore Government College Department of Geography Hugli India Graphic Era Deemed to Be University Department of Computer Science and Engineering Dehradun India Lovely Professional University Department of Computer Science and Engineering Phagwara India

ISBN: (纸本)9781665493963

Image analysis tasks use salient object detection because it not only identifies important elements of a visual scene but also lessens computational complexity by removing unimportant elements. In this research, we propose a novel salient object recognition method based on a deep learning network that maintains picture information in the mid and low regions. Using a deep learning model, our technique generates a coarse saliency map for the entire target image. The map is then fine-Tuned utilising low-To-mid level information particular to the image. For detection of salient objects, we use a U-Net as our architecture. The saliency map can be predicted pixel by pixel, reducing low-level visual information loss. Our results show that our system regularly outperforms other approaches for detecting salient objects, resulting in superior precision and recall rates. © 2022 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Discriminatively Constrained Semi-Supervised Multi-View Nonnegative Matrix Factorization with Graph Regularization

引用

Big data Mining and Analytics 2024年第1期7卷 55-74页

作者： Guosheng Cui Ye Li Jianzhong Li Jianping Fan Shenzhen Institute of Advanced Technology Chinese Academy of SciencesShenzhen 518055China Joint Engineering Research Center for Health Big Data Intelligent Analysis Technology Shenzhen 518055China University of Chinese Academy of Sciences Beijing 100049China School of Computer Science and Control Engineering Shenzhen Institute of Advanced TechnologyChinese Academy of SciencesShenzhen 518055China

Nonnegative Matrix Factorization(NMF)is one of the most popular feature learning technologies in the field of machine learning and pattern *** has been widely used and studied in the multi-view clustering tasks because of its *** study proposes a general semi-supervised multi-view nonnegative matrix factorization *** algorithm incorporates discriminative and geometric information on data to learn a better-fused representation,and adopts a feature normalizing strategy to align the different *** specific implementations of this algorithm are developed to validate the effectiveness of the proposed framework:Graph regularization based Discriminatively Constrained Multi-View Nonnegative Matrix Factorization(GDCMVNMF)and Extended Multi-View Constrained Nonnegative Matrix Factorization(ExMVCNMF).The intrinsic connection between these two specific implementations is discussed,and the optimization based on multiply update rules is *** on six datasets show that the effectiveness of GDCMVNMF and ExMVCNMF outperforms several representative unsupervised and semi-supervised multi-view NMF approaches.

关键词： multi-view semi-supervised clustering discriminative information geometric information feature normalizing strategy

来源：评论

学校读者我要写书评

暂无评论

Robust Intelligent System for COVID-19 Detection using CT-Scan

Robust Intelligent System for COVID-19 Detection using CT-Sc...

引用

2023 International Conference Automatics and Informatics, ICAI 2023

作者： Al Smadi, Ahmad Abugabah, Ahed Al-Smadi, Ahmad Mohammad Zarqa University Dep. of Data Science and Artificial Intelligence Zarqa13100 Jordan College of Technological Innovation Zayed University Abu Dhabi United Arab Emirates Al-Balqa Applied University Ajloun University College Department of Computer Science Jordan

ISBN: (纸本)9798350312911

In the beginning of 2020, the world witnessed the rapid spread of the new coronavirus, COVID-19, affecting millions of people globally. However, at the outset, the availability of corona test kits was scarce, leading researchers to explore alternative detection methods. Among these methods, the COVID-19 detection approach using CT-scans emerged, and artificial intelligence (AI)-based solutions proved to offer superior outcomes. Despite the potential of AI-based models, the issue of overfitting arose, significantly impacting model performance. In response to this challenge, we present a coherent and cohesive solution in this paper, utilizing a Convolutional Neural Network (CNN)-based approach for accurate classification of COVID-19 vs. non-COVID cases. To enhance the model's robustness, we incorporated data augmentation and batch normalization techniques for regularization. To evaluate the effectiveness of our proposed model, we conducted experiments with four different data splitting ratios (50%-50;70%-30;75%-25;80%-20) for training and testing. As a result, our suggested model achieved an impressive classification accuracy of 98.56% for distinguishing between COVID-19 and non-COVID cases. These promising results highlight the efficacy of our CNN-based approach with regularization techniques. Furthermore, we conducted a comparative analysis with other deep learning-based algorithms, and our model consistently outperformed them, demonstrating its superiority in COVID-19 detection. By providing such reliable and accurate results, our proposed model contributes significantly to the ongoing efforts in combating the COVID-19 pandemic and holds the potential to aid healthcare professionals in timely and precise diagnosis. © 2023 IEEE.

关键词： COVID-19

来源：评论

学校读者我要写书评

暂无评论

Design Automation for Continuous-Flow Lab-on-a-Chip Systems: A One-Pass Paradigm

Design Automation for Continuous-Flow Lab-on-a-Chip Systems:...

引用

作者： Huang, Xing Pan, Youlin Chen, Zhen Guo, Wenzhong Wang, Lu Li, Qingshan Wille, Robert Ho, Tsung-Yi Schlichtmann, Ulf Technical University of Munich Electronic Design Automation Munich80333 Germany Fuzhou University College of Computer and Data Science Fuzhou350116 China Xidian University School of Computer Science and Technology Xi'an710071 China Technical University of Munich Chair for Design Automation Munich80333 Germany Software Competence Center Hagenberg GmbH Hagenberg4232 Austria The Chinese University of Hong Kong Department of Computer Science and Engineering Hong Kong

Owing to the high complexity of chip architecture and assay protocol, considerable effort has been directed toward the design automation of continuous-flow microfluidics over the past decade. Existing methods, however, perform the corresponding design tasks, including binding, scheduling, placement, and routing separately, leading to serious gaps between different steps and potentially even cause design failure. To overcome these drawbacks, in this article, we propose a one-pass design paradigm for continuous-flow microfluidic lab-on-a-chip systems, integrating all the design steps into an 'organic whole,' which has never been considered in prior work. With the proposed paradigm, all the design tasks can be synchronized seamlessly and performed in a combined manner, thereby eliminating the gaps between design steps. Consequently, optimized biochip architectures can be generated without any design adjustments and modifications. The experimental results demonstrate the effectiveness of the proposed automation flows. © 1982-2012 IEEE.

关键词： Microfluidics

来源：评论

学校读者我要写书评

暂无评论

Neural Collapse in Multi-label Learning with Pick-all-label Loss

arXiv

引用

arXiv 2023年

作者： Li, Pengyu Li, Xiao Wang, Yutong Qu, Qing Department of Electrical Engineering & Computer Science Michigan Institute for Data Science University of Michigan United States

We study deep neural networks for the multi-label classification (M-lab) task through the lens of neural collapse (NC). Previous works have been restricted to the multi-class classification setting and discovered a prevalent NC phenomenon comprising of the following properties for the last-layer features: (i) the variability of features within every class collapses to zero, (ii) the set of feature means form an equi-angular tight frame (ETF), and (iii) the last layer classifiers collapse to the feature mean upon some scaling. We generalize the study to multi-label learning, and prove for the first time that a generalized NC phenomenon holds with the "pick-all-label" formulation, which we term as M-lab NC. While the ETF geometry remains consistent for features with a single label, multi-label scenarios introduce a unique combinatorial aspect we term the "tag-wise average" property, where the means of features with multiple labels are the scaled averages of means for single-label instances. Theoretically, under proper assumptions on the features, we establish that the only global optimizer of the pick-all-label cross-entropy loss satisfy the multi-label NC. In practice, we demonstrate that our findings can lead to better test performance with more efficient training techniques for M-lab learning. Copyright © 2023, The Authors. All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Language-Guided Instance-Aware Domain-Adaptive Panoptic Segmentation

arXiv

引用

arXiv 2024年

作者： Mansour, Elham Amin Unal, Ozan Saha, Suman Bejar, Benjamin Van Gool, Luc Computer Vision Lab ETH Zurich Switzerland Swiss Data Science Center PSI Switzerland

The increasing relevance of panoptic segmentation is tied to the advancements in autonomous driving and AR/VR applications. However, the deployment of such models has been limited due to the expensive nature of dense data annotation, giving rise to unsupervised domain adaptation (UDA). A key challenge in panoptic UDA is reducing the domain gap between a labeled source and an unlabeled target domain while harmonizing the subtasks of semantic and instance segmentation to limit catastrophic interference. While considerable progress has been achieved, existing approaches mainly focus on the adaptation of semantic segmentation. In this work, we focus on incorporating instance-level adaptation via a novel instance-aware cross-domain mixing strategy IMix. IMix significantly enhances the panoptic quality by improving instance segmentation performance. Specifically, we propose inserting high-confidence predicted instances from the target domain onto source images, retaining the exhaustiveness of the resulting pseudo-labels while reducing the injected confirmation bias. Nevertheless, such an enhancement comes at the cost of degraded semantic performance, attributed to catastrophic forgetting. To mitigate this issue, we regularize our semantic branch by employing CLIP-based domain alignment (CDA), exploiting the domain-robustness of natural language prompts. Finally, we present an end-to-end model incorporating these two mechanisms called LIDAPS, achieving state-of-the-art results on all popular panoptic UDA benchmarks. © 2024, CC BY-NC-ND.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Open-Vocabulary Calibration for Fine-tuned CLIP

arXiv

引用

arXiv 2024年

作者： Wang, Shuoyuan Wang, Jindong Wang, Guoqing Zhang, Bob Zhou, Kaiyang Wei, Hongxin Department of Statistics and Data Science Southern University of Science and Technology Shenzhen China Department of Computer and Information Science University of Macau Taipa China William & Mary WilliamsburgVA United States School of Computer Science and Engineering University of Electronic Science and Technology of China China Department of Computer Science Hong Kong Baptist University Hong Kong

Vision-language models (VLMs) have emerged as formidable tools, showing their strong capability in handling various open-vocabulary tasks in image recognition, text-driven visual content generation, and visual chatbots, to name a few. In recent years, considerable efforts and resources have been devoted to adaptation methods for improving the downstream performance of VLMs, particularly on parameter-efficient fine-tuning methods like prompt learning. However, a crucial aspect that has been largely overlooked is the confidence calibration problem in fine-tuned VLMs, which could greatly reduce reliability when deploying such models in the real world. This paper bridges the gap by systematically investigating the confidence calibration problem in the context of prompt learning and reveals that existing calibration methods are insufficient to address the problem, especially in the open-vocabulary setting. To solve the problem, we present a simple and effective approach called Distance-Aware Calibration (DAC), which is based on scaling the temperature using as guidance the distance between predicted text labels and base classes. The experiments with 7 distinct prompt learning methods applied across 11 diverse downstream datasets demonstrate the effectiveness of DAC, which achieves high efficacy without sacrificing the inference speed. Our code is available at https://***/mlstat-Sustech/CLIP Calibration. © 2024, CC0.

关键词： Calibration

来源：评论

学校读者我要写书评

暂无评论

E2E-MFERC:AMulti-Face Expression Recognition Model for Group Emotion Assessment

引用

computers, Materials & Continua 2024年第4期79卷 1105-1135页

作者： Lin Wang Juan Zhao Hu Song Xiaolong Xu Jiangsu Key Laboratory of Big Data Security&Intelligent Processing Nanjing University of Posts and TelecommunicationsNanjing210042China School of Network Security Jinling Institute of TechnologyNanjing211169China State Grid Jiangsu Electric Power Company Limited Nanjing210000China School of Computer Science Nanjing University of Posts and TelecommunicationsNanjing210042China

In smart classrooms, conducting multi-face expression recognition based on existing hardware devices to assessstudents’ group emotions can provide educators with more comprehensive and intuitive classroom effect analysis,thereby continuouslypromotingthe improvementof teaching ***,most existingmulti-face expressionrecognition methods adopt a multi-stage approach, with an overall complex process, poor real-time performance,and insufficient generalization ability. In addition, the existing facial expression datasets are mostly single faceimages, which are of low quality and lack specificity, also restricting the development of this research. This paperaims to propose an end-to-end high-performance multi-face expression recognition algorithm model suitable forsmart classrooms, construct a high-quality multi-face expression dataset to support algorithm research, and applythe model to group emotion assessment to expand its application value. To this end, we propose an end-to-endmulti-face expression recognition algorithm model for smart classrooms (E2E-MFERC). In order to provide highqualityand highly targeted data support for model research, we constructed a multi-face expression dataset inreal classrooms (MFED), containing 2,385 images and a total of 18,712 expression labels, collected from smartclassrooms. In constructing E2E-MFERC, by introducing Re-parameterization visual geometry group (RepVGG)block and symmetric positive definite convolution (SPD-Conv) modules to enhance representational capability;combined with the cross stage partial network fusion module optimized by attention mechanism (C2f_Attention),it strengthens the ability to extract key information;adopts asymptotic feature pyramid network (AFPN) featurefusion tailored to classroomscenes and optimizes the head prediction output size;achieves high-performance endto-end multi-face expression detection. Finally, we apply the model to smart classroom group emotion assessmentand provide design refe

关键词： Multi-face expression recognition smart classroom end-to-end detection group emotion assessment

来源：评论

学校读者我要写书评

暂无评论

Prefetching-based Adaptive Video Streaming Strategy on Edge Computing

Prefetching-based Adaptive Video Streaming Strategy on Edge ...

引用

International Conference on Parallel and Distributed Systems (ICPADS)

作者： Xuanyu Yi Jipeng Zhou Department of Computer Science Jinan University Guangzhou China Department of Computer Science School of Data Science of Guangzhou Huashang College Jinan University Guangzhou China

Dynamic Adaptive Streaming over HTTP (DASH) is a widely adopted video streaming protocol. Adaptive Bitrate Streaming (ABR) algorithm is utilized to dynamically switch between different bitrates. However, traditional ABR algorithms have gradually failed to meet users’ demands for high-quality video transmission, especially in complex network environments. Thus, the current research focus has shifted towards enhancing the accuracy of algorithm. The advent of edge computing has brought new possibilities for DASH transmission. Edge computing-based algorithms can make decisions from a more macroscopic perspective, which can enhance algorithm efficiency. In this paper, we propose an edge computing-based strategy that enables the edge server to obtain the actual file size of the next segment at every bitrate. As a result, edge servers can obtain more informations. We further propose an edge-based system model that assists the ABR algorithm in achieving better operational efficiency. With additional information, the algorithm located at the edge node possesses the capability to make more precise decisions, which is advantageous for enhancing the quality of experience (QoE) for users. Our experiments demonstrate that the proposed strategy can significantly enhance QoE and network resource utilization.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A New Time Optimal Control based Tracking Differentiator with Small Phase Lag 43

A New Time Optimal Control based Tracking Differentiator wit...

引用

43rd Chinese Control Conference, CCC 2024

作者： Zhang, Louyue Zhang, Hehong Zhai, Chao Wang, Xi Dan, Zhihong Beihang University School of Energy and Power Engineering Beijing102206 China Fuzhou University College of Computer and Data Science Fuzhou350108 China China University of Geosciences School of Automation Wuhan430074 China AECC Sichuan Gas Turbine Establishment Mianyang Science and Technology on Altitude Simulation Laboratory 621703 China

ISBN: (纸本)9789887581581

Real-time filtering and derivative signals with as small phase lag as possible are of great significance for control performances. In this work, an enhanced time optimal control (referred as f sa) based tracking differentiator (TD) is proposed, which can guarantee a small phase lag in both filtering and differential signals for a class of input signals with different levels of noises. This characteristic is achieved by introducing a damping coefficient to the boundary transformation function of the time-optimal control algorithm. Simulation results show that the proposed f s a based TD outperforms the fhan based TD in both filtering and differentiation © 2024 Technical Committee on Control Theory, Chinese Association of Automation.

关键词： Damping

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：