检索结果-内蒙古大学图书馆

international conference on pattern recognition

作者： Mengyuan Ding Shanshan Zhang Jian Yang PCA Lab Key Lab of Intelligent Perception and Systems for High-Dimensional Information of Ministry of Education Jiangsu Key Lab of Image and Video Understanding for Social Security China

ISBN: (纸本)9781728188089;9781728188096

Pedestrian detection is a canonical instance of object detection in computer vision. In practice, scale variation is one of the key challenges, resulting in unbalanced performance across different scales. Recently, the High-Resolution Network (HRNet) has become popular because high-resolution feature representations are more friendly to small objects. However, when we apply HRNet to pedestrian detection, we observe that it improves for small pedestrians on one hand, but hurts the performance for larger ones on the other hand. To overcome this problem, we propose a learnable Dynamic HRNet (DHRNet) aiming to generate different network paths adaptive to different scales. Specifically, we construct a parallel multi-branch architecture and add a soft conditional gate module allowing for dynamic feature fusion. Both branches share all the same parameters except the soft gate module. Experimental results on CityPersons and Caltech benchmarks indicate that our proposed dynamic HRNet is more capable of dealing with pedestrians of various scales, and thus improves the performance across different scales consistently.

关键词： Computer vision Adaptive systems Object detection Computer architecture Logic gates Benchmark testing pattern recognition

来源：评论

学校读者我要写书评

暂无评论

A Privacy-Protecting Indoor Emergency Monitoring System based on Floor Vibration 20

A Privacy-Protecting Indoor Emergency Monitoring System base...

引用

ACM international Joint conference on Pervasive and Ubiquitous computing (UbiComp) / ACM international Symposium on Wearable Computers (ISWC)

作者： Yu, Yang Weis, Torben Univ Duisburg Essen Distributed Syst Grp Duisburg Germany

ISBN: (纸本)9781450380768

In this work we present an indoor emergency context monitoring system based on ground vibration caused by persons in the target area. The system is designed for production plants and large buildings to perceive the safety status of this area. Our approach is privacy-protecting, because it requires neither video nor sound. Instead, piezo sensors on the floor measure vibrations, which are analyzed with machine learning to compute the safety status of the covered area. This way our system can determine whether an emergency occurred, but it is not straight forward possible to attach names to the detected persons. We compare the impact of different feature extraction methods and different types of classifiers on the classification results. Our experiments show that we can determine an emergency event with an average F1 score of 0.97.

关键词： vibration signal pattern recognition deep learning privacy protection emergency detecton ontext recognition piezo sensor

来源：评论

学校读者我要写书评

暂无评论

RSAC: Regularized Subspace Approximation Classifier for Lightweight Continuous Learning

RSAC: Regularized Subspace Approximation Classifier for Ligh...

引用

international conference on pattern recognition

作者： Chih-Hsing Ho Shang-Ho Lawrence Tsai National Chiao Tung University Hsinchu Taiwan

ISBN: (纸本)9781728188089;9781728188096

Continuous learning seeks to perform the learning on the data that arrives from time to time. While prior works have demonstrated several possible solutions, these approaches require excessive training time as well as memory usage. This is impractical for applications where time and storage are constrained, such as edge computing. In this work, a novel training algorithm, regularized subspace approximation classifier (RSAC), is proposed to achieve lightweight continuous learning. RSAC contains a feature reduction module and classifier module with regularization. Extensive experiments show that RSAC is more efficient than prior continuous learning works and outperforms these works on various experimental settings.

关键词： Training Memory management Training data Approximation algorithms Classification algorithms pattern recognition Usability

来源：评论

学校读者我要写书评

暂无评论

Handwritten Mathematical Expression recognition with Self-Attention 21

Handwritten Mathematical Expression Recognition with Self-At...

引用

Proceedings of the 2021 4th international conference on Algorithms, computing and Artificial Intelligence

作者： Xueke Chi Da-Han Wang Yuefeng Wu Yun Wu Fujian Key Laboratory of Pattern Recognition and Image Understanding China and School of Computer and Information Engineering Xiamen University of Technology China

ISBN: (纸本)9781450385053

Attention-based encoder-decoder models have made great success on handwritten mathematical expression recognition in recent years. However, this kind of method has the problem of attention drift, because under the local attention mechanism based on RNN, the high similarity between coding features can cause attention confusion. To settle this problem, we propose an encoder-decoder model with self-attention, which captures the global information of the feature map and fuses the local information of the CNN as complementary features. Experiments are conducted on the CROHME2014 and CROHME 2016 competition datasets. The experimental results show that, when only using the official training dataset, the proposed method achieves recognition accuracies of 51.98% and 50.74% on the CROHME2014 and CROHME2016 competition datasets, respectively, which outperforms the other methods significantly. The improvements demonstrate the effectiveness of the self-attention module.

关键词： handwritten mathematical expression offline recognition non-local self-attention

来源：评论

学校读者我要写书评

暂无评论

Russian Advances in Fuzzy Systems and soft computing: Selected Contributions to the 8th international conference on "Fuzzy Systems, soft soft computing and Intelligent Technologies",FSSCIT 2020

Russian Advances in Fuzzy Systems and Soft Computing: Select...

引用

Russian Advances in Fuzzy Systems and soft computing: Selected Contributions to the 8th international conference on "Fuzzy Systems, soft soft computing and Intelligent Technologies",FSSCIT 2020

The proceedings contain 39 papers. The topics discussed include: (dis)similarity-based correlation functions on polar scales;logic of estimates for fuzzy logics sentences;architecture of some models for optimization problems under conditions of hybrid uncertainty;inference method for MISO-structure systems with fuzzy inputs using parallel computing techniques;method for classification of objects with fuzzy values of features;analysis of the multifactorial phenomena based on fuzzy Bayesian model;fuzzy clustering in the problem of technological state assessment for phosporose production;algebraic Bayesian networks: a frequentist approach to knowledge pattern parameters machine learning;empirical approach to assessing sensitivity of local posteriori inference of algebraic Bayesian network;and new trends in measurement theory: Bayesian intelligent measurement and its application in the digital economy.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Methods for Analysis of EMG Signals: Systematic Review

Methods for Analysis of EMG Signals: Systematic Review

引用

Mediterranean conference on Embedded computing (MECO)

作者： Medina Herak Tarik Zubč ević Faculty of Engineering and Natural Sciences International Burch University Sarajevo Bosnia and Herzegovina

ISBN: (数字)9781665468282

ISBN: (纸本)9781665468299

The aim of this paper was to get familiar with the basic methods used for the EMG signal analysis. The research followed PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines for a clearer and concise search strategy. Grading of Recommendations of Assessment, Development and Evaluation strategy was used to ensure evidence certainty. Data that was summarized and presented in the review are gathered from three different electronic databases. Doing this systematic review, we summarized methods such as Fourier transform, wavelet analysis, the use of artificial networks, etc. We discussed the main advantages and disadvantages when it comes to each of the methods, but also a comparison between some of these methods. It will aid researchers in choosing the most appropriate method of EMG signal analysis which is important when it comes to neurological diagnostics, biomedical research, and prosthetic arm control.

关键词： Wavelet transforms Fourier transforms Systematics Wavelet analysis Search problems Electromyography pattern recognition

来源：评论

学校读者我要写书评

暂无评论

JECL: Joint Embedding and Cluster Learning for Image-Text Pairs

JECL: Joint Embedding and Cluster Learning for Image-Text Pa...

引用

international conference on pattern recognition

作者： Sean T. Yang Kuan-Hao Huang Bill Howe University of Washington University of California Los Angeles

We propose JECL, a method for clustering image-caption pairs by training parallel encoders with regularized clustering and alignment objectives, simultaneously learning both representations and cluster assignments. These image-caption pairs arise frequently in high-value applications where structured training data is expensive to produce, but free-text descriptions are common. JECL trains by minimizing the Kullback-Leibler divergence between the distribution of the images and text to that of a combined joint target distribution and optimizing the Jensen-Shannon divergence between the soft cluster assignments of the images and text. Regularizers are also applied to JECL to prevent trivial solutions. Experiments show that JECL outperforms both single-view and multi-view methods on large benchmark image-caption datasets, and is remarkably robust to missing captions and varying data sizes.

关键词： Training Sensitivity Computational modeling Training data Benchmark testing Robustness pattern recognition

来源：评论

学校读者我要写书评

暂无评论

APT: An Automated Probe Tracker From Gene Expression Data

引用

IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS 2021年第5期18卷 1864-1874页

作者： Kumar, Gautam Kumar, Rajnish Pal, Manoj Kumar Pramanik, Nilotpal Lahiri, Tapobrata Gupta, Ankita Pandey, Saket Indian Inst Informat Technol Allahabad Dept Appl Sci Comp Ctr 2 Room 4302 Allahabad 211015 Uttar Pradesh India

Out of currently available semi-automatic tools for detecting diagnostic probes relevant to a pathophysiological condition, ArrayMining and GEO2R of NCBI are most popular. The shortcomings of ArrayMining and GEO2R are that both tools list the probes ordering them on the basis of their individual statistical level of significances with only difference of statistical methods used by them. While the latest tool GEO2R outputs either top 250 or all genes following its own ranking mechanism, ArrayMining requires number of probes to be inputted by the user. This study provided a way for automatic selection of probe-set that can be obtained from the voting of outputs resulted from statistical methods, t-Test, Mann-Whitney Test and Empirical Bayes Moderated t-test. It was also intriguing to find that the parameters of these statistical methods can be represented as a mathematical function of group fisher's discriminant ratio of a disease-control expression data-pair. Result of this fully automatic method, APT shows 88.97 percent success in comparison to 80.40 and 87.60 percent successes of ArrayMining and GEO2R respectively to include reported probes. Furthermore, out of 10 fold cross validation and 5 new test cases, APT shows a better performance than both ArrayMining and GEO2R in regards to sensitivity and specificity.

关键词： Probes Bioinformatics Genomics Gene expression Tools Diseases Arrays pattern recognition automatic probe selection differential gene expression t-test Mann-Whitney test empirical Bayes moderated t-test Fisher's discriminant ratio non-linear polynomial regression curve fitting

来源：评论

学校读者我要写书评

暂无评论

Palmprint based pattern recognition Using Fast ICA

Palmprint based Pattern Recognition Using Fast ICA

引用

international conference on Intelligent computing and Control Systems (ICICCS)

作者： Patil, Jayshri P. Pawar, Chandrashekhar S. Gangamai Polytech Dept Comp Engn Dhule India CHARUSAT Dept Comp Sci & Engn Devang Patel Inst Adv Technol & Res Changa Gujarat India

ISBN: (纸本)9781728148762

The pattern of Palmprint has high dissimilarity because of its numerous determination features like ridges, principle lines, datum points, and wrinkles and so on. In this paper, a blend of Haar Wavelet, DCT and Fast ICA have been utilized to recognize the palmprint pattern using fast ICA. The images of Palm were initially decomposed by the Haar Wavelet and frequency band, independent of different image resolutions chosen for dimensionality decrease. Next on LL, band DCT was use. Lastly, Fast ICA use to discover the independent components of the DCT features. To find matching image, Euclidean Distance (ED) has been used. The procedure is validated on Polytechnic-University (Poly U) storage as well as the results show the efficiency and accuracy of the proposed method.

关键词： Independent Component Analysis (ICA) Palmprint Discrete Cosine Transform (DCT) Haar Wavelet

来源：评论

学校读者我要写书评

暂无评论

3rd international conference on Smart Innovations in Communications and Computational Sciences, ICSICCS 2020

3rd International Conference on Smart Innovations in Communi...

引用

3rd international conference on Smart Innovations in Communications and Computational Sciences, ICSICCS 2020

ISBN: (纸本)9789811553448

The proceedings contain 29 papers. The special focus in this conference is on Smart Innovations in Communications and Computational Sciences. The topics include: Multipath routing using improved grey wolf optimizer (igwo)-based ad hoc on-demand distance vector routing (aodv) algorithm on manet;analysis and prediction of fuel oil in a terminal;an android malware detection method based on native libraries;fire early warning system based on precision positioning technology;the construction of learning resource recommendation system based on recognition technology;review of ultra-low-power cmos amplifier for bio-electronic sensor interface;an enhancement of underwater images based on contrast restricted adaptive histogram equalization for image enhancement;hybridization of social spider optimization (sso) algorithm with differential evolution (de) using super-resolution reconstruction of video images;enhanced convolutional neural network (ecnn) for maize leaf diseases identification;facial expression recognition using improved local binary pattern and min-max similarity with nearest neighbor algorithm;visibility improvement of hazy image using fusion of multiple exposure images;comparative analysis of consensus algorithms and issues in integration of blockchain with iot;use of hadoop for sentiment analysis on twitter’s big data;enhanced mobility management model for mobile communications;multi-objective-based lion optimization algorithm for relay selection in wireless cooperative communication networks;embpred30: Assessing 30-days readmission for diabetic patients using categorical embeddings;prediction of different classes of skin disease using machine learning techniques;feature selection and classification based on swarm intelligence approach to detect malware in android platform;preface;trust-based fuzzy bat optimization algorithm for attack detection in manet;transfer learning: Survey and classification.

关键词：

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：