检索结果-内蒙古大学图书馆

HuayuNavi: A mobile chinese learning application based on intelligent character recognition

Lecture Notes in computer science

作者： Kuo, Jen-Ho Huang, Cheng-Ming Liao, Wen-Hung Huang, Chun-Chieh Dept. of Computer Science National Chengchi University Taipei Taiwan Master's Program in Digital Contents and Technologies National Chengchi University Taipei Taiwan Funwish Co. Taipei Taiwan

ISBN: (纸本)9783642234552

With the recent rise of China, Chinese is becoming a dominant language in the world. People are pursuing an efficient and effective means to learn the Chinese language. Most of the traditional learning platforms such as textbooks, laptop applications and language learning centers are not portable and interactive simultaneously. In this paper, we attempt to develop a new language-learning platform that not only creates a better user experience but also promotes better efficiency in language training. We present a mobile application named HuayuNavi that integrates a touch-based user interface with intelligent character recognition techniques to achieve real-time image content understanding. Specifically, the feature vector is formed by accumulating localized gradients of different orientations in the image. Recognition is achieved by employing support vector machine (SVM) with probability estimation to obtain candidate characters, which is then refined using domain-specific vocabulary models. The overall operation can be completed in 3 seconds. Initial tests on the specific subject of Taiwanese snacks indicate that the recognition rate can reach 83% using handwritten samples as well as signboards containing characters of diverse fonts. © 2011 Springer-Verlag.

关键词： User interfaces

来源：评论

学校读者我要写书评

暂无评论

Design patterns for lazy evaluation

SIGCSE Bulletin (Association for Computing Machinery, Specia...

引用

SIGCSE Bulletin (Association for Computing Machinery, Special Interest Group on computer science Education) 2000年 21-25页

作者： Nguyen, Dung Wong, Stephen B. Dept. of Computer Science Rice University Houston TX United States Computer Science Program Oberlin College Oberlin OH United States

We propose an object-oriented (OO) formulation and implementation of lazy/delayed evaluation by reusing and extending an existing linear recursive structure (LRS) framework with the help of the strategy, decorator and factory design patterns. The result is a robust, flexible framework that can handle both infinite and finite lists and to which existing algorithms for finite lists can be applied without modification. The OO techniques used to develop this model are effective tools for teaching abstraction and design of data structures.

关键词： computer science

来源：评论

学校读者我要写书评

暂无评论

Design patterns for lazy evaluation 00

Design patterns for lazy evaluation

引用

Proceedings of the thirty-first SIGCSE technical symposium on computer science education

作者： Dung Nguyen Stephen B. Wong Dept. of Computer Science Rice University Houston TX Computer Science Program Oberlin College Oberlin OH

ISBN: (纸本)9781581132137

We propose an object-oriented (OO) formulation and implementation of lazy/delayed evaluation by reusing and extending an existing linear recursive structure (LRS) framework with the help of the strategy, decorator and factory design patters. The result is a robust, flexible framework that can handle both infinite and finite lists and to which existing algorithms for finite lists can be applied without modification. The OO techniques used to develop this model are effective tools for teaching abstraction and design of data structures.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Robust Map Fusion with Visual Attention Utilizing Multi-agent Rendezvous

Robust Map Fusion with Visual Attention Utilizing Multi-agen...

引用

IEEE International Conference on Robotics and Automation (ICRA)

作者： Jaein Kim Dong-Sig Han Byoung-Tak Zhang Interdisciplinary Program in Neuroscience Seoul National University Artificial Intelligence Institute Seoul National University Dept. of Computer Science and Engineering Seoul National University

The map fusion for multi-robot simultaneous localization and mapping (SLAM) consistently combines robot maps built independently into the global map. An established approach to map fusion is utilizing rendezvous, which refers to an encounter between multiple agents, to calculate the transformation into the global map. However, previous works using rendezvous have a limitation in that they are unreliable for certain circumstances, where the amount of agent observations or overlapping landmarks is limited. This work proposes a novel map fusion system which robustly fuses local maps in challenging rendezvous that lack shared information. Our system utilizes the single visual perception from rendezvous and estimates the relative pose between agents with the DOPE. Then our scheme transforms local maps with an estimated relative pose and predicts the misalignment from approximated maps by utilizing the attention mechanism of the vision transformer. Comparisons with the Hough transform-based method show that ours is significantly better when the overlap between local maps is insufficient. We also verify the robustness of our system against a similar real-world scenario.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Optimum sensor position for inverse load recovery on a composite plate

Optimum sensor position for inverse load recovery on a compo...

引用

1st International Conference on Advances and Trends in Engineering Materials and their Applications, AES-ATEMA'2007

作者： Coates, Cameron W. Hashemi, Ray R. Daniell, Darryl W. Engineering Studies Program Armstrong Atlantic State University Savannah GA 31419 United States Dept. of Computer Science Armstrong Atlantic State University Savannah GA 31419 United States

ISBN: (纸本)0978047907

An inverse method using strain values obtained at finite locations is used to predict coefficients of an unknown continuous static load function applied to a composite sandwich plate. The strain values obtained from a Finite Element Model (FEM) are used to simulate strain sensor readings. A Back-Propagation Neural Network (BPNN) is developed in order to predict optimum strain sensor locations necessary for accurate load function coefficient recovery using the inverse method. Coefficient values output from the inverse method, as well as material, and geometrical properties are used as input into the BPNN. The BPNN utilizes a back propagation algorithm that incorporates a linear transfer function. The BPNN developed predicts locations for sensor placement such that load function coefficients are recovered by the inverse method with good accuracy. © AES-Advanced Engineering Solutions.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

Prediction Tool on Fine Particle Pollutants and Air Quality for Environmental Engineering

引用

SN computer science 2022年第3期3卷 184页

作者： Varde, Aparna S. Pandey, Abidha Du, Xu Dept. of Computer Science Environmental Science and Management PhD Program Montclair State University Montclair NJ United States Visiting Researcher at Max Planck Institute for Informatics Saarbrücken Germany Department of Computer Science Montclair State University Montclair NJ United States Department of Earth and Environmental Science Environmental Science and Management PhD Montclair State University Montclair NJ United States

This article focuses on the research, design and implementation of a prediction tool for air quality to estimate pollutant concentrations, contributing to environmental engineering. It addresses prediction of fine particle air pollutants of diameter less than 2.5 µm (particulate matter 2.5), their concentration being substantially influenced by urban traffic. We collect worldwide multicity data from health-related public sources on which mining is performed using classical data mining/machine learning paradigms: association rules, clustering and classification. Challenges include adapting appropriate techniques based on data, and capturing subtle domain-specific aspects. The prediction tool is built using knowledge discovered by mining, leveraging health standards, catering to novice, intermediate and expert users. The prediction output is accurate, efficient, interpretable and useful as evident from our experiments. The tool is helpful for urban decision support. This work is beneficial in developing software systems such as intelligent tutors, mobile device apps and smart city tools. It contributes to smart environment, mobility and living, making a positive impact on smart cities and sustainability. In this work, we claim that classical computational paradigms in their fundamental form can be adapted to solve environmental engineering problems, with easy comprehension, as per the Occam’s razor principle that advocates simplicity. This article constitutes applied research: using computational techniques to solve domain-specific problems. Future work includes exploring models in deep learning such as CNN and Bi-LSTM, and considering different types of pollutants as well as other sources besides multicity traffic data, to conduct further studies. This would address additional challenges with enhancements. © 2022, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.

关键词： Applied research Decision support Environmental engineering Health informatics Machine learning Urban studies

来源：评论

学校读者我要写书评

暂无评论

Efficient clustering with limited distance information

Efficient clustering with limited distance information

引用

作者： Voevodski, Konstantin Balcan, Maria-Florina Röglin, Heiko Teng, Shang-Hua Xia, Yu Dept. of Computer Science Boston University Boston MA 02215 United States College of Computing Georgia Institute of Technology Atlanta GA 30332 United States Dept. of Quantitative Economics Maastricht University Maastricht Netherlands Computer Science Dept. University of Southern California Los Angeles CA 90089 United States Bioinformatics Program Dept. of Chemistry Boston University Boston MA 02215 United States

ISBN: (纸本)9780974903965

Given a point set S and an unknown metric d on S, we study the problem of efficiently partitioning S into k clusters while querying few distances between the points. In our model we assume that we have access to one versus all queries that given a point s Ε S return the distances between s and all other points. We show that given a natural assumption about the structure of the instance, we can efficiently find an accurate clustering using only O(k) distance queries. We use our algorithm to cluster proteins by sequence similarity. This setting nicely fits our model because we can use a fast sequence database search program to query a sequence against an entire dataset. We conduct an empirical study that shows that even though we query a small fraction of the distances between the points, we produce clusterings that are close to a desired clustering given by manual classification.

关键词： Search engines

来源：评论

学校读者我要写书评

暂无评论

A Comparative Study of Meningioma Tumors Segmentation Methods from MR Images 3

A Comparative Study of Meningioma Tumors Segmentation Method...

引用

3rd International Conference on Signal Processing and Information Security, ICSPIS 2020

作者： Alkhodari, Mohanad Hassanin, Omnia Dhou, Salam Biomed. Eng. Graduate Program American University of Sharjah Sharjah United Arab Emirates Dept. of Computer Science and Engineering American University of Sharjah Sharjah United Arab Emirates

ISBN: (纸本)9781728189987

Brain tumor segmentation from magnetic resonance (MR) images can have a great impact on improving diagnostics, growth rate prediction, and treatment planning. In this paper, we provide a comparative study of four well-known segmentation algorithms, namely k-means clustering, histogram thresholding (Otsu), fuzzy c-means thresholding, and region growing. For the region growing algorithm, the seed selection process is automated and enhanced by preprocessing the images and approximating the tumor regions using initial clustering and/or thresholding approaches. The evaluation and comparison of the algorithms is conducted using a data-set of T1-Weighted Contrast-Enhanced magnetic resonance imaging (MRI) brain images. Ground truth tumor images were provided by three experienced radiologists and are used in the evaluation process. Results showed that the enhanced region growing method had the highest mean dice similarity coefficient with a score of 0.87, and the lowest under-segmentation rate (17.46%). The fuzzy c-means thresholding method had the lowest over-segmentation rate (0.03%). This study serves as a baseline for other advanced tumor segmentation studies such as the ones using the emergent machine learning approaches. © 2020 IEEE.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Unsupervised learning by a nonlinear network with Hebbian excitatory and anti-Hebbian inhibitory neurons

arXiv

引用

arXiv 2018年

作者： Sebastian, H. Seung Neuroscience Institute and Computer Science Dept. Princeton University PrincetonNJ08544 United States

This paper introduces a rate-based nonlinear neural network in which excitatory (E) neurons receive feedforward excitation from sensory (S) neurons, and inhibit each other through disynaptic pathways mediated by inhibitory (I) interneurons. Correlation-based plasticity of disynaptic inhibition serves to incompletely decorrelate E neuron activity, pushing the E neurons to learn distinct sensory features. The plasticity equations additionally contain "extra" terms fostering competition between excitatory synapses converging onto the same postsynaptic neuron and inhibitory synapses diverging from the same presynaptic neuron. The parameters of competition between S → E connections can be adjusted to make learned features look more like "parts" or "wholes." The parameters of competition between I − E connections can be adjusted to set the typical decorrelatedness and sparsity of E neuron activity. Numerical simulations of unsupervised learning show that relatively few I neurons can be sufficient for achieving good decorrelation, and increasing the number of I neurons makes decorrelation more complete. Excitatory and inhibitory inputs to active E neurons are approximately balanced as a result of learning. Copyright © 2018, The Authors. All rights reserved.

关键词： Neurons

来源：评论

学校读者我要写书评

暂无评论

An interactive environment for data partitioning and distribution 5

An interactive environment for data partitioning and distrib...

引用

5th Distributed Memory Computing Conference, DMCC 1990

作者： Balasundaram, Vasanth Fox, Geoffrey Kennedy, Ken Kremer, Ulrich Caltech Concurrent Computation Program Mail Stop 206-49 Caltech PasadenaCA91125 United States Dept. of Computer Science Rice University HoustonTX77251 United States

ISBN: (纸本)0818621133

An approach to distributed memory parallel programming that has recently become popular is one where the programmer explicitly specifies the data decomposition using language extensions, and a, compiler generates all the communication. While this frees the programmer from the tedium of thinking about message-passing, no assistance is provided in determining the data decomposition scheme that gives the best performance on the target machine. In this paper, we propose an interactive software tool that provides assistance for this very task. The proposed tool also computes performance estimates for any chosen data partitioning scheme, allowing the programmer to experiment with several different strategies without ever running the program on the machine. ©1990 IEEE.

关键词： Parallel programming

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：