检索结果-内蒙古大学图书馆

16th IAPR International Conference on Document Analysis and Recognition (ICDAR)

作者： Wang, Jiaming Wang, Qing Du, Jun Zhang, Jianshu Wang, Bin Ren, Bo Univ Sci & Technol China Hefei Anhui Peoples R China Guangdong Artificial Intelligence & Digital Econ Pazhou Lab Guangzhou Peoples R China Tencent Youtu Lab Hefei Anhui Peoples R China

ISBN: (纸本)9783030863340;9783030863333

Recently, attention based encoder-decoder methods have been widely used in online handwritten mathematical expression recognition, which achieve significant improvements compared to traditional methods. The encoder-decoder methods usually employ string decoders to generate the recognition result, which are not well matched for tree-structured languages like math expression. A novel sequential relation decoder (SRD) was introduced to recognize the online mathematical expression as a math tree, which can be decomposed into a subtree sequence and each subtree consists of a relation node and two symbol nodes (related symbol node and primary symbol node). However, the alignments between these two symbol nodes were implemented by spatial attention probabilities, leading to incorrect recognition if spatial attention is not accurate. In this paper, we propose a memory relation decoder (MRD), equipped with a memory based attention model to determine the correspondence between two symbol nodes. Specifically, at each decoding step, this memory based attention finds the corresponding primary symbol node in the memory and treats it as the related symbol node, which actually achieves the alignments between two symbol nodes in an explicit manner. Besides, we propose to introduce global visual information while calculating attention probabilities to help alleviate the ambiguous problems in online handwritten mathematical expression recognition. Evaluated on a benchmark published by CROHME competition, the proposed approach can substantially outperform previous encoder-decoder methods.

关键词： Handwritten mathematical expression recognition encoder-decoder Tree structure Memory based attention

来源：评论

学校读者我要写书评

暂无评论

Synthesizing DNAs of a system's binary files from its functional and structural representation

Synthesizing DNAs of a system's binary files from its functi...

引用

73rd IEEE National Aerospace and Electronics Conference (NAECON)

作者： Ngwobia, Sunday Cosmos Ralescu, Anca Kapp, David Kebede, Temesgen Univ Cincinnati EECS Dept Cincinnati OH 45221 USA AFRL Dayton OH USA

ISBN: (纸本)9781665448598

Malware infects system programs or files basically by injecting malicious code into them via a dynamic or static approach. This malicious code will cause the system to malfunction or crash while compromising invaluable vital data. Engineering techniques to curb malicious attacks has been an area of research interest in cybersecurity and malware analysis. After careful examination of the existing work in this domain, we conclude that most of the state-of-the-art research work concentrate on the separate analysis of the behavior of benign and malware programs in order to obtain discriminating feature sets between them. We propose a novel approach to study the behaviors of non-malicious programs (.exe, .dll, etc.) when infected and not infected by malware and then compare those with the behavior of the malware itself. The paper aims to capture these feature sets associated with these three distinct behavioral dynamics, analyze and synthesize them into DNA-like representation (data sets) akin to biological Deoxyribonucleic acid(DNA). These data sets will then be piped into another future phase of this research that aims at developing a malware mitigation model that leverages the CRISPR defense concept, detects zero-day malware and enables recovering (repairing) files infected by malware.

关键词： Profiling DNA sequence encoder-decoder graph similarly API calls

来源：评论

学校读者我要写书评

暂无评论

Simple Online Unmanned Aerial Vehicle Tracking with Transformer 20

Simple Online Unmanned Aerial Vehicle Tracking with Transfor...

引用

20th IEEE International Conference on Trust, Security and Privacy in Computing and Communications (IEEE TrustCom)

作者： Liu, Yang Wang, Ershen Xu, Song Wang, Zhi Liu, Meizhi Shu, Wansen Zhejiang Key Lab Gen Aviat Operat Technol Jiande Peoples R China Civil Aviat Management Inst China Dept Gen Aviat CAAC Key Lab Gen Aviat Operat Beijing Peoples R China Shenyang Aerosp Univ Sch Elect & Informat Engn Shenyang Peoples R China Shenyang Aerosp Univ Sch Artificial Intelligence Shenyang Peoples R China

ISBN: (纸本)9781665416580

Multi-Object Tracking (MOT) mainly implements certain complex multi-step tracking by using detection algorithms, which respectively perform target detection, feature extraction and data association. The main challenges of Unmanned Aerial Vehicle (UAV) tracking are complex background and mutual occlusion. In this work, we propose a simple online UAV tracking with Transformer. It uses the encoder-decoder mechanism to introduce a set of object query in the pipeline, which can achieve the detection of new targets accurately. This method uses an online joint-detection-and-tracking pipeline based on the encoder-decoder mechanism. The complex and multi-step components in the previous method are simplified. Further, it is a new structure based on Transformer. Object query detects the object in the current frame. The object feature query in the previous frame associates these current objects with the previous object. We proposed a method that can be well applied to small targets such as UAVs, achieved 64.1 % MOTA competition on the MOTIS challenge dataset.

关键词： Unmanned Aerial Vehicle (UAV) transformer encoder-decoder multiple object tracking

来源：评论

学校读者我要写书评

暂无评论

2D Semantic Segmentation of the Prostate Gland in Magnetic Resonance Images using Convolutional Neural Networks 11th

2D Semantic Segmentation of the Prostate Gland in Magnetic R...

引用

11th IFAC Symposium on Biological and Medical Systems (BMS)

作者： Vacacela, Silvia P. Benalcazar, Marco E. Escuela Politec Nacl Artificial Intelligence & Comp Vis Res Lab Dept Comp Sci & Informat Quito Ecuador

Convolutional Neural Networks is one of the most commonly used methods for automatic prostate segmentation. However, few studies focus on the segmentation of the two main zones of the prostate: the central gland and the peripheral zone. This work proposes and evaluates two models for 2D semantic segmentation of these two zones of the prostate. The first model (Model-A) uses an encoder-decoder architecture based on the global U-net and the local U-net architectures. The global U-net segments the whole prostate, whereas the local U-net segments the central gland. The peripheral zone is obtained by subtracting the central gland from the whole prostate. On the other hand, the second model (Model-B) uses an encoder-classifier architecture based on the VGG16 network. Model-B performs segmentation by classifying each pixel of a Magnetic Resonance Image (MRI) into three categories: background, central gland, and peripheral zone. Both models are tested using MRIs from the dataset NCI-ISBI 2013 Challenge. The experimental results show a superior segmentation performance for Model-A, encoder-decoder architecture, (DSC = 96.79% +/- 0.15% and IoU = 93.79% +/- 0.29%) compared to Model-B, encoder-classifier architecture, (DSC = 92.50% +/- 1.19% and IoU = 86.13% +/- 2.02%). Copyright (C) 2021 The Authors.

关键词： Convolutional Neural Networks Prostate Segmentation Central Gland Peripheral Zone MRIs encoder-decoder U-net encoder-Classifier VGG16 NCI-ISBI 2013

来源：评论

学校读者我要写书评

暂无评论

TWO-STAGE PRE-TRAINING FOR SEQUENCE TO SEQUENCE SPEECH RECOGNITION

TWO-STAGE PRE-TRAINING FOR SEQUENCE TO SEQUENCE SPEECH RECOG...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Fan, Zhiyun Zhou, Shiyu Xu, Bo Chinese Acad Sci Inst Automat Beijing Peoples R China Univ Chinese Acad Sci Sch Artificial Intelligence Beijing Peoples R China

ISBN: (纸本)9780738133669

The attention-based encoder-decoder structure is popular in automatic speech recognition (ASR). However, it relies heavily on transcribed data. In this paper, we propose a novel pre-training strategy for the encoder-decoder sequence-to-sequence (seq2seq) model by utilizing unpaired speech and transcripts. The pre-training process consists of two stages, acoustic pre-training and linguistic pre-training. In the acoustic pre-training stage, we use a large amount of speech to pre-train the encoder by predicting masked speech feature chunks with their contexts. In the linguistic pre-training stage, we first generate synthesized speech from a large number of transcripts using a text-to-speech (TTS) system and then use the synthesized paired data to pretrain the decoder. The two-stage pre-training is conducted on the AISHELL-2 dataset, and we apply this pre-trained model to multiple subsets of AISHELL-1 and HKUST for post-training. As the size of the subset increases, we obtain relative character error rate reduction (CERR) from 38.24% to 7.88% on AISHELL-1 and from 12.00% to 1.20% on HKUST.

关键词： pre-training speech recognition encoder-decoder sequence-to-sequence

来源：评论

学校读者我要写书评

暂无评论

A DEEP NEURAL NETWORK MODEL FOR LEARNING RUNTIME FREQUENCY RESPONSE FUNCTION USING SENSOR MEASUREMENTS 16

A DEEP NEURAL NETWORK MODEL FOR LEARNING RUNTIME FREQUENCY R...

引用

16th ASME International Manufacturing Science and Engineering Conference (MSEC)

作者： Qu, Yongzhi Vogl, Gregory W. Wang, Zechao Univ Minnesota Duluth MN 55812 USA NIST Gaithersburg MD 20899 USA Wuhan Univ Technol Wuhan Peoples R China

ISBN: (纸本)9780791885079

The frequency response function (FRF), defined as the ratio between the Fourier transform of the time-domain output and the Fourier transform of the time-domain input, is a common tool to analyze the relationships between inputs and outputs of a mechanical system. Learning the FRF for mechanical systems can facilitate system identification, condition-based health monitoring, and improve performance metrics, by providing an input-output model that describes the system dynamics. Existing FRF identification assumes there is a one-to-one mapping between each input frequency component and output frequency component. However, during dynamic operations, the FRF can present complex dependencies with frequency cross-correlations due to modulation effects, nonlinearities, and mechanical noise. Furthermore, existing FRFs assume linearity between inputoutput spectrums with varying mechanical loads, while in practice FRFs can depend on the operating conditions and show high nonlinearities. Outputs of existing neural networks are typically low-dimensional labels rather than real-time high-dimensional measurements. This paper proposes a vector regression method based on deep neural networks for the learning of runtime FRFs from measurement data under different operating conditions. More specifically, a neural network based on an encoder-decoder with a symmetric compression structure is proposed. The deep encoder-decoder network features simultaneous learning of the regression relationship between input and output embeddings, as well as a discriminative model for output spectrum classification under different operating conditions. The learning model is validated using experimental data from a high-pressure hydraulic test rig. The results show that the proposed model can learn the FRF between sensor measurements under different operating conditions with high accuracy and denoising capability. The learned FRF model provides an estimation for sensor measurements when a physical sensor

关键词： Frequency response function encoder-decoder Neural network Deep learning

来源：评论

学校读者我要写书评

暂无评论

Edge Intelligence with Deep Learning in Greenhouse Management 10

Edge Intelligence with Deep Learning in Greenhouse Managemen...

引用

10th International Conference on Smart Cities and Green ICT Systems (SMARTGREENS)

作者： Proietti, Massimiliano Bianchi, Federico Marini, Andrea Menculini, Lorenzo Termite, Loris Francesco Garinei, Alberto Biondi, Lorenzo Marconi, Marcello Idea Re Srl Perugia Italy Guglielmo Marconi Univ Dept Sustainabil Engn Rome Italy K Digitale Srl Perugia Italy

ISBN: (纸本)9789897585128

This paper presents a methodology to control greenhouse operations based on deep learning. The proposed methodology employs Artificial Intelligence algorithms working on edge devices, allowing the detection of anomalies in plants growth and greenhouse control equipment, in view of taking possible corrective actions. Edge Intelligence allows the greenhouse to work independently of the network to which it is connected. It also guarantees privacy to the processed data and contributes to fast and efficient decision-making. In this work, a Long-Short Time Memory encoder-decoder architecture is used for greenhouse anomaly detection. The best performance is achieved when using one LSTM layer and 64 LSTM units.

关键词： Greenhouse Farming Deep Learning Computer Vision Edge Intelligence Anomaly Detection encoder-decoder Smart Local Systems

来源：评论

学校读者我要写书评

暂无评论

基于SBAS-InSAR与神经网络-数据同化机场跑道沉降预测研究

基于SBAS-InSAR与神经网络-数据同化机场跑道沉降预测研究

引用

作者：王志鹏中国民用航空飞行学院

学位级别：硕士

机场跑道沉降将增加飞机起降风险,使飞机在该过程中遭受震动和冲击,进而导致飞机机身受损,甚至是机毁人亡。尽管既有激光测量技术、GPS监测技术、地面震动测试技术、分布式光纤技术等跑道沉降监测预警技术具有高精度优势,但不是存在成... 详细信息

机场跑道沉降将增加飞机起降风险,使飞机在该过程中遭受震动和冲击,进而导致飞机机身受损,甚至是机毁人亡。尽管既有激光测量技术、GPS监测技术、地面震动测试技术、分布式光纤技术等跑道沉降监测预警技术具有高精度优势,但不是存在成本高就是存在干扰跑道正常运行等问题,因而如何建立低成本、非干扰式的跑道沉降监测预警模型,是当前研究的一个热点。针对该问题,本文提出了一种基于SBAS-In SAR和神经网络-数据同化的机场跑道沉降预测模型。(1)通过SBAS-In SAR技术克服现有沉降监测方法检测范围受监测点影响的缺点,充分利用In SAR技术的监测全面性,为沉降预测提供准确的基础数据。(2)通过三次样条插值方法对SBAS-In SAR获取的沉降时间序列数据进行插值处理,符合实际沉降趋势的同时扩充数据数量,为沉降预测模型提供多量数据。(3)构建LSTM和BP人工神经网络数据驱动模型,实现自主学习数据特征,提高预测模型的泛化能力。(4)引入En KF数据同化方法,构建En KF-LSTM和En KF-BP人工神经网络数据驱动模型。通过En KF代替LSTM和BP模型中使用梯度反向传播的优化方法,避免模型梯度优化产生的梯度爆炸或梯度消失问题,融入观测值实现模型的可持续性学习功能。(5)构建En KF-LSTM-ED模型,通过实现encoder-decoder结构,预测方法自循环预测更改为序列到序列的预测,提高En KF-LSTM模型针对时序数据中短期预测能力。结果表明:本文提出模型具有一定的预测准确性。本模型中,SBAS-In SAR技术获取了康定机场和双流机场跑道沉降时序数据,分析主要沉降区域和数据变化趋势,设计实验案例对比说明,En KF-LSTM模型和En KF-BP模型较未优化模型的预测准确性提升约12%,并且En KF-LSTM模型较En KF-BP模型具有更强的预测能力;序列到序列的En KF-LSTM-ED预测方法较循环预测方法可信预测结果提高到30天。

关键词：机场跑道安全沉降预测 SBAS-InSAR 数据同化神经网络 encoder-decoder

来源：评论

学校读者我要写书评

暂无评论

Machine learning and deep learning-based sentiment analysis of IMDB user reviews

Machine learning and deep learning-based sentiment analysis ...

引用

作者： Shengjie Xia School of Information Science and Engineering Lanzhou University

Movie reviews have always been a popular and enduring subject of interest among researchers. Sentiment analysis plays a significant role in this domain. The utilization of machine learning and natural language processing techniques can provide valuable insights into the emotional responses of audiences towards movies, as well as facilitate the appraisal of their reputation and market potential. This is achieved through the analysis of sentiment expressed in movie reviews. Furthermore, this approach is highly valuable in various application domains such as data mining, web mining, and social media analysis. This paper aims to conduct a comparative analysis by utilizing typical models based on machine learning and neural networks,along with the integration of natural language processing techniques. The IMDB database, which contains 50,000 reviews, will be used, and data preprocessing will be performed before applying these models. By comparing the accuracy of each model, insights regarding movie reviews can be derived.

关键词： Machine learning encoder-decoder sentiment analysis movies reviews binary classification

来源：评论

学校读者我要写书评

暂无评论

Split, Embed and Merge: An accurate table structure recognizer

引用

PATTERN RECOGNITION 2022年 126卷 108565-108565页

作者： Zhang, Zhenrong Zhang, Jianshu Du, Jun Wang, Fengren Univ Sci & Technol China Natl Engn Res Ctr Speech & Language Informat Proc 96 JinZhai Rd Hefei Anhui Peoples R China IFLYTEK Res Hefei Anhui Peoples R China

Table structure recognition is an essential part for making machines understand tables. Its main task is to recognize the internal structure of a table. However, due to the complexity and diversity in their structure and style, it is very difficult to parse the tabular data into the structured format which machines can understand, especially for complex tables. In this paper, we introduce Split, Embed and Merge (SEM), an accurate table structure recognizer. SEM is mainly composed of three parts, splitter, embedder and merger. In the first stage, we apply the splitter to predict the potential regions of the table row/column separators, and obtain the fine grid structure of the table. In the second stage, by taking a full consideration of the textual information in the table, we fuse the output features for each table grid from both vision and text modalities. Moreover, we achieve a higher precision in our experiments through providing additional textual features. Finally, we process the merging of these basic table grids in a self-regression manner. The corresponding merging results are learned through the attention mechanism. In our experiments, SEM achieves an average F1-Measure of 97 . 11% on the SciTSR dataset which outperforms other methods by a large margin. We also won the first place of complex tables and third place of all tables in Task-B of ICDAR 2021 Competition on Scientific Literature Parsing. Extensive experiments on other publicly available datasets further demonstrate the effectiveness of our proposed approach. (c) 2022 Elsevier Ltd. All rights reserved.

关键词： Table structure recognition Self-regression Attention mechanism encoder-decoder

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：