检索结果-内蒙古大学图书馆

2nd World Conference on Intelligent and 3D Technologies, WCI3DT 2023

作者： Tang, Haochun Zhang, Jindong Yang, Jing College of Software Jilin University Changchun130012 China College of Computer Science and Technology Jilin University Changchun130012 China Key Laboratory of Symbol Computation and Knowledge Engineering of the Ministry of Education Jilin University Changchun130012 China

ISBN: (纸本)9789819970100

This paper focuses on the optimization and improvement of the line-of-sight tracking algorithm b.sed on monocular vision, and aims to achieve a series of complex functions through image analysis b.sed on line-of-sight tracking technology and more in-depth analysis of monocular images. Therefore, we use the web.aze framework and use face detection models and feature point recognition lib.aries to analyze the recognition rate and provide further development of the technology. © 2024, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Research on the Parsing Algorithm of Monocular Visual Structured Data b.sed on YOLOv5 2nd

Research on the Parsing Algorithm of Monocular Visual Struct...

引用

2nd World Conference on Intelligent and 3D Technologies, WCI3DT 2023

作者： Lu, Wanli Zhang, Wen Sun, Mingrui Zhang, Jindong College of Computer Science and Technology Jilin University Changchun130012 China College of Software Jilin University Changchun130012 China Key Laboratory of Symbol Computation and Knowledge Engineering of the Ministry of Education Jilin University Changchun130012 China

ISBN: (纸本)9789819721436

Visual perception plays an important role in autonomous driving technology. The two key factors in visual perception tasks are monocular ob.ect detection and structured data analysis. In this paper, a structured data parsing algorithm b.sed on monocular vision is proposed. Firstly, using the single-stage YOLOv5 ob.ect detection model, the target was located and its kind was identified. Secondly, the 3D position of the ob.ect in the camera coordinate system is calculated according to the mapping relationship b.tween the imaging point and the target point. Finally, the distance b.tween the ob.ect and the camera is calculated b. using the pinhole camera model, and the size of the ob.ect and the horizontal declination Angle relative to the optical center of the camera are estimated b. geometric derivation. This method not only has a simple principle, b.t also guarantees the precision of structured data output. Experimental findings support the method's viab.lity and efficiency. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Migration: An Efficient Explorer Operator to Guide Swarm Evolution

Migration: An Efficient Explorer Operator to Guide Swarm Evo...

引用

Medical Artificial Intelligence (MedAI), IEEE International Conference on

作者： Xiaosong Han Chengyang Zhou Yanchun Liang Renchu Guan Key Laboratory for Symbol Computation and Knowledge Engineering of National Education Ministry College of Computer Science and Technology Jilin University Changchun China Key Laboratory for Symbol Computation and Knowledge Engineering of National Education Ministry College of Computer Science and Technology Zhuhai Laboratory of Key Laboratory for Symbol Computation and Knowledge Engineering of Ministry of Education Jilin University Zhuhai College of Science and Technology Zhuhai China

Particle Swarm Optimization with Migration (MPSO) is proposed to solve the issue that PSO will encounter unb.arab.e time cost prob.ems when dealing with High-dimension, Expensive and b.ack-b.x ob.ective function tasks. The Migration operator is inspired b. the migration of Salmon. Salmon will start a dangerous journey from the ocean to the home rivers for reproduction. The process of the entire b.havior is similar to the reduction and recovery of dimension. Therefore, we design the Migration operator where a pretrained Wasserstein Autoencoders (WAE) is applied to simulate the migration b.havior to accelerate the process of evolution in PSO, and we use Least-Squares Regression in lower space to produce b.tter generation. In comparison with famous b.seline methods in some b.nchmark functions, MPSO converges faster and more accurately, which shows the great potential of migration operations.

关键词：

来源：评论

学校读者我要写书评

暂无评论

PNESR-DDI: An Effective Drug-Drug Interaction Prediction Model b.sed on Pretraining Method and Enhanced Sub.raph Reconstruction

PNESR-DDI: An Effective Drug-Drug Interaction Prediction Mod...

引用

IEEE International Conference on b.oinformatics and b.omedicine (b.b.)

作者： Ke Chen Xiaosong Han Xiaoran Li Yanchun Liang Dong Xu Renchu Guan Key Laboratory for Symbol Computation and Knowledge Engineering of National Education Ministry College of Software Jilin University Changchun China Key Laboratory for Symbol Computation and Knowledge Engineering of National Education Ministry College of Computer Science and Technology Jilin University Changchun China Zhuhai Laboratory of Key Laboratory of Symbol Computation and Knowledge Engineering of Ministry of Education School of Computer Science Zhuhai College of Science and Technology Zhuhai China Department of Electrical Engineering and Computer Science Christopher S. Bond Life Sciences Center University of Missouri Columbia USA

ISBN: (数字)9798350386226

ISBN: (纸本)9798350386233

Drug-Drug Interaction (DDI) task plays a crucial role in clinical treatment and drug development. Recently, deep learning methods have b.en successfully applied for DDI prediction. However, training deep learning models always need large amount of data, while known DDIs are scarce. To address this challenge, a graph neural network-b.sed DDI prediction model named PNESR-DDI is proposed, which compensates for the lack of DDIs b. enriching drug representations. First, to ob.ain initial node representations that incorporate rich semantic information from the b.omedical knowledge graph (KG), a link prediction pre-training method on external KG is proposed in the node emb.dding pre-training module. Then, considering the large scale of the KG, sub.raph extraction for the target drug pairs is introduced to reduce noise and decrease computational complexity in the sub.raph anchoring module. After that, the sub.raph is updated, and node similarities are propagated in the sub.raph reconstruction module. b.sed on the node similarity scores, the sub.raph is pruned and reconstructed, which adjusts node representations to b. more conducive to DDI prediction. Finally, the drug emb.ddings, sub.raph representations, and drug fingerprint features are concatenated to predict DDIs. PNESRDDI is evaluated on two b.nchmark DDI datasets: Drugb.nk and TWOSIDES. Experiment results show that PNESR-DDI achieves b.tter performance than b.selines. Ab.ation results validate the effectiveness of the pre-training method and the adaptive sub.raph reconstruction strategy.

关键词： Drugs Deep learning Adaptation models Visualization b.ological system modeling Noise knowledge graphs Predictive models Fingerprint recognition Transformers

来源：评论

学校读者我要写书评

暂无评论

An AutoEncoder-b.sed Matrix Factorization Approach to Estimating Cell Proportion from b.lk Tumor RNA-seq Data

An AutoEncoder-Based Matrix Factorization Approach to Estima...

引用

2021 IEEE International Conference on b.oinformatics and b.omedicine, b.b. 2021

作者： Xu, Yingze Wang, Yan Xie, Xuping Wang, Feilong Chen, Qiong Sun, Huiyan Jilin University College of Computer Science and Technology Key Laboratory of Symbol Computation and Knowledge Engineering of Ministry of Education Changchun China Jilin University School of Artificial Intelligence Key Laboratory of Symbol Computation and Knowledge Engineering of Ministry of Education Changchun China

ISBN: (纸本)9781665401265

The deconvolution of infiltrating immune cells and stromal cells from complex tumor tissues is significant for studying the impact of these cells on tumor development, as well as assisting cancer therapies. Integrating cell-specific marker genes from several references, we put forward an AutoEncoder-b.sed matrix factorization method to estimate the cell proportion in heterogeneous samples. The proportion predicted b. our method achieved over 0.95 Pearson correlation coefficient (PCC) with the ground truth. Moreover, through analyzing association b.tween cell proportion of tumor tissues and clinical information of the tumor patients, we found that the proportion of cancer-associated fib.ob.ast (CAF) gradually increased with the progress of tumor, and the proportion of b.cell in tissues was significantly related to the five-year survival rate of tumor patients. © 2021 IEEE.

关键词： RNA

来源：评论

学校读者我要写书评

暂无评论

Independent and Collab.rative Demosaicking Neural Networks 5

Independent and Collaborative Demosaicking Neural Networks

引用

5th ACM International Conference on Multimedia in Asia, MMAsia 2023

作者： Niu, Yan Zhang, Lixue Li, Chenlai State Key Laboratory of Symbol Computation and Knowledge Engineering Ministry of Education College of Computer Science and Technology Jilin University Jilin Changchun China College of Computer Science and Technology Jilin University Changchun China School of Artificial Intelligence Shenzhen Polytechnic University Shenzhen China

ISBN: (纸本)9798400702051

Existing demosaicking neural models generally reconstruct the red, green and b.ue channels b. one unified network. This has the advantage of allowing the RGb.channels to share latent features. However, it is unnoticed that the training samples are severely unb.lanced among the three channels. As a consequence, a unified model trained to fit the red or b.ue sample distrib.tions is prone to under-fit the green samples. In this paper, we demonstrate the existence of such conflicts, and analyze the reason b.hind this phenomenon. To solve this disadvantage, we decouple the traditional three-in-one demosaicking to three independent b.t collab.rative sub.tasks, respecting the distrib.tion of samples in each spectrum. We show that, b. our decoupling strategy, a very simple architecture can achieve high accuracy with high efficiency. For sub.tantiation, we construct a model that has merely 1.3M parameters in total, and compares its accuracy to state of the art methods of similar or larger model size, across a wide variety of b.nchmark datasets. The proposed independent and collab.rative architecture gains performance improvement over comparab.e models in b.th accuracy and inference time. © 2023 Copyright held b. the owner/author(s).

关键词： Network architecture

来源：评论

学校读者我要写书评

暂无评论

Crossed-Time Delay Neural Network for Speaker Recognition 1

引用

27th International Conference on MultiMedia Modeling, MMM 2021

作者： Chen, Liang Liang, Yanchun Shi, Xiaoshu Zhou, You Wu, Chunguo Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education College of Computer Science and Technology Jilin University Changchun130012 China Zhuhai Laboratory of Key Laboratory of Symbol Computation and Knowledge Engineering of Ministry of Education School of Computer Zhuhai College of Jilin University Zhuhai519041 China

ISBN: (数字)9783030678326

ISBN: (纸本)9783030678319

Time Delay Neural Network (TDNN) is a well-performing structure for deep neural network-b.sed speaker recognition systems. In this paper we introduce a novel structure, named Crossed-Time Delay Neural Network (CTDNN) to enhance the performance of current TDNN for speaker recognition. Inspired b. the multi-filters setting of convolution layers from convolution neural networks, we set multiple time delay units with different context size at the b.ttom layer and construct a multilayer parallel network. The proposed CTDNN gives significant improvements over original TDNN on b.th speaker verification and identification tasks. It outperforms in VoxCeleb. dataset in verification experiment with a 2.6% ab.olute Equal Error Rate improvement. In few shots condition, CTDNN reaches 90.4% identification accuracy, which doub.es the identification accuracy of original TDNN. We also compare the proposed CTDNN with another new variant of TDNN, Factorized-TDNN, which shows that our model has a 36% ab.olute identification accuracy improvement under few shots condition. Moreover, the proposed CTDNN can handle training with a larger b.tch more efficiently and hence, utilize calculation resources economically. © 2021, Springer Nature Switzerland AG.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Nutrition feature b.sis of food classification in traditional chinese medicine 7

Nutrition feature basis of food classification in traditiona...

引用

7th International Conference on b.g Data Computing and Communications, b.gCom 2021

作者： Zhao, Fucheng Han, Xiaosong Zhang, Yikun Niu, Peiyu Cheng, Jun Yi, Wan Jilin University Key Laboratory for Symbol Computation and Knowledge Engineering of National Education Ministry College of Computer Science and Technology Changchun130012 China

ISBN: (纸本)9781665442527

In TCM (Traditional Chinese Medicine) theory, the cold-and-hot property of food is considered as an important information to guide people's daily diet and keep them healthy. b.t the classification of this property mostly relies on experience from medical b.oks, which lacks corresponding molecular chemistry b.sis and makes differentiation very difficult for ordinary people. To solve ab.ve question, in this paper, we collected 500 kinds of food and their nutrients, used ANOVA algorithm for feature selection to find out the nutrient feature that have a greater correlation with the cold and hot properties of food, and then used LightGb. algorithm to classify the processed data set. Finally, we constructed a reliab.e and accurate classification model and found the related nutritional elements are Tyrosine, Arginine, Thiamine, Vitamin b. and Selenium. © 2021 IEEE.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

2.5D ASF-UNet: Adjacent Slice Spatial Feature Fusion Model for WMH Segmentation from 3D MR b.ain Image

2.5D ASF-UNet: Adjacent Slice Spatial Feature Fusion Model f...

引用

IEEE International Conference on b.oinformatics and b.omedicine (b.b.)

作者： Lan Huang Yinglu Sun Ziqi Zhao Chunjie Guo Yan Wang Key Laboratory of Symbol Computation and Knowledge Engineering of Ministry of Education College of Computer Science and Technology Jilin University Changchun China Department of Radiology the First Hospital of Jilin University Changchun China

ISBN: (数字)9798350386226

ISBN: (纸本)9798350386233

Segmenting b.ain white matter hyperintensities (WMH) from 3D Magnetic Resonance (MR) images is crucial for the diagnosis, treatment, and prognosis of Multiple Sclerosis (MS). Unlike common 2D images, this task is more challenging and time-consuming. Classical deep learning methods for 3D image segmentation face two main challenges: 1) Pure 3D networks have more parameters and are prone to overfitting. 2) When using 2D networks to segment slices of 3D images, the lack of 3D structural information results in sub.ptimal segmentation after reconstruction. To address these difficulties, we propose the 2.5D ASF-UNet, which employs the 2.5D workflow and uses adjacent slices as the input for 2D segmentation network ASF-UNet. In ASF-UNet, separate down-sampling paths are used for the adjacent slices, and the Local Spatial Attention Module (LSAM) is designed to more effectively integrate 3D spatial information into the 2D network. Additionally, the Conv_Spectral_b.ock (CSb. is designed to extract and integrate local and glob.l features. It allows the model to capture glob.l spatial structures while preserving detailed information. Experimental results on MICCAI MSSEG 2016 and Local MS datasets show that 2.5D ASF-UNet achieve b.tter segmentation performance than other deep learning methods.

关键词： Deep learning Image segmentation Solid modeling Three-dimensional displays Multiple sclerosis Feature extraction b.ain modeling White matter Prognostics and health management Overfitting

来源：评论

学校读者我要写书评

暂无评论

A Survey on Performance Optimization of High-Level Synthesis Tools

引用

Journal of Computer Science & Technology 2020年第3期35卷 697-720页

作者： Lan Huang Da-Lin Li Kang-Ping Wang Teng Gao Adriano Tavares College of Computer Science and Technology Jilin UniversityChangchun 130012China Key Laboratory of Symbol Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun 130012China Zhuhai Laboratory of Key Laboratory of Symbol Computation and Knowledge Engineering of Ministry of Education Zhuhai College of Jilin University Zhuhai 519041China Algorithm Center University of MinhoGuimaraes 4800058Portugal

Field-programmab.e gate arrays(FPGAs)have recently evolved as a valuab.e component of the heterogeneous *** register transfer level(RTL)design flows demand the designers to b. experienced in hardware,resulting in a possib.e failure of ***-level synthesis(HLS)permits designers to work at a higher level of ab.traction through synthesizing high-level language programs to RTL *** provides a promising approach to solve these ***,the performance of HLS tools still has *** example,designers remain exposed to various aspects of hardware design,development cycles are still time consuming,and the quality of results(QoR)of HLS tools is far b.hind that of RTL *** this paper,we survey the literature pub.ished since 2014 focusing on the performance optimization of HLS *** with previous work,we extend the scope of the performance of HLS tools,and present a set of three-level evaluation criteria,covering from ease of use of the HLS tools to promotion on specific metrics of *** also propose performance evaluation equations for describ.ng the relation b.tween the performance optimization and the *** find that it needs more efforts on the ease of use for efficient HLS *** suggest that it is b.tter to draw an analogy b.tween the HLS development process and the emb.dded system design process,and to provide more elastic HLS methodology which integrates FPGAs virtual machines.

关键词： evaluation criterion field-programmab.e gate array(FPGA) high-level synthesis(HLS) performance optimization quality of results(QoR)

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：