检索结果-内蒙古大学图书馆

International Automatic Control Conference

作者： Xiao-Wei Huang Xiu-Zhi Chen Yen-Lin Chen Master Program in Artificial Intelligence Technology National Taipei University of Technology Taipei Taiwan Department of Computer Science and Information Engineering National Taipei University of Technology Taipei Taiwan

Traditional perception systems for TJA (Traffic Jam Assistance) are mostly implemented by fusing images with radar or lidar. As computer vision techniques become more powerful, cameras can almost replace the need for radar and lidar in perception tasks, which reduces the hardware cost of the system. In this research, we propose a camera-only perception system for TJA, which is able to provide the information of the vehicles ahead and the drivable area. The proposed system has been evaluated through real-world scenario sequences, and proved that it achieves high robustness, which is highly possible to be adopted for TJA development.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Robust Object Detection Model for UAV Application

Robust Object Detection Model for UAV Application

引用

International Automatic Control Conference

作者： Chun Ku Xiu-Zhi Chen Yen-Lin Chen Master Program in Artificial Intelligence Technology National Taipei University of Technology Taipei Taiwan Department of Computer Science and Information Engineering National Taipei University of Technology Taipei Taiwan

Not only common issues on object detection task need to be deal with, for Unmanned Aerial Vehicle (UAV) applications, small object is one of the critical problems that needs to be solved. YOLOv7 is a powerful network architecture that provides high efficiency and accuracy object detection results. This paper adopts YOLOv7 as an object detection model for two different kinds of targets, one is vehicle, and the other is ocean flotsam. By training the model with open datasets and fine-tuning the model with self-collected datasets, we prove through sequences collected from real-world scenarios that YOLOv7 is able to provide robust and accurate object detection results, including vehicles and ocean flotsam, with real-time efficiency. Based on such experimental result, we confirmed that YOLOv7 can be the baseline for object detection model development.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Evolving artificial neural network using simple augmenting weight matrix method

Evolving artificial neural network using simple augmenting w...

引用

IEEE International Conference on Industrial Informatics (INDIN)

作者： Dong-Hyun Lee Ju-Jang Lee Robotics Program Korea Advanced Institute of Science and Technology Daejeon South Korea Electrical Engineering and Computer Science Korea Advanced Institute of Science and Technology Daejeon South Korea

In this paper, a simple method is proposed to evolve artificial neural networks(ANNs) using augmenting weight matrix method(AWMM). ANNs' architecture and connection weights can be evolved simultaneously by AWMM, and their structures incrementally are growing up from minimal structure. It is a non-mating method. It employs 5 mutation operators: add connection, add node, delete connection, delete node, and new initial weight. And the connection weight is trained by the simplified alopex method, which is a correlation based method for solving optimization problem. In AWMM, structural information is encoded to weighting matrix, and the matrix is augmenting as the hidden nodes are added.

关键词：

来源：评论

学校读者我要写书评

暂无评论

No-boundary thinking: a viable solution to ethical data-driven AI in precision medicine

引用

AI and ethics 2022年第4期2卷 635-643页

作者： Tayo Obafemi-Ajayi Andy Perkins Bindu Nanduri Donald C Wunsch Ii James A Foster Joan Peckham Engineering Program Missouri State University Springfield MO USA. Department of Computer Science and Engineering Mississippi State University Starkville MS USA. Department of Comparative Biomedical Sciences College of Veterinary Medicine Mississippi State University Starkville MS USA. Electrical & Computer Engineering Department Missouri University of Science and Technology Rolla MO USA. Biological Sciences Department University of Idaho Moscow ID USA. Computer Science & Statistics Department University of Rhode Island Kingston RI USA.

Today Artificial Intelligence (AI) supports difficult decisions about policy, health, and our personal lives. The AI algorithms we develop and deploy to make sense of information, are informed by data, and based on models that capture and use pertinent details of the population or phenomenon being analyzed. For any application area, more importantly in precision medicine which directly impacts human lives, the data upon which algorithms are run must be procured, cleaned, and organized well to assure reliable and interpretable results, and to assure that they do not perpetrate or amplify human prejudices. This must be done without violating basic assumptions of the algorithms in use. Algorithmic results need to be clearly communicated to stakeholders and domain experts to enable sound conclusions. Our position is that AI holds great promise for supporting precision medicine, but we need to move forward with great care, with consideration for possible ethical implications. We make the case that a no-boundary or convergent approach is essential to support sound and ethical decisions. No-boundary thinking supports problem definition and solving with teams of experts possessing diverse perspectives. When dealing with AI and the data needed to use AI, there is a spectrum of activities that needs the attention of a no-boundary team. This is necessary if we are to draw viable conclusions and develop actions and policies based on the AI, the data, and the scientific foundations of the domain in question.

关键词： Artificial intelligence Data science Ethics Precision medicine

来源：评论

学校读者我要写书评

暂无评论

Cost-Efficient Snubbers for Boost Converters

Cost-Efficient Snubbers for Boost Converters

引用

Asia-Pacific Symposium on Electromagnetic Compatibility (APEMC)

作者： Yeu-Torng Yau Thanh-Phu. Luu Department of Ph.D. Program Prospective Technology of Electrical Engineering and Computer Science National Chin-Yi University of Technology Taichung Taiwan Department of Electrical Engineering National Chin-Yi University of Technology Taichung Taiwan

ISBN: (数字)9784885523472

ISBN: (纸本)9798350349498

Conventional boost converters operating with hard-switching result in low conversion efficiency and increased electromagnetic interference emissions. In this paper, a cost-efficient passive snubber is proposed with a few additional components: two diodes, one capacitor, and one inductor. Moreover, because these snubber components are not located on the main power processing path, they only require low ratings, resulting in improved cost-effectiveness.

关键词： Snubbers Electromagnetic interference Capacitors Switches Electromagnetic compatibility Inductors

来源：评论

学校读者我要写书评

暂无评论

Calculation of the radar ambiguity function from time-domain measurement data for real-time, amplifier-in-the-loop waveform optimization

Calculation of the radar ambiguity function from time-domain...

引用

ARFTG Conference

作者： Matthew Fellows Charles Baylis Lawrence Cohen Robert J. Marks Department of Electrical and Computer Engineering Baylor University Wireless and Microwave Circuits and Systems Program Waco TX USA Naval Research Laboratory Radar Division Washington DC USA

The ambiguity function is a measure of a radar's range and Doppler detection capability. Cognitive radar systems require the capability to adjust the waveform in realtime to obtain desired range/Doppler detection capability while meeting stringent spectral requirements. While the ambiguity function of the waveform input to the transmitter can be simulated, it is the ambiguity function of the transmitter power amplifier's output waveform that will be used for the detection. As such, it is very helpful to be able to measure the ambiguity function output from a power amplifier in the optimization process. This paper describes a technique that can be used to quickly calculate the ambiguity function for the output waveform from the radar amplifier as measured on an oscilloscope. Brief examination is also given to the effect of amplifier nonlinearity on the ambiguity function.

关键词： Chirp Optimization Radar detection Radar measurements Doppler radar Distortion measurement

来源：评论

学校读者我要写书评

暂无评论

Real-time tool wear identification using sensor integration with neural network

Real-time tool wear identification using sensor integration ...

引用

IEEE Conference on Decision and Control

作者： N. Levy MengChu Zhou Yu Quan Department of Mechanical and Industrical Engineering New Jersey Institute of Technology Newark NJ USA Department of Electrical and Computer Engineering New Jersey Institute of Technology Newark NJ USA Manufacturing Engineering Program New Jersey Institute of Technology Newark NJ USA

Real-time identification of tool wear in shop floor environment is essential for optimization of machining processes and implementation of automated manufacturing systems. In this paper. the signals obtained from acoustic emission and power sensors during machining processes are analyzed and a set of feature parameters characterizing the tool wear condition are extracted. In order to realize the realtime tool wear condition monitoring for different cutting conditions, a sensor integration strategy which combines the information from multiple sensors (acoustic emission sensor and power sensor) and machining parameters is proposed. A neural network based on improved back-propogation algorithm is developed and a prototype scheme for realtime identification of tool wear is implemented. Experiments under different conditions have proved that a higher rate of tool wear identification can be achieved by using the sensor integration model with neural network. The results also indicated that the neural network is a very effective method of sensor integration for online monitoring of tool abnormalities.< >

关键词： Wearable sensors Sensor phenomena and characterization Acoustic sensors Machining Neural networks Acoustic emission Real time systems Manufacturing systems Signal processing Signal analysis

来源：评论

学校读者我要写书评

暂无评论

Mean-shift tracker with face-adjusted model

Mean-shift tracker with face-adjusted model

引用

International Conference on Control, Automation and Systems ( ICCAS)

作者： Jeehyun Goya Choe Joon-Hong Seok Ju-Jang Lee Robotics Program Korea Advanced Institute of Science and Technology Daejeon South Korea Electrical Engineering and Computer Science Korea Advanced Institute of Science and Technology Daejeon South Korea

Mean-shift algorithm shows robust performances in various object-tracking technologies including face tracking. Due to its robustness and accuracy, mean-shift algorithm is regarded as one of the best ways to apply in object-tracking technology in computer vision fields. However, it has a drawback of getting into a bottleneck state when faced with a speedy object moving beyond its window size within one image frame interval time. The time required to calculate mean-shift vector could be much lessened with lesser memory when color model is adjusted to the previously known target information. This paper shows the building process of target-adjusted model with a non-uniform quantization. The target color model dealt in this paper is the one used for deriving mean-shift vector. It is a kernel model containing both the color and distance information. This paper gives scheme to efficiently deal with color information in the model. Through a proper selection of color bins, unimportant color values were reduced to a small amount. As a result, the computing time of the mean-shift vector in face-tracking was shortened while maintaining robustness and accuracy.

关键词： Target tracking Face detection Robotics and automation Electronic mail Robustness Quantization Kernel Pixel Focusing Face recognition

来源：评论

学校读者我要写书评

暂无评论

Orchard: building large cancer phylogenies using stochastic combinatorial search

arXiv

引用

arXiv 2023年

作者： Kulman, Ethan Kuang, Rui Morris, Quaid Computational and Systems Biology Program Sloan Kettering Institute New YorkNY United States Department of Computer Science and Engineering University of Minnesota MinneapolisMN United States

Phylogenies depicting the evolutionary history of genetically heterogeneous subpopulations of cells from the same cancer, i.e., cancer phylogenies, offer valuable insights about cancer development and guide treatment strategies. Many methods exist that reconstruct cancer phylogenies using point mutations detected with bulk DNA sequencing. However, these methods become inaccurate when reconstructing phylogenies with more than 30 mutations, or, in some cases, fail to recover a phylogeny altogether. Here, we introduce Orchard, a cancer phylogeny reconstruction algorithm that is fast and accurate using up to 1000 mutations. Orchard samples without replacement from a factorized approximation of the posterior distribution over phylogenies, a novel result derived in this paper. Each factor in this approximate posterior corresponds to a conditional distribution for adding a new mutation to a partially built phylogeny. Orchard optimizes each factor sequentially, generating a sequence of incrementally larger phylogenies that ultimately culminate in a complete tree containing all mutations. Our evaluations demonstrate that Orchard outperforms state-of-the-art cancer phylogeny reconstruction methods in reconstructing more plausible phylogenies across 90 simulated cancers and 14 B-progenitor acute lymphoblastic leukemias (B-ALLs). Remarkably, Orchard accurately reconstructs cancer phylogenies using up to 1,000 mutations. Additionally, we demonstrate that the large and accurate phylogenies reconstructed by Orchard are useful for identifying patterns of somatic mutations and genetic variations among distinct cancer cell subpopulations. © 2023, CC BY.

关键词： Plant diseases

来源：评论

学校读者我要写书评

暂无评论

An efficient heuristic for calculating a protected path with specified nodes

An efficient heuristic for calculating a protected path with...

引用

International Workshop on Reliable Networks Design and Modeling (RNDM)

作者： Lúcia Martins Teresa Gomes David Tipper INESC Coimbra DEEC Coimbra Portugal Department of Electrical and Computer Engineering University of Coimbra Coimbra Portugal Graduate Telecommunications and Networking Program University of Pittsburgh Pittsburgh PA USA

The problem of determining a path between two nodes in a network that must visit specific intermediate nodes arises in a number of contexts. For example, one might require traffic to visit nodes where it can be monitored by deep packet inspection for security reasons. In this paper a new recursive heuristic is proposed for finding the shortest loopless path, from a source node to a target node, that visits a specified set of nodes in a network. In order to provide survivability to failures along the path, the proposed heuristic is modified to ensure that the calculated path can be protected by a node-disjoint backup path. The performance of the heuristic, calculating a path with and without protection, is evaluated by comparing with an integer linear programming (ILP) formulation for each of the considered problems. The ILP solver may fail to obtain a solution in a reasonable amount of time, especially in large networks, which justifies the need for effective, computationally efficient heuristics for solving these problems. Our numerical results are also compared with previous heuristics in the literature.

关键词： Algorithm design and analysis Amplitude shift keying Integer linear programming Routing Routing protocols Heuristic algorithms Electronic mail

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：