检索结果-内蒙古大学图书馆

Root identification in minirhizotron imagery with multiple instance learning

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Yu, Guohao Zare, Alina Sheng, Hudanyun Matamala, Roser Reyes-Cabrera, Joel Frischi, Felix B. Juenger, Thomas E. Department of Electrical and Computer Engineering University of Florida GainesvilleFL32611 United States Argonne National Laboratory LemontIL60439 United States Division of Plant Sciences University of Missouri ColumbiaMO65211 United States Department of Integrative Biology University of Texas at Austin AustinTX78712 United States

In this paper, multiple instance learning (MIL) algorithms to automatically perform root detection and segmentation in minirhizotron imagery using only image-level labels are proposed. Root and soil characteristics vary from location to location, thus, supervised machine learning approaches that are trained with local data provide the best ability to identify and segment roots in minirhizotron imagery. However, labeling roots for training data (or otherwise) is an extremely tedious and time-consuming task. This paper aims to address this problem by labeling data at the image level (rather than the individual root or root pixel level) and train algorithms to perform individual root pixel level segmentation using MIL strategies. Three MIL methods (MI-ACE, miSVM, MIForests) were applied to root detection and compared to non-MIL approches. The results show that MIL methods improve root segmentation in challenging minirhizotron imagery and reduce the labeling burden. In our results, miSVM outperformed other methods. The MI-ACE algorithm was a close second with an added advantage that it learned an interpretable root signature which identified the traits used to distinguish roots from soil and did not require parameter selection. Note to Practitioners-Minirhizotrons provide an efficient and non-destructive way to collect plant roots for studying root system dynamically. However existing software used to extract roots from minirhizotron image require significant, tedious manual marking of roots and soil in the collected imagery. Due to this slow manual process, the ability to collect useful information from a large number of minirhizotron images is bottlenecked. In this paper, we propose an automated approach to segment roots from minirhizotron images. The proposed methods not only automatically identifies and segments root pixels in imagery, but also allow for an efficient approach to label training data. This allows one to be able to re-train the models for adaptation to

关键词： Soils

Recursive network estimation from binary-valued observation data

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Xing, Yu He, Xingkang Fang, Haitao Johansson, Karl Henrik Key Laboratory of Systems and Control Academy of Mathematics and Systems Science Chinese Academy of Sciences Beijing100190 China School of Mathematical Sciences University of Chinese Academy of Sciences Beijing100049 China Division of Decision and Control Systems School of Electrical Engineering and Computer Science KTH Royal Institute of Technology StockholmSE-10044 Sweden

This paper studies the problem of recursively estimating the weighted adjacency matrix of a network out of a temporal sequence of binary-valued observations. The observation sequence is generated from nonlinear networked dynamics in which agents exchange and display binary outputs. Sufficient conditions are given to ensure stability of the observation sequence and identifiability of the system parameters. It is shown that stability and identifiability can be guaranteed under the assumption of independent standard Gaussian disturbances. Via a maximum likelihood approach, the estimation problem is transformed into an optimization problem, and it is verified that its solution is the true parameter vector under the independent standard Gaussian assumption. A recursive algorithm for the estimation problem is then proposed based on stochastic approximation techniques. Its strong consistency is established and convergence rate analyzed. Finally, numerical simulations are conducted to illustrate the results and to show that the proposed algorithm is insensitive to small unmodeled factors. Copyright © 2019, The Authors. All rights reserved.

关键词： Approximation theory

Transient Response Analysis of Metropolis Learning in Games

学校读者我要写书评

暂无评论

IFAC-PapersOnLine 2017年第1期50卷 9661-9667页

作者： Jaleel H. Shamma J.S. Computer Electrical and Mathematical Sciences & Engineering (CEMSE) Division King Abdullah University of Science and Technology Thuwal Saudi Arabia

The objective of this work is to provide a qualitative description of the transient properties of stochastic learning dynamics like adaptive play, log-linear learning, and Metropolis learning. The solution concept used in these learning dynamics for potential games is that of stochastic stability, which is based on the stationary distribution of the reversible Markov chain representing the learning process. However, time to converge to a stochastically stable state is exponential in the inverse of noise, which limits the use of stochastic stability as an effective solution concept for these dynamics. We propose a complete solution concept that qualitatively describes the state of the system at all times. The proposed concept is prevalent in control systems literature where a solution to a linear or a non-linear system has two parts, transient response and steady state response. Stochastic stability provides the steady state response of stochastic learning rules. In this work, we study its transient properties. Starting from an initial condition, we identify the subsets of the state space called cycles that have small hitting times and long exit times. Over the long time scales, we provide a description of how the distributions over joint action profiles transition from one cycle to another till it reaches the globally optimal state. © 2017

关键词： game theory Learning theory Sensor networks Stochastic control

Two-level Clustering-based Target Detection Through Sensor Deployment

学校读者我要写书评

暂无评论

Two-level Clustering-based Target Detection Through Sensor D...

International Conference on Information Fusion

作者： Chase Q. Wu Wuji Liu Satyabrata Sen Nageswara S. V. Rao Richard R. Brooks Guthrie Cordone Dept of Computer Science New Jersey Institute of Technology Newark NJ Computational Sciences and Engineering Division Oak Ridge National Laboratory Oak Ridge TN Dept of Electrical and Computer Engineering Clemson University Clemson SC

Target detection is one fundamental problem in many sensor network-based applications, and is typically tackled in two separate stages for sensor deployment and data fusion. We propose an integrated solution, referred to as SSEM, which combines 2-level clustering-based sensor deployment and Source Strength Estimate Map-based data fusion for the detection of a single static or moving target SSEM conducts the first level of clustering to determine a sensor deployment scheme and the second level of clustering to divide the deployed sensors into multiple subsets. For each sensor, the source strength is estimated at each grid point of the entire region based on a signal attenuation model, and for each subset of sensors, the target location is estimated using a strength distribution map-based statistical analysis method. A. final detection decision is made by thresholding the clustering degree of the target location estimates computed by all subsets of sensors. Compared with traditional grid-based target detection methods, SSEM significantly reduces the computation complexity and improves the detection performance through an integrated optimization strategy. Extensive simulation results show the performance superiority of the proposed solution over several well-known methods for target detection.

关键词： Sensor networks Source detection Sensor deployment Cluster analysis

Quantum Nanophotonics with Group IV defects in Diamond

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Bradac, Carlo Gao, Weibo Forneris, Jacopo Trusheim, Matt Aharonovich, Igor School of Mathematical and Physical Sciences Faculty of Science University of Technology Sydney 2007 Australia Division of Physics and Applied Physics School of Physical and Mathematical Sciences Nanyang Technological University Singapore637371 Singapore Torino10125 Italy Department of Electrical Engineering and Computer Science Massachusetts Institute of Technology CambridgeMA02139 United States

Diamond photonics is an ever-growing field of research driven by the prospects of harnessing diamond and its colour centres as suitable hardware for solid-state quantum applications. The last two decades have seen the field been shaped by the nitrogen-vacancy (NV) centre both with breakthrough fundamental physics demonstrations and practical realizations. Recently however, an entire suite of other diamond defects has emerged. They are M-V colour centres, where M indicates one of the elements in the IV column of the periodic table—Si, Ge, Sn and Pb, and V indicates lattice vacancies, i.e. missing next-neighbour carbon atoms. These centres exhibit a much stronger emission into the zero-phonon line then the NV centre, they display inversion symmetry, and can be engineered using ion implantation—all attractive features for scalable quantum photonic architectures based on solid-state, single-photon sources. In this perspective, we highlight the leading techniques for engineering and characterizing these diamond defects, discuss the current state-of-the-art of group IV-based devices and provide an outlook of the future directions the field is taking towards the realisation of solid-state quantum photonics with diamond. Copyright © 2019, The Authors. All rights reserved.

关键词： Diamonds

Distributed design of Robust Kalman filters over corrupted channels

学校读者我要写书评

暂无评论

arXiv 2019年

作者： He, Xingkang Johansson, Karl H. Fang, Haitao Division of Decision and Control Systems School of Electrical Engineering and Computer Science KTH Royal Institute of Technology StockholmSE-100 44 Sweden LSC NCMIS Academy of Mathematics and Systems Science Chinese Academy of Sciences Beijing100190 China School of Mathematical Sciences University of Chinese Academy of Sciences Beijing100049 China

We study distributed filtering for a class of uncertain systems over corrupted communication channels. We propose a distributed robust Kalman filter with stochastic gains, through which upper bounds of the conditional mean square estimation errors are calculated online. We present a robust collective observability condition, under which the mean square error of the distributed filter is proved to be uniformly upper bounded if the network is strongly connected. For better performance, we modify the filer by introducing a switching fusion scheme based on a sliding window. It provides a smaller upper bound of the conditional mean square error. Numerical simulations are provided to validate the theoretical results and show that the filter scales to large networks. Copyright © 2019, The Authors. All rights reserved.

关键词： Mean square error

Controlled spin-torque driven domain wall motion using staggered magnetic nanowires

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Mohammed, H. Risi, S. Al Jin, T.L. Kosel, J. Piramanayagam, S.N. Sbiaa, R. King Abdullah University of Science and Technology Computer Electrical and Mathematical Science and Engineering Thuwal23955-6900 Saudi Arabia Department of Physics Sultan Qaboos University P.O. Box 36 PC 123 Muscat Oman Division of Physics and Applied Physics School of Physical and Mathematical Sciences Nanyang Technological University Singapore

The growing demand for storage, due to big data applications, cannot be met by hard disk drives. Domain wall (DW) memory devices such as racetrack memory offer an alternative route to achieve high capacity storage. In DW memory, control of domain wall positions and their motion using spin-transfer torque are important challenges. In this paper, we demonstrate controlled domain wall motion using spin-transfer torque in staggered magnetic nanowires. The devices, fabricated using electron-beam lithography, were tested using a magneto-optical Kerr microscopy and electrical transport measurements. The depinning current, pinning potential and thermal stability were found to depend on the device dimensions of the staggering nanowires. Thus, the proposed staggering configuration helps to fine-tune the properties of domain wall devices for memory applications. Copyright © 2019, The Authors. All rights reserved.

关键词： Domain walls

A Statistical Framework for Optimizing and Evaluating MRI T1 and T2 Relaxometry Approaches

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Liu, Yang Buck, John R. Zheng, Shaokuan Consumer Electronics Division Bose Corporation 100 the Mountain Rd FraminghamMA01701 United States Department of Electrical and Computer Engineering University of Massachusetts Dartmouth 285 Old Westport Rd DartmouthMA02747 United States Department of Radiology University of Massachusetts Medical School WorcesterMA01655 United States

This paper proposes a statistical framework to optimize and evaluate the MR parameter T1 and T2 mapping capabilities for quantitative MRI relaxometry approaches. This analysis explores the intrinsic MR parameter estimate precision per unit scan time, termed the T1,2-to-noise ratio (TNR) efficiency, for different ranges of biologically realistic relaxation times. The TNR efficiency is defined in terms of the Cramer-Rao bound (CRB), a statistical lower bound on the parameter estimate variance. Geometrically interpreting the new TNR efficiency definition reveals a more complete model describing the factors controlling the T1/T2 mapping capabilities. This paper compares T1 mapping approaches including the inversion recovery (IR) family sequences and the Look-Locker (LL) sequence and simultaneous T1 and T2 mapping approaches including the spin-echo inversion recovery (SEIR) and driven equilibrium single pulse observation of T1/T2 (DESPOT) sequences. All pulse parameters are optimized to maximize the TNR efficiency within different T1 and T2 ranges of interest. Monte Carlo simulations with nonlinear least square estimation (NLSE) of T1/T2 validated the theoretical predictions on the estimator performances. Copyright © 2019, The Authors. All rights reserved.

关键词： Efficiency

Unsupervised decoding of spinal motor neuron spike trains for estimating hand kinematics following targeted muscle reinnervation

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Andalib, Arash Farina, Dario Vujaklija, Ivan Negro, Francesco Aszmann, Oskar C. Bashirullah, Rizwan Principe, Jose C. Department of Electrical and Computer Engineering University of Florida GainesvilleFL32611 United States Department of Bioengineering Imperial College London London United Kingdom Department of Clinical and Experimental Sciences Università degli Studi di Brescia Italy Laboratory of Bionic Extremity Reconstruction Division of Plastic and Reconstructive Surgery University of Vienna Austria

The performance of upper-limb prostheses is currently limited by the relatively poor functionality of unintuitive control schemes. This paper proposes to extract, from multichannel electromyographic signals (EMG), motor neuron spike trains and project them into lower dimensional continuous signals, which are used as multichannel proportional inputs to control the prosthetic’s actuators. These control signals are an estimation of the common synaptic input that the motor neurons receive. We use the simplest of metric learning approaches known as principal component analysis (PCA), as a linear unsupervised metric projection to extract the spectral information of high dimensional data into the subspace of the prosthetic hand degrees of freedom. We also investigate the importance of a rotation in the projection space that best aligns the PCA subspace with the space of degrees of freedom of the prosthetic hand, to attempt to approximate the sparseness properties of the motor axes (which are orthogonal), while no sparseness is enforced by PCA. Proof of concept for the feedforward path (open loop) is given by successful estimation of concurrent movements with up to three degrees of freedom. We also analyze and quantify the performance of the proposed decoding algorithm under implementation constraints and hardware limitations and propose a space-time subsampling strategy, to maximize projection fidelity, in each extracted source over time. The results confirm that the proposed decoding approach can directly project the motor neuron spike trains into kinematics for simultaneous and proportional prosthesis control that can be extended to multiple degrees of freedom. We show that the method is robust to reducing the training data in space (number of sources) and time, which makes it potentially suitable for clinical applications. Copyright © 2019, The Authors. All rights reserved.

关键词： Neurons