检索结果-内蒙古大学图书馆

Single-shot fringe projection profilometry based on deep learning and computer graphics

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Wang, Fanzhou Wang, Chenxing Guan, Qingze School of Automation Southeast University 2# Sipailou Xuanwu Nanjing210096 China Key laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Southeast University Nanjing210096 China Department of Computer Science University College London Gower St Bloomsbury LondonWC1E 6BT United Kingdom

Multiple works have applied deep learning to fringe projection profilometry (FPP) in recent years. However, to obtain a large amount of data from actual systems for training is still a tricky problem, and moreover, the network design and optimization still worth exploring. In this paper, we introduce computer graphics to build virtual FPP systems in order to generate the desired datasets conveniently and simply. The way of constructing a virtual FPP system is described in detail firstly, and then some key factors to set the virtual FPP system much close to the reality are analyzed. With the aim of accurately estimating the depth image from only one fringe image, we also design a new loss function to enhance the quality of the overall and detailed information restored. And two representative networks, U-Net and pix2pix, are compared in multiple aspects. The real experiments prove the good accuracy and generalization of the network trained by the data from our virtual systems and the designed loss, implying the potential of our method for applications. Copyright © 2021, The Authors. All rights reserved.

关键词： computer graphics

AI drives the assessment of lung cancer microenvironment composition

学校读者我要写书评

暂无评论

Journal of Pathology Informatics 2024年 15卷 100400-100400页

作者： Gallo, Enzo Guardiani, Davide Betti, Martina Arteni, Brindusa Ana Maria Di Martino, Simona Baldinelli, Sara Daralioti, Theodora Merenda, Elisabetta Ascione, Andrea Visca, Paolo Pescarmona, Edoardo Lavitrano, Marialuisa Nisticò, Paola Ciliberto, Gennaro Pallocca, Matteo Department of Pathology IRCCS Regina Elena National Cancer Institute Rome Italy Biostatistics Bioinformatics and Clinical Trial Center IRCCS Regina Elena National Cancer Institute Rome Italy Department of Computer Control and Management Engineering La Sapienza University of Rome Rome Italy UOC Anatomy Pathology Biobank IRCCS Regina Elena National Cancer Institute Istituti Fisioterapici Ospitalieri IFO Rome Italy Department of Radiological Oncological and Pathological Sciences Sapienza University of Rome Policlinico Umberto I Rome Italy Department of Experimental Medicine Sapienza University of Rome Policlinico Umberto I Rome Italy School of Medicine and Surgery University of Milano-Bicocca Monza 20900 Italy Tumor Immunology and Immunotherapy Unit IRCCS Regina Elena National Cancer Institute Rome Italy Scientific Direction IRCCS Regina Elena National Cancer Institute Rome Italy Institute of Experimental Endocrinology and Oncology National Research Council Naples Italy

Purpose: The abundance and distribution of tumor-infiltrating lymphocytes (TILs) as well as that of other components of the tumor microenvironment is of particular importance for predicting response to immunotherapy in lung cancer (LC). We describe here a pilot study employing artificial intelligence (AI) in the assessment of TILs and other cell populations, intending to reduce the inter- or intra-observer variability that commonly characterizes this evaluation. Design: We developed a machine learning-based classifier to detect tumor, immune, and stromal cells on hematoxylin and eosin-stained sections, using the open-source framework QuPath. We evaluated the quantity of the aforementioned three cell populations among 37 LC whole slide images regions of interest, comparing the assessments made by five pathologists, both before and after using graphical predictions made by AI, for a total of 1110 quantitative measurements. Results: Our findings indicate noteworthy variations in score distribution among pathologists and between individual pathologists and AI. The AI-guided pathologist's evaluations resulted in reduction of significant discrepancies across pathologists: three comparisons showed a loss of significance (p > 0.05), whereas other four showed a reduction in significance (p > 0.01). Conclusions: We show that employing a machine learning approach in cell population quantification reduces inter- and intra-observer variability, improving reproducibility and facilitating its use in further validation studies. © 2024 The Authors

关键词： computer-aided tool Digital pathology Lung cancer Machine learning NSCLC Pathology image QuPath Tumor-infiltrating lymphocytes Whole slide images

Mobile robot navigation for complete coverage of an environment

学校读者我要写书评

暂无评论

Mobile robot navigation for complete coverage of an environm...

作者： Šelek, Ana Seder, Marija Petrović, Ivan Department of Control and Computer Engineering Faculty of Electrical Engineering and Computing University of Zagreb Croatia

Inspired by the Spanning Tree Covering (STC) algorithm of Gabriely and Rimon, a novel algorithm of complete coverage for known environments is developed. Unlike the original STC algorithm, we detect dynamic elements of the environment and efficiently update the solution when changes are observed. The contribution is a path replanning algorithm that reduces overlapping when the part of the environment is changed. Experiments show that the proposed algorithm is capable of planning complete coverage robot paths with 70.27% coverage and without overlapping. © 2018

关键词： Mobile robots

Can information technology affects people’s behavior and opinion? A way to identify it.

学校读者我要写书评

暂无评论

IFAC-PapersOnLine 2019年第25期52卷 498-503页

作者： Georgios Mandellos Dimitris Tsipianitis Dimitrios Lymberopoulos Telecommunications & Information Technology Division Electrical & Computer Engineering Department University of Patras Rion GR-26504 Greece Systems & Automatic Control Division Electrical & Computer Engineering Department University of Patras Rion GR-26504 Greece

The scope of this study is to determine if the information technology can affects people’s behavior and or opinion. A way to identify it is using a tool to collect answers using or not using the technology. How? Making surveys. Nowadays there are 2 main categories of survey administration. The traditional paper data collection method and the relatively new web-based method. Although the two acquisition processes are completely different, each one has its own advantages and disadvantages, facility or difficulty, adoption or not. The scope of this paper is to study factors that affect each survey type and find the preferable questionnaire administration method. To compare survey media, responses from University students were taken. We tried to investigate the reliability, honesty, item mean, response rate, completeness of data, and also made factor analysis comparisons across survey media. The experiment was applied on the same audience and theme, using different methods. From the data analyses the results between paper and online media in this experiment appear to be major, and seem to have a significant effect on overall results. The majority of differences found on questions related with personal criticism (where anonymity plays a significant role), and on response rates.

关键词： information technology people’s opinion surveys people’s behavior

Bio-inspired polarization event camera

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Haessig, Germain Joubert, Damien Haque, Justin Chen, Yingkai Milde, Moritz B. Delbruck, Tobi Gruev, Viktor AIT Austrian Institute of Technology Center for Vision Automation & Control High-performance Vision Systems Vienna Austria Institute of Neuroinformatics University of Zurich ETH Zurich Switzerland International Centre for Neuromorphic Systems The MARCS Institute Western Sydney University Sydney Australia Department of Electrical and Computer Engineering University of Illinois at Urbana-Champaign UrbanaIL United States

The stomatopod (mantis shrimp) visual system0 has recently provided a blueprint for the design of paradigm-shifting polarization and multispectral imaging sensors0, enabling solutions to challenging medical0 and remote sensing problems0. However, these bioinspired sensors lack the high dynamic range (HDR) and asynchronous polarization vision capabilities of the stomatopod visual system, limiting temporal resolution to ∼12 ms and dynamic range to ∼72 dB. Here we present a novel stomatopod-inspired polarization camera which mimics the sustained and transient biological visual pathways to save power and sample data beyond the maximum Nyquist frame rate. This bio-inspired sensor simultaneously captures both synchronous intensity frames and asynchronous polarization brightness change information with sub-millisecond latencies over a million-fold range of illumination. Our PDAVIS camera is comprised of 346x260 pixels, organized in 2-by-2 macropixels, which filter the incoming light with four linear polarization filters offset by 45°. Polarization information is reconstructed using both low cost and latency event-based algorithms and more accurate but slower deep neural networks. Our sensor is used to image HDR polarization scenes which vary at high speeds and to observe dynamical properties of single collagen fibers in bovine tendon under rapid cyclical loads. © 2021, CC BY.

关键词： Cameras

Diabatic Quantum Annealing for the Frustrated Ring Model

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Côté, Jeremy Sauvage, Frédéric Larocca, Martín Jonsson, Matías Cincio, Lukasz Albash, Tameem Theoretical Division Los Alamos National Laboratory Los AlamosNM87545 United States Institut Quantique Département de Physique Université de Sherbrooke SherbrookeQCJ1K 2R1 Canada Center for Nonlinear Studies Los Alamos National Laboratory Los AlamosNM87545 United States Department of Physics Carnegie Mellon University PittsburghPA15217 United States Department of Electrical and Computer Engineering University of New Mexico AlbuquerqueNM87131 United States Department of Physics and Astronomy Center for Quantum Information and Control University of New Mexico AlbuquerqueNM87131 United States

Quantum annealing is a continuous-time heuristic quantum algorithm for solving or approximately solving classical optimization problems. The algorithm uses a schedule to interpolate between a driver Hamiltonian with an easy-to-prepare ground state and a problem Hamiltonian whose ground state encodes solutions to an optimization problem. The standard implementation relies on the evolution being adiabatic: keeping the system in the instantaneous ground state with high probability and requiring a time scale inversely related to the minimum energy gap between the instantaneous ground and excited states. However, adiabatic evolution can lead to evolution times that scale exponentially with the system size, even for computationally simple problems. Here, we study whether non-adiabatic evolutions with optimized annealing schedules can bypass this exponential slowdown for one such class of problems called the frustrated ring model. For sufficiently optimized annealing schedules and system sizes of up to 39 qubits, we provide numerical evidence that we can avoid the exponential slowdown. Our work highlights the potential of highly-controllable quantum annealing to circumvent bottlenecks associated with the standard implementation of quantum annealing. Copyright © 2022, The Authors. All rights reserved.

关键词： Ground state

Foundation models and intelligent decision-making: Progress, challenges, and perspectives

学校读者我要写书评

暂无评论

Innovation 2025年第6期6卷 100948页

作者： Huang, Jincai Xu, Yongjun Wang, Qi Wang, Qi [Cheems] Liang, Xingxing Wang, Fei Zhang, Zhao Wei, Wei Zhang, Boxuan Huang, Libo Chang, Jingru Ma, Liantao Ma, Ting Liang, Yuxuan Zhang, Jie Guo, Jian Jiang, Xuhui Fan, Xinxin An, Zhulin Li, Tingting Li, Xuefei Shao, Zezhi Qian, Tangwen Sun, Tao Diao, Boyu Yang, Chuanguang Yu, Chenqing Wu, Yiqing Li, Mengxian Zhang, Haifeng Zeng, Yongcheng Zhang, Zhicheng Zhu, Zhengqiu Lv, Yiqin Li, Aming Chen, Xu An, Bo Xiao, Wei Bai, Chenguang Mao, Yuxing Yin, Zhigang Gui, Sheng Su, Wentao Zhu, Yinghao Gao, Junyi He, Xinyu Li, Yizhou Jin, Guangyin Ao, Xiang Zhai, Xuehao Tan, Haoran Yun, Lijun Shi, Hongquan Li, Jun Fan, Changjun Huang, Kuihua Harrison, Ewen Leung, Victor C.M. Qiu, Sihang Dong, Yanjie Zheng, Xiaolong Wang, Gang Zheng, Yu Wang, Yuanzhuo Guo, Jiafeng Wang, Lizhe Cheng, Xueqi Wang, Yaonan Yang, Shanlin Fu, Mengyin Fei, Aiguo Institute of Computing Technology Chinese Academy of Sciences Beijing100190 China Department of Automation Tsinghua University Beijing100084 China Huazhong University of Science and Technology Wuhan430074 China School of Automation Beijing Institute of Technology Beijing100081 China School of Information Science and Engineering Dalian Polytechnic University Dalian116034 China National Engineering Research Center for Software Engineering Peking University Beijing100871 China Department of Oral Implantology Peking University School and Hospital of Stomatology Beijing100081 China Guangzhou511453 China College of Information and Electrical Engineering China Agricultural University Beijing100083 China IDEA Research International Digital Economy Academy Shenzhen518057 China Institute of Automation Chinese Academy of Sciences Beijing100190 China The State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China College of Science National University of Defense Technology Changsha410073 China Center for Systems and Control College of Engineering Peking University Beijing100871 China Gaoling School of Artificial Intelligence Renmin University of China Beijing100872 China College of Systems Engineering National University of Defense Technology Changsha410073 China Laboratory for Big Data and Decision National University of Defense Technology Changsha410073 China College of Computing and Data Science Nanyang Technological University Singapore639798 Singapore School of Electrical Engineering Chongqing University Chongqing400044 China Department of Mathematics The University of Hong Kong Hong Kong SAR999077 China School of Food Science and Technology Dalian Polytechnic University Dalian116034 China Centre for Medical Informatics University of Edinburgh EdinburghEH16 4UX United Kingdom Department of Stomatology Peking Union Medical College Hospital Chinese Academ

Intelligent decision-making (IDM) is a cornerstone of artificial intelligence (AI) designed to automate or augment decision processes. Modern IDM paradigms integrate advanced frameworks to enable intelligent agents to make effective and adaptive choices and decompose complex tasks into manageable steps, such as AI agents and high-level reinforcement learning. Recent advances in multimodal foundation-based approaches unify diverse input modalities—such as vision, language, and sensory data—into a cohesive decision-making process. Foundation models (FMs) have become pivotal in science and industry, transforming decision-making and research capabilities. Their large-scale, multimodal data-processing abilities foster adaptability and interdisciplinary breakthroughs across fields such as healthcare, life sciences, and education. This survey examines IDM's evolution, advanced paradigms with FMs and their transformative impact on decision-making across diverse scientific and industrial domains, highlighting the challenges and opportunities in building efficient, adaptive, and ethical decision systems. © 2025 The Authors

关键词： Reinforcement learning

Existence theorem of finite krasner hyperfields

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Tsafack, Surdive Atamewoue Amassayoga, Ogadoa Onasanya, Babatunde O. Feng, Yuming School of Three Gorges Artificial Intelligence Chongqing Three Gorges University Chongqing404100 China Department of Mathematics Higher Teacher Training College University of Yaounde 1 Cameroon Department of Mathematics Faculty of Science University of Yaounde 1 Cameroon Key Laboratory of Intelligent Information Processing and Control Chongqing Three China Gorges University Chongqing404100 China Department of Mathematics Uni-versity of Ibadan Nigeria School of Computer Science and Engineering Chongqing Three Gorges University China

关键词： Algebra

Robust State Estimations in controlled ARMA Processes with the Non-Gaussian Noises: Applications to the Delayed Dynamics

学校读者我要写书评

暂无评论

IFAC-PapersOnLine 2021年第20期54卷 334-339页

作者： Vadim Azhmyakov Jose Perea Arango Moises Bonilla Raymundo Juarez del Toro Stefan Pickl Institut für Theoretische Informatik Mathematik und Operations Research Universität der Bundeswehr München München Germany National Research University Higher School of Economics MIEM School of Computer Engineering Moscow Russia Department of Basic Sciences Universidad de Medellin Medellin Colombia CINVESTAV-IPN Control Automatico UMI 3575 CINVESTAV-CNRS Mexico UPIIC Instituto Politécnico Nacional San Buenaventura Mexico

This paper deals with a novel theoretic approach to the robust state estimations in discrete-time dynamic systems with the non-Gaussian correlated stochastic noises. The methodology we develop is based on the so-called ”worst case” robust Kalman Filter (KF) approach proposed in [2,3]. We are interested in the robust state estimation for the controlled ARMA models under assumption of the colored noises. Since an ARMA model involves the correlated noises in the equivalent Linear Model (LM) representation, the resulting dynamic system also includes the correlated stochastic variables. These two crucial properties of the ARMA models under consideration imply the impossibility of application of the classic KF-type state estimations. We use the modified ”instrumental variable” method and derive an auxiliary LM with the uncorrelated noises. Application of the robust KF to this auxiliary LM makes it possible to derive a guaranteed state estimation in the initially given ARMA model. The proposed non-standard KF based state estimations are finally applied to the linear stochastic dynamic systems with the time delays.

关键词：