基于多节点拓扑重叠测度高阶MRF模型的图像分割

针对低阶马尔科夫随机场(Markov random field, MRF)模型难以有效表达自然图像中复杂的先验知识而造成误分割问题, 提出一种基于多节点拓扑重叠测度高阶MRF模型(Higher-order MRF model with multi-node topological overlap measure, MTOM-HMRF)的图像分割方法. 首先, 为描述图像局部区域内多像素蕴含的复杂空间拓扑结构信息, 利用多节点拓扑重叠测度建立图像局部区域的高阶先验模型; 其次, 利用较大的局部区域包含更多的标签节点信息能力, 基于Pairwise MRF模型建立基于局部区域的部分二阶Potts先验模型, 提高分割模型的抗噪能力; 再次, 为有效描述观察图像场与其标签场的似然特征分布, 研究利用局部区域内邻接像素的Hamming距离引入图像局部空间相关性, 建立局部空间一致性约束的高斯混合分布; 最后, 基于MRF框架建立用于图像分割的多节点拓扑重叠测度高阶MRF模型, 采用Gibbs采样算法对提出模型进行优化. 实验结果表明, 提出模型不仅能有效抵抗图像强噪声和复杂的纹理突变干扰, 鲁棒性更好, 而且具有更准确的图像分割结果.     

A novel call admission control scheme in cellular/WLAN integration and performance analysis

In order to achieve the Quality of Service (QoS) provisioning and efficient resource utilization in cellular network and Wireless Local Area Network (WLAN) integration, an Integrated Service-Based Call Admission Control (ISB-CAC) scheme is proposed in this paper. The integrated network is modeled by using multi-dimensional Markov chains. The numerical analysis is presented to evaluate the important performance measures such as the blocking probability of originating calls, the dropping probability, and the average transfer time, etc. The steady-state probabilities of the multi-dimensional Markov chains are obtained by using an iterative approach, and the CAC parameters are optimally designed. The analytical model is validated by the computer simulation. It is shown that compared with the conventional WLAN-First Call Admission Control (WF-CAC) scheme, the proposed ISB-CAC scheme not only provides better QoS for mobile users but also utilizes the bandwidth resources more efficiently.     

A Brief Overview of ChatGPT: The History,Status Quo and Potential Future Development

ChatGPT, an artificial intelligence generated content (AIGC) model developed by OpenAI, has attracted worldwide attention for its capability of dealing with challenging language understanding and generation tasks in the form of conversations. This paper briefly provides an overview on the history, status quo and potential future development of ChatGPT, helping to provide an entry point to think about ChatGPT. Specifically, from the limited open-accessed resources, we conclude the core techniques of ChatGPT, mainly including large-scale language models, in-context learning, reinforcement learning from human feedback and the key technical steps for developing ChatGPT. We further analyze the pros and cons of ChatGPT and we rethink the duality of ChatGPT in various fields. Although it has been widely acknowledged that ChatGPT brings plenty of opportunities for various fields, mankind should still treat and use ChatGPT properly to avoid the potential threat, e.g., academic integrity and safety challenge. Finally, we discuss several open problems as the potential development of ChatGPT.      

隐Markov模型的多算法交叉耦合自适应算法

以离子通道信号重构为例,扩展HMM为矢量隐Markov模型,利用随机逼近原理对期望最大算法进行自适应改造,估计离子通道的动力学特征参数;递归辅助变量算法估计背景噪声的统计特征;卡尔曼滤波预测背景噪声;三种算法交叉耦合重构离子通道信号。该算法能够克服滤波器和背景噪声的影响,在低信噪比情况下得到了较高精度的估计参数和重构信号,具有鲁棒性和一致收敛特性。     

基于改进变量预测模型的变压器故障诊断

针对变压器故障诊断中的小样本、非线性、参数寻优难等问题,提出改进的变量预测模型的变压器故障诊断方法.分析变量预测模型和布谷鸟搜索算法结合解决小样本和非线性问题,指出其后期收敛速度慢,稳定性差,收敛精度不高,易陷入局部极小值问题,在此基础上在谷鸟搜索算法位置更新中引入变异操作,提高解的多样性.引入动态步长和动态发现概率提...     

盈余过程服从跳扩模型下的破产概率

为了更真实地反映市场随机变化趋势,将经典的扩散模型推广到跳跃-扩散模型.用布朗运动和复合泊松过程共同驱动保险公司的盈余过程,并考虑盈余过程的系数受马尔可夫链干扰的情况.采用鞅方法和微积分方法研究保险公司的破产概率,得到破产概率所满足的偏微分方程.     

A centralized tool for managing,archiving, and serving point-in-time data in ecological research laboratories

The recent proliferation of software tools that aid researchers in various phases of data tracking and analysis undoubtedly contribute to successful development of increasingly complex and data-intensive scientific investigations. However, the lack of fully integrated solutions to data acquisition and storage, quality assurance/control, visualization, and provenance tracking of heterogeneous temporal data streams collected at numerous geospatial locations continues to occupy a general problem area for scientists and data managers working in the environmental sciences. We present a new Service Oriented Architecture (SOA) that allows users to: 1) automate the process of pushing real-time data streams from networks of environmental sensors or other data sources to an electronic data archive; 2) to perform basic data management and quality control tasks; and 3) to publish any subset of the data to existing cyberinfrastructure platforms for global discovery and distribution via the World Wide Web. The approach outlined here supports management of: 1) repeated field observations, 2) data from laboratory analysis of field samples, 3) simulation results, and 4) derived values. We describe how the use of Hypertext Transfer Protocol (HTTP) Application Programming Interfaces (APIs) Representational State Transfer (REST) methods for data model objects and Resource Query Language (RQL) interfaces respond to a basic problem area in environmental modelling by enabling researchers to integrate an electronic data repository with existing workflows, simulation models, or third-party software.     

Learning hierarchical spatio-temporal pattern for human activity prediction

Human activity prediction has become increasingly valuable in many applications. This paper, initially from the perspective of cognition science, presents a novel approach to learning a hierarchical spatio-temporal pattern of human activities to predict ongoing activities from videos that contain only the onsets of the activities. Spatio-temporal pattern can be learned by a Hierarchical Self-Organizing Map (HSOM), which consists of two self-organizing maps (i.e., action map and actionlet map) connected via associative links trained by Hebbian learning. Ongoing activities can be predicted by Variable order Markov Model (VMM), which provides the means for capturing both large and small order Markov dependencies based on the training actionlet sequences. Experiments of the proposed method on four challenging 3D action datasets captured by commodity depth cameras show promising results.     

Reliability and maintainability analysis to improve the operation of the limoncello production line

The reliability, maintainability, failure rate and repair rate models of a limoncello automated production line for a period of 8 months at machine and entire line level were carried out. It was pointed out that (i) the real operating time of the limoncello production line (LPL) is 89.35%, whereas for the rest 10.65%, the LPL is under repair, (ii) the Weibull distribution provided the best fit of the failure data for the LPL to describe the time‐between‐failures (TBFs), whereas the time‐to‐repairs (TTRs) are lognormal distributed, and (iii) the failure rate of the production line increases, implying that the current maintenance strategy is not adequate and must be improved. This particular methodology can also be utilised in the beverage industry sector by the machinery manufacturers and the manufacturers of bottled products to improve the design and operation management of the bottling production line.