ID 原文 译文
55827 图像的微分信息包含了图像的细节信息以及像素的相对空间位置信息,提高了模型建立时信息的利用率,能够提高目标模型的建模精度。 Image differential information contains the image information and the details of the pixel relative space position information, increase the efficiency of modeling time information, can improve the modeling precision of target model.
55828 仿真实验结果表明,与传统均值漂移算法相比,改进算法在复杂的背景情况下具有更强的抗干扰性能,能够有效提高目标跟踪的稳定性。 Simulation experiment results show that compared with the traditional mean shift algorithm, the improved algorithm under the condition of complicated background has better anti-jamming performance, can effectively improve the stability of target tracking.
55829 针对网络化飞行器故障检测系统在多工作点间切换情况,考虑时延和丢包导致的滤波器切换时刻滞后于飞行器模型切换时刻问题,提出一种基于异步切换系统的故障检测算法。 For networked aircraft fault detection system to switch between multiple working point, considering the time delay and packet loss caused by filter switching time lags behind the aircraft model switching time problem, put forward a kind of fault detection algorithm based on asynchronous switching system.
55830 将网络环境下大包线内多工作点飞行器建模为异步切换系统,基于李亚普诺夫函数和平均驻留时间法,给出了具有异步切换形式的故障检测系统全局一致渐近稳定的充分条件,进一步利用线性矩阵不等式(linear matrix inequation,LMI)工具求取滤波器参数。 To obtain more aircraft bag line under the network environment modeling for asynchronous switching system, based on the lyapunov function and the average dwell time method, the fault detection system with asynchronous switching form is given the sufficient condition of global uniform asymptotic stability, further using linear matrix inequality (LMI) LMI (linear matrix inequation,) tool to calculate the filter parameters.
55831 数值仿真验证并分析了所提方法的有效性。 Numerical simulation and analysis of the effectiveness of the proposed method.
55832 提出一种复杂场景极化合成孔径雷达(polarimetric synthetic aperture radar,PolSAR)图像机场跑道多级分类检测方法。 In this paper, a complex scene polarization synthetic aperture radar (polarimetric synthetic aperture radar, PolSAR) image airport runway multi-stage classification method.
55833 首先利用先验信息进行h/q第一级分类,得到图像中各类训练样本模板;然后利用PolSAR图像极化相干矩阵的统计特性进行第二级分类;再根据跑道的弱回波特性,利用极化总功率检测器完成第三级分类,提取图像中疑似机场跑道区域;最后根据机场跑道的尺寸及结构特征进行判别,确定机场跑道区域。 First use of a priori information classification is the first step on the h/q, get the image in all kinds of training sample template;Then by using the statistical properties of PolSAR image polarization coherent matrix secondary classification;Again according to the properties of weak echo the runway, completed the third grade classification, using polarization total power detector to extract the image suspected airport runway area;According to the size and structure characteristics of the airport runway the discriminant, airport runway area is determined. In U. S.
55834 采用美国UAVSAR机载系统获取的多组实测数据对算法进行验证,并与现有的两种方法进行比较,结果表明,本文算法能有效地检测出跑道,并保持跑道结构完整、轮廓清晰且虚警率低。 UAVSAR airborne system for multiple sets of experimental data to verify this algorithm, and comparing with the existing two methods, the results show that this algorithm can effectively detect the runway, and keep the track structure, complete and with clear outline and low false alarm rate.
55835 提出一种基于多Agent粒子群优化支持向量回归机(support vector regression,SVR)参数的优化算法,并利用该算法建立多步预测控制模型,对非线性系统进行预测控制。 Put forward a kind of based on Agent particle swarm optimization support vector regression machine (support vector regression, the SVR) parameter optimization algorithm, and the multi-step predictive control model was established based on the algorithm, the predictive control of nonlinear system is presented.
55836 通过预测控制的机理推导出满足滚动优化目标函数的多步预测输出的控制律。 Through the mechanism of predictive control meet the rolling optimization target function is deduced multi-step prediction output control law.