Manifold structure preservative for hyperspectral target detection

A nonparametric method termed as manifold structure preservative (MSP) is proposed in this paper for hyperspectral target detection. MSP transforms the feature space of data to maximize the separation between target and background signals. Moreover, it minimizes the reconstruction error of targets and preserves the topological structure of data in the projected feature space. MSP does not need to consider any distribution for target and background data. So, it can achieve accurate results in real scenarios due to avoiding unreliable assumptions. The proposed MSP detector is compared to several popular detectors and the experiments on a synthetic data and two real hyperspectral images indicate the superior ability of it in target detection.     

A robust object-based shadow detection method for cloud-free high resolution satellite images over urban areas and water bodies

Unwanted contrast in high resolution satellite images such as shadow areas directly affects the result of further processing in urban remote sensing images. Detecting and finding the precise position of shadows is critical in different remote sensing processing chains such as change detection, image classification and digital elevation model generation from stereo images. The spectral similarity between shadow areas, water bodies, and some dark asphalt roads makes the development of robust shadow detection algorithms challenging. In addition, most of the existing methods work on pixel-level and neglect the contextual information contained in neighboring pixels. In this paper, a new object-based shadow detection framework is introduced. In the proposed method a pixel-level shadow mask is built by extending established thresholding methods with a new C₄ index which enables to solve the ambiguity of shadow and water bodies. Then the pixel-based results are further processed in an object-based majority analysis to detect the final shadow objects. Four different high resolution satellite images are used to validate this new approach. The result shows the superiority of the proposed method over some state-of-the-art shadow detection method with an average of 96% in F-measure.     

在二元柔壁自适应风洞中利用面元法进行三元模型实验

简要叙述了在西北工业大学二元柔壁自适应风洞中利用面元法进行二元半机翼模型试验的研究情况，分析了面元法的基本思想和部分试验结果。     

质量矩导弹的神经网络及自适应非线性控制

本文对质量矩控制导弹的动力学方程简化和控制系统的设计进行了研究。详细阐述了所提出的控制方法，介绍了径向基单隐层神经网络进行误差补偿的算法原理，用Lyapunov理论证明了所采用方法的稳定性。在给出了质量矩导弹的六自由度动力学方程和简化后的方程后，用所提出的方法设计控制系统。通过仿真分析，得出了本文所提出的方法具有较好的误差补偿能力的结论。     

故障检测与分离的广义一致性算子法

本文提出了故障检测与分离的广义一致性算子法的一般形式。得出了两个故障可分离的充要条件和多个故障可分离的充要条件。最后给出了检测和分离故障的过程。     

一种在线神经网络在补偿飞机模型不确定性误差中的应用

讨论了一种基于神经网络动态逆的直接自适应控制方法，并应用于超机动飞机的飞行控制中。基本控制律采用非线性动态逆方法进行设计，对由于模型不准确导致的逆误差采用单隐层神经网络进行在线补偿。仿真结果表明，神经网络通过补偿由于模型不准确引起的逆误差，弥补了非线性动态逆要求精确数学模型的缺点，提高了整个控制系统的鲁棒性，而且可以大大简化动态逆控制律的设计。     

神经网络在未来超分辨雷达中的应用

最新的技术发展已为研制超分辨雷达创造了条件。它能够突破普通雷达基于匹配滤波原理的分辨能力,而实现超分辨。超分辨雷达的巨大计算量通常来自求解非线性最小二乘问题时的多维搜索。本文针对这一问题,研究了雷达信号处理中的两个典型例子:(1)利用阵列处理检测个数已知而方向未知的雷达目标;(2)对波音727飞机进行超分辨距离多普勒成像。说明Hop-field神经网络通过集体运算能够求解各种困难的最优化问题,因而在未来的超分辨雷达中有广阔的应用前景。     

一种适于动态环境的盲多用户检测算法

比较两种盲多用户检测算法在动态环境中的收敛性能,并通过计算机进行模拟仿真。其中一种是能随输入信号矢量变化而变化的变步长最小输出能量盲算法;另一种是既保持最小输出能量检测器的全局收敛性,又具有最小均方误差检测器高输出信干比优点的判决反馈变步长盲算法。利用前者在动态强干扰下收敛快及后者在稳态时输出信干比高的优势,将二者结合起来应用到动态环境中,提高系统性能,同时抑制多址干扰和窄带干扰。     

自适应控制方法的直接命中目标本体的平行接近制导律

本文在最少量测条件下采用视线角速度ｑ趋于零的指标函数和低阶等效辨识模型与快速自适应控制方法实现了战术导弹的平行接近制导律。这种最优制导律可以实现直接命中目标本体，实现一弹多用，可以用于反导弹的控制。能使制导装置简化并使其尺寸和重量减小，算法简单易于工程使用。经过大量数学和半实物仿真试验表明，上述成果可应用于实际工程当中。     

自适应天线互耦影响研究

用等效网络法推导了考虑互耦影响时阵列的输出信号解析式和信噪比,仿真了不同阵元间距时,自适应直线阵的方向图和输出信噪比,得到了一些有用的结论。