基于Choquet模糊积分的运动目标检测算法

本文提出了一种基于Choquet模糊积分的运动目标检测算法（CIMOD, Choquet Integrate-based Moving Object Detection）。将模糊测度和模糊积分理论应用于运动目标与背景分类中,提出了自适应阈值的Choquet积分算法,实现了图像的颜色特征和纹理特征相融合;选择YCbCr颜色空间代替传统RGB空间,将图像亮度与色度分离,降低了光照变化对运动检测的影响;利用局部二元模式（LBP,Local Binary Pattern）纹理特征对亮度级的单调的变化具有不变性的特点,将其融合到检测算法中,有效抑制了阴影的干扰。仿真实验结果表明,即使在光照变化、阴影干扰等复杂背景情况下,该算法也能够准确的检测出运动区域。     

基于神经网络的时延驱动版图规划算法

本文用离散的网格代替连续的版图规划平面,把长宽比可变的软模块对应成多个长度和宽度均确定的硬模块,给出了相应的时延驱动版图规划问题的形式化描述,并提出了基于均场退火网络的新的求解算法.算法用一个三维二值换位矩阵将问题映射为神经网络,建立包含时延约束、重叠约束和优化目标的能量函数,再用均场退火方程迭代求解.对应于同一软模块的硬模块有且只有一个能且只能放置在版图规划平面一个位置上的约束用神经元归一化的方法解决.本算法已用Visual C++编程实现,实验结果表明,这是一种有效的方法.     

Hierarchical scale-based multiobject recognition of 3-D anatomical structures

Segmentation of anatomical structures from medical images is a challenging problem, which depends on the accurate recognition (localization) of anatomical structures prior to delineation. This study generalizes anatomy segmentation problem via attacking two major challenges: 1) automatically locating anatomical structures without doing search or optimization, and 2) automatically delineating the anatomical structures based on the located model assembly. For 1), we propose intensity weighted ball-scale object extraction concept to build a hierarchical transfer function from image space to object (shape) space such that anatomical structures in 3-D medical images can be recognized without the need to perform search or optimization. For 2), we integrate the graph-cut (GC) segmentation algorithm with prior shape model. This integrated segmentation framework is evaluated on clinical 3-D images consisting of a set of 20 abdominal CT scans. In addition, we use a set of 11 foot MR images to test the generalizability of our method to the different imaging modalities as well as robustness and accuracy of the proposed methodology. Since MR image intensities do not possess a tissue specific numeric meaning, we also explore the effects of intensity nonstandardness on anatomical object recognition. Experimental results indicate that: 1) effective recognition can make the delineation more accurate; 2) incorporating a large number of anatomical structures via a model assembly in the shape model improves the recognition and delineation accuracy dramatically; 3) ball-scale yields useful information about the relationship between the objects and the image; 4) intensity variation among scenes in an ensemble degrades object recognition performance.     

A novel algorithm and architecture for adaptive digital beamforming

A novel algorithm and architecture are described which have specific application to high performance, digital, adaptive beamforming. It is shown how a simple, linearly constrained adaptive combiner forms the basis for a wide range of adaptive antenna subsystems. The function of such an adaptive combiner is formulated as a recursive least squares minimization operation and the corresponding weight vector is obtained by means of theQ-Rdecomposition algorithm using Givens rotations. An efficient pipelined architecture to implement this algorithm is also described. It takes the form of a triangular systolic/wavefront array and has many desirable features for very large scale integration (VLSI) system design.     

联合瞬时姿态估计的可见光成像定位技术

针对可见光成像定位中目标位置估计所涉及的高维空间搜索收敛慢的问题,引入降维设计思想,联合惯导单元(IMU)的瞬时姿态估计,将搜索空间缩小到带约束特征的样本空间,据此提出一种联合目标瞬时姿态估计的高维空间搜索方法。该方法以瞬时姿态信息特征构建目标寻优函数作为约束条件,缩小遗传算法中个体优秀基因的进化范围,显著提高了进化速度,同时规避了对初始样本空间的依赖性,提高了算法的鲁棒性。仿真结果表明,算法的位置估计均方误差优于40 mm。     

基于三维直方图的改进Camshift目标跟踪算法

经典的连续自适应均值漂移算法Camshift通过HSV空间的色调Hue分量建立一维直方图,在有光照变化及有相似颜色目标或背景的干扰下,跟踪效果不好。提出一种融合HSV空间中色调、饱和度以及反应物体形状信息的边缘梯度的三维直方图特征,并基于背景模型自适应调整特征直方图三种分量的权重值,提高了算法的跟踪准确度。通过与传统Camshift跟踪实验比较,提出的改进算法在光照变化及相似颜色目标/背景干扰下具有更好的鲁棒性,同样也满足跟踪系统的实时性要求。     

3D human posture segmentation by spectral clustering with surface normal constraint

In this paper, we propose a new algorithm for partitioning human posture represented by 3D point clouds sampled from the surface of human body. The algorithm is formed as a constrained extension of the recently developed segmentation method, spectral clustering (SC). Two folds of merits are offered by the algorithm: (1) as a nonlinear method, it is able to deal with the situation that data (point cloud) are sampled from a manifold (the surface of human body) rather than the embedded entire 3D space; (2) by using constraints, it facilitates the integration of multiple similarities for human posture partitioning, and it also helps to reduce the limitations of spectral clustering. We show that the constrained spectral clustering (CSC) still can be solved by generalized eigen-decomposition. Experimental results confirm the effectiveness of the proposed algorithm.     

Robust adaptive beamforming for large-scale arrays

For a large-scale adaptive array, heavy computational load and high-rate data transmission are two challenges in the implementation of an adaptive digital beamforming system. Moreover, the large-scale array becomes extremely sensitive to array imperfections. First, based on a restructured recursive linearly constrained minimum variance algorithm and a gradient-based optimization method, a new robust recursive linearly constrained minimum variance (RRLCMV) algorithm is proposed in this paper. The computational load of the RRLCMV algorithm is on the order of o(N), which is less than that of the conventional gradient-based robust adaptive algorithm. Then, a new efficient parallel robust recursive linearly constrained minimum variance (PRRLCMV) adaptive algorithm is proposed by appropriately partitioning the RRLCMV algorithm into a number of operational modules. It can be easily executed in a distributed-parallel-processing fashion, sequentially and in parallel. As a result, the PRRLCMV algorithm provides an effective solution that can alleviate the bottleneck of high-rate data transmission and reduce the computational cost. Finally, an implementation scheme of the PRRLCMV algorithm based on a distributed-parallel-processing system is also proposed. The simulation results demonstrate that the new PRRLCMV algorithm can significantly reduce the degradation due to various array errors.     

Combining static and dynamic features using neural networks and edge fusion for video object extraction

Semantic object representation is an important step for digital multimedia applications such as object-based coding, content-based access and manipulations. The authors propose an image sequence segmentation scheme which provides region information for the semantic object representation of those applications. The objective is to develop a hardware-friendly segmentation algorithm by combining static and dynamic features simultaneously in one scheme. In the initial stage, a multiple feature space is transformed to one-dimensional label space by using self-organising feature map (SOFM) neural networks. The next stage is an edge fusion process in which edge information is incorporated into the neural network outputs to generate more precisely located boundaries of segmentation. The proposed algorithm differs from existing methods as follows: it can segment textured images with low-dimensional features; leads to more meaningful segmentation region boundaries; and is easier to map into hardware than existing methods. Experimental results are compared with an existing segmentation method using evaluation metrics to clarify the advantages of the proposed algorithm objectively.     

多信息融合的红外弱小目标检测

红外弱小目标检测是图像处理的难点之一,许多研究人员提出了不少检测方法.针对复杂背景与强杂波干扰下图像信杂比(Signal-to-Clutter Ratio,SCR)低造成的目前检测方法易受伪目标干扰、虚警率高的问题,提出了一种多信息融合的红外弱小目标检测算法.首先,构建八向局部灰度残差信息图;其次,设计一个滑动窗口遍历整个图像,将图像分为一系列局部图像块,对局部图像块的强度均值进行约束,获得局部强度均值约束信息图;然后,将局部图像块进一步划分为12个方向块,对每个方向块中像素的梯度方向进行约束,获取梯度方向约束信息图;最后,上述3个信息图像通过点积运算得到最终显著图,并利用阈值分割实现弱小目标的分离.将该算法与3种其它不同算法从信杂比增益(Signal-to-Clutter Ratio Gain,SCRG)、背景抑制因子(Background Suppression Factor,BSF)以及检测率与虚警率的接受者操作特征(Receiver Operating Characteristic,ROC)曲线方面进行对比.实验结果表明:该算法具有更高的SCRG、BSF和ROC曲线下面积(Area Under the Curve,AUC),不仅能有效地抑制背景杂波、剔除伪目标,而且能准确地检测出红外弱小目标,具有较高的检测率.     

Linearly constrained block adaptive filtering algorithm withoptimum convergence factors

A new linearly constrained adaptive filtering algorithm, the linearly constrained optimum block adaptive (LCOBA) algorithm, is presented. The LCOBA algorithm processes data in blocks and uses variable convergence factors which are optimised in a least square sense. It is superior to Frost's linearly constrained least mean squares algorithm at achieving the conflicting goals of fast convergence with little steady-state error. In addition, its computational requirements generally tend to be smaller than that of the Frost algorithm, as the block length is increased     

A maximum-likelihood estimator to simultaneously unwrap, geocode, and fuse SAR interferograms from different viewing geometries into one digital elevation model

This paper presents theory, algorithm, and results of a maximum-likelihood algorithm that is capable to fuse a number of heterogeneous synthetic aperture radar interferograms into a single digital elevation model (DEM) without the need for the critical phase-unwrapping step. The fusion process takes place in the object space, i.e., the map geometry, and considers the periodic likelihood function of each individual interferometric phase sample. The interferograms may vary regarding their radar wavelength, their baseline, their heading angle (ascending or descending), and their incidence angle. Geometric baseline error estimates and a priori knowledge from other estimates like existing DEMs are incorporated seamlessly into the estimation process. The presented approach significantly differs from the standard DEM generation method where each interferogram is first phase-unwrapped individually, then geocoded into a common map geometry, and finally averaged with DEMs generated from other interferograms. By avoiding the phase-unwrapping step, the proposed algorithm does not depend on gradients between samples and is therefore capable to reconstruct the arbitrary height of each single scatterer. Because the height of each DEM sample is determined individually, spatial propagation of phase-unwrapping errors is avoided. The algorithm is targeted to fuse an ensemble of interferometric multiangle or multibaseline observations in areas of rugged terrain or highly ambiguous data where algorithms based on phase unwrapping may fail. The algorithm is explained, and examples with real data from the Shuttle Radar Topography Mission are given. Conditions of future missions are simulated, and optimization criteria for the viewing geometry are discussed.     

A modified block FTF adaptive algorithm with applications tounderwater target detection

In this paper, the problem of weighted block recursive least squares (RLS) adaptive filtering is formulated in the context of a block fast transversal filter (FTF) algorithm. This “modified block FTF algorithm” is derived by modifying the constrained block-LS cost function to guarantee global optimality. This new soft-constrained algorithm provides an efficient way of transferring weight information between blocks of data. The tracking ability of the algorithm can be controlled by varying the block length and/or a soft constrained parameter. This algorithm is computationally more efficient compared with other LS-based schemes. The effectiveness of this algorithm is tested on a real-life problem dealing with underwater target identification from acoustic backscatter. The process involves the identification of the presence of resonance in the acoustic backscatter from a target of unknown shape submerged in water     

A computationally efficient frequency-domain LMS algorithm withconstraints on the adaptive filter

The frequency domain implementation of the LMS algorithm is attractive due to both the reduced computational complexity and the potential of faster convergence compared with the time domain implementation. Another advantage is the potential of using frequency-domain constraints on the adaptive filter, such as limiting its magnitude response or limiting the power of its output signal. This paper presents a computationally efficient algorithm that allows the incorporation of various frequency domain constraints into the LMS algorithm. A penalty function formulation is used with a steepest descent search to adapt the filter so that it converges to the new constrained minimum. The formulation of the algorithm is derived first, after which the use of some practical constraints with this algorithm and a simulation example for adaptive blind equalization are described     

Simplified adaptive algorithm for constrained notch filters with guaranteed stability

The stability issue in adaptive notch filters is investigated from a structural viewpoint. A new adaptive algorithm for direct frequency estimation is developed using a constrained pole-zero notch filter. The basic characteristic of this algorithm is that the notch filter is realised in a cascade form, with each stage implemented in a normal realisation. A simplified version of the gradient signals is derived, which makes the algorithm very efficient. It is shown that this proposed adaptive notch filter is bounded-input-bounded-output stable. Numerical examples with simulations, where the proposed algorithm performs well in terms of convergence and guaranteed stability, are given to support the theoretical results     

基于多分辨率格网的三维物体识别方法

本文首先提出了一种改进的三维物体表达方法,它将一个三维物体表面网格与其它表面网格的几何关系表示为一个二维矩阵,称为距离角度图.这种表达能够描述任意形态物体,抑制杂乱背景和遮挡,几何意义直观,且适应不同分辨率、非规则的三角格网.然后,以这种表达方法为基础,本文阐述了一种基于多分辨率格网的,由粗到精的三维物体识别方法.它先在场景和模型的低分辨率格网上进行粗匹配以得到模型候选集合,之后在已匹配网格的高分辨率格网邻域上筛选模型候选集合,最后综合考虑多个网格对应的模型候选以得到最终模型候选的确认和验证.这种识别方法具有运算量小,准确可靠等优点,实验证明该方法正确有效.     

NEMP毁伤效应数据自适应建模技术研究

为了在三维地理信息系统中实现NEMP毁伤效应的可视化,利用空间聚类技术对毁伤数据进行分析,得到毁伤数据的空间聚类检验指数;利用多分辨率建模技术的聚合-解聚方法和自适应算法,对毁伤目标空间进行剖分,得到MRRTP模型;该模型与传统的TEN模型混合,实现了NEMP毁伤效应的三维空间数据模型构建;为高效、完整地实现NEMP毁伤效应的三维表达以及空间分析提供了技术支持。     

Object classification in 3-D images using alpha-trimmed mean radialbasis function network

We propose a pattern classification based approach for simultaneous three-dimensional (3-D) object modeling and segmentation in image volumes. The 3-D objects are described as a set of overlapping ellipsoids. The segmentation relies on the geometrical model and graylevel statistics. The characteristic parameters of the ellipsoids and of the graylevel statistics are embedded in a radial basis function (RBF) network and they are found by means of unsupervised training. A new robust training algorithm for RBF networks based on alpha-trimmed mean statistics is employed in this study. The extension of the Hough transform algorithm in the 3-D space by employing a spherical coordinate system is used for ellipsoidal center estimation. We study the performance of the proposed algorithm and we present results when segmenting a stack of microscopy images.