Application of multi-layer perceptron neural networks to vision problems

This paper discusses the application of a class of feed-forward Artificial Neural Networks (ANNs) known as Multi-Layer Perceptrons(MLPs) to two vision problems: recognition and pose estimation of 3D objects from a single 2D perspective view; and handwritten digit recognition. In both cases, a multi-MLP classification scheme is developed that combines the decisions of several classifiers. These classifiers operate on the same feature set for the 3D recognition problem whereas different feature types are used for the handwritten digit recognition. The backpropagationlearning rule is used to train the MLPs. Application of the MLP architecture to other vision problems is also briefly discussed.     

An efficient and robust integration technique for applied random vibration analysis

The mode-based finite element formulation of the equations of motion is usually adopted for linear random vibration analysis (RVA). In general, the RVA of large systems requires a large number of numerical integrations which is very time-consuming for a reasonable level of desired accuracy. Moreover, conventional numerical integration methods may fail to converge when the integrands are highly oscillatory due to slow propagation velocities. In this paper, a robust general-purpose RVA integration technique which can overcome these drawbacks is presented. Multi-point base and nodal excitations including wave passage effect and frequency-independent spatial correlation can be taken into account in the analysis. The proposed technique is based on the closed-form solutions for polynomial-type power spectral density functions and has been verified to be efficient and accurate for many engineering problems. This paper describes the implementation details, presents two examples taken from engineering applications and demonstrates the dramatic time-saving in the computation compared to numerical integration solutions. Response statistics, such as standard deviation of structural responses, are computed and displayed over the entire structures for these examples.     

基于两级内容表征的数字图像内容保护算法

文中首先通过动态加权平均得到图像的象素级内容表征(Pixel-level Image Content Map，PICM)，在此基础上，通过图像分割技术得到图像的区域级内容表征(Regional Image Content Map，RICM)，并将表征图像RICM的Hu不变矩编码为内容7F-gF45息嵌入到原图像中，从而实现对原始图像的内容保护。实验结果表明，由于该两级内容表征方法及其矩特征的良好性能，算法对图像的非内容级图像操作(特别是仿射变换)具有很强的鲁棒性，且能够敏感的检测出对所保护图像的内容篡改。     

灰度图像矩的一种快速算法

大多数矩的快速算法仅适用于二值图像。本文给出了一种适用于灰度图像的矩快速算法。对N×N大小的灰度图像 ,引入 6个大小为N的数组 ,将求矩过程中重复计算的量一次计算后存入数组 ,减少了重复和冗余计算。虽然增加了 6个数组 ,内存的相对增加仅为 6/N。该算法原理简单 ,求矩结果精确 ,计算量大大减少。和直接法相比 ,加法和乘法运算次数仅为其 40 4%和2 2 6%。和Yang s算法相比 ,加法次数相同 ,而乘法次数约为其 67 6%。     

基于循环平稳的复调频信号检测性能研究

针对加性色噪声和乘性色噪声联合干扰下的雷达运动目标检测问题进行了研究.首先利用循环平稳信号处理原理,分析了加性噪声和乘性噪声联合干扰下的复调频信号的循环统计特性,在此基础上构造了基于二阶循环矩的检测统计量,推导了检测统计量的输出信噪比与观测噪声统计特性的关系.最后利用Monte-Carlo仿真试验得到检测性能的曲线,并与基于时频分布的Wigner-Hough检测器做了比较,仿真试验进一步验证了本文提出的二阶循环矩检测器性能要优于Wigner-Hough检测器.     

Design Strategies for Piled Rafts Subjected to Nonuniform Vertical Loading

The piled raft is a geotechnical composite construction, consisting of the three elements piles, raft, and soil, which is applied for the foundation of tall buildings in an increasing number. In a parametric study, 259 different piled raft configurations have been analyzed by means of three-dimensional elastoplastic finite element analyses. In the study, the pile positions, the pile number, the pile length, and the raft-soil stiffness ratio as well as the load distribution on the raft has been varied. In the scope of this paper, the results of the parametric study are presented and design strategies for an optimized design of piled rafts subjected to nonuniform vertical loading are discussed.     

Exact solutions for vibration and buckling of an SS-C-SS-C rectangular plate loaded by linearly varying in-plane stresses

Exact solutions are presented for the free vibration and buckling of rectangular plates having two opposite edges (x=0 and a) simply supported and the other two (y=0 and b) clamped, with the simply supported edges subjected to a linearly varying normal stress σ_x=−N₀[1−α(y/b)]/h, where h is the plate thickness. By assuming the transverse displacement (w) to vary as sin(mπx/a), the governing partial differential equation of motion is reduced to an ordinary differential equation in y with variable coefficients, for which an exact solution is obtained as a power series (the method of Frobenius). Applying the clamped boundary conditions at y=0 and b yields the frequency determinant. Buckling loads arise as the frequencies approach zero. A careful study of the convergence of the power series is made. Buckling loads are determined for loading parameters α=0,0.5,1,1.5,2, for which α=2 is a pure in-plane bending moment. Comparisons are made with published buckling loads for α=0,1,2 obtained by the method of integration of the differential equation (α=0) or the method of energy (α=1,2). Novel results are presented for the free vibration frequencies of rectangular plates with aspect ratios a/b=0.5,1,2 subjected to three types of loadings (α=0,1,2), with load intensities N₀/N_cr=0,0.5,0.8,0.95,1, where N_cr is the critical buckling load of the plate. Contour plots of buckling and free vibration mode shapes are also shown.     

Shot boundary detection using Zernike moments in multi-GPU multi-CPU architectures

This paper presents an analysis of a Multi-GPU Multi-CPU environment, along with the different possible hybrid combinations. The analysis has been performed for a shot boundary detection application, based on Zernike moments, although it is general enough to be applied to many different application areas. A deep study of the performance, bottlenecks and design challenges is carried out showing the validity of this approach and achieving very high frame per second rates. In this paper, Zernike calculations are carried out on GPUs, taking advantage of a packing strategy proposed to minimize host-device communication time.     

Algorithmic estimation of risk factors in financial markets with stochastic drift

We assume a financial market governed by a diffusion process reverting to a stochastic mean which is itself governed by an unobservable ergodic diffusion, similar to those observed in electricity and other energy markets. We develop a moment method algorithm for the estimation of the parameters of both the observable process and the unobservable stochastic mean. Our approach is contrasted with other methods for parameter estimation of partially observed diffusions, and applications to the modelling of interest rates and commodity prices are discussed.     

Visual-based quadrotor control by means of fuzzy cognitive maps

By applying an image-based visual servoing (IBVS) method, the intelligent image-based controlling of a quadrotor type unmanned aerial vehicle (UAV) tracking a moving target is studied in this paper. A fuzzy cognitive map (FCM) is a soft computing method which is classified as a fuzzy neural system and exploits the main aspects of fuzzy logic and neural network systems; so it seems to be a suitable choice for implementing a vision-based intelligent technique. An FCM has been employed in implementing an IBVS scheme on a quadrotor UAV, so that the UAV can track a moving target on the ground. For this purpose, by properly combining the perspective image moments, some features with the desired characteristics for controlling the translational and yaw motions of a UAV have been presented. In designing a vision-based control method for a UAV quadrotor, there are some challenges, including the target mobility and not knowing the height of UAV above the target. Also, no sensor has been installed on the moving object and the changes of its yaw angle are not available. Despite all the stated challenges, the proposed method, which uses an FCM in controlling the translational motion and the yaw rotation of a UAV, adequately enables the quadrotor to follow the moving target. The simulation results for different paths show the satisfactory performance of the designed controller.