基于统计正交性分析的彩色人脸识别

提出了一个用于彩色人脸识别的特征提取方法。该方法根据Fisher准则按顺序分别计算彩色人脸图像集中红(R)、绿(G)、蓝(B)三个彩色分量集的投影变换,同时使获得的三个投影变换互相统计正交。该方法能够加强三个彩色分量鉴别特征之间的互补性,移除相关性。AR和第二版的FRGC两个彩色人脸图像数据库上的实验结果表明,该方法的识别效果超越相关的几个彩色人脸特征提取方法。

彩色图像统计正交分析方法

彩色图像含有比灰度图像更丰富的信息,因此在图像识别中扮演重要的角色。RGB彩色空间是使用最为广泛的彩色空间。通常R、G、B三分量间存在相关性。彩色图像识别技术的关键是如何有效使用分量间的补信息、消除冗余,并且提取有效的鉴别特征。文中提出了一种新的彩色图像特征提取方法,即彩色图像统计正交分析(CISOA)。该方法按照R、G、B的顺序依次提取三分量的鉴别特征,并保证各分量所提取的特征满足统计正交约束。在彩色人脸和掌纹图像数据库的实验结果表明此方法具有较好的识别效果。

基于散度差的彩色人脸图像统计正交分析方法

由于彩色人脸图像比灰度人脸图像包含了更多的信息,彩色人脸图像识别方法越来越受到学者的重视。而对于研究最多的RGB彩色空间,通常R(红)、G(绿)、B(蓝)三分量间存在很大的相关性。为了最大程度去除各个分量之间的相关性从而提高识别效果,有学者提出了基于统计正交投影变换(SOA)的彩色人脸图像识别方法。然而,该方法在特征提取的过程中不可避免地存在奇异性问题。为了解决这个问题,文中提出了一种基于散度差的彩色人脸图像统计正交分析方法(SDFSOA)。此外,对所涉及的参数进行了合理的设置。实验结果表明所提方法能取得更好的识别效果。

岩溶管道与裂隙交叉渗流特性试验研究

研制了光滑岩溶管道和裂隙交叉的水力特性室内试验模型。采用正交试验设计方法安排试验,测得了48组岩溶管道和裂隙不同交叉、不同水力情况下的渗流水力特性数据。采用正交统计分析法分析绘制试验因子关系曲线图,详细研究裂隙宽度、岩溶管道直径及进水水头对试验成果的影响。定量分析了岩溶管道与裂隙交叉时其折算渗透系数与长度、直径等因素之间的关系,并在量纲分析的基础上,采用变结构遗传算法建立了计算岩溶管道和裂隙交叉时管道汇流量的数学表达式。最后将计算值与试验值对比,结果表明,所建立的公式能较好地反映影响管道汇流量各因素之间的内在联系。

迭代无味卡尔曼滤波的目标跟踪算法

针对目标跟踪迭代无味卡尔曼滤波(IUKF)算法中跟踪精度较差的问题,提出一种基于状态扩展技术的改进迭代无味卡尔曼滤波(IIUKF)算法.新算法首先将观测噪声扩展进状态,构造关于扩展状态的零噪声观测方程,然后在观测迭代过程中将最新的扩展状态后验估计代入更新公式,进行观测迭代更新.相比IUKF算法,IIUKF算法不仅形式上更为简洁,而且避免了IUKF算法中先验估计和观测噪声非统计正交的问题,滤波精度更高.数值仿真表明,IIUKF算法的跟踪误差比IUKF算法减小了20%以上.

栽培因素对烤烟经济性状的效应分析

利用正交设计与统计分析研究了栽培因素对烤烟经济性状的效应 ,结果表明 :在一定范围内 ,施氮量对烤烟产量的影响是主要因素 ,留叶数是次重要因素 ,种植密度是次要因素 ,其中施氮量效应大于留叶数效应 ;施氮量对烤烟产值的影响是主要因素 ,种植密度是次重要因素 ,留叶数是次要因素 。

基于片上坐标旋转数字计算机的正交验证模型

为了提高坐标旋转数字计算机的验证效率,提出了一种新的验证模型,该模型采用正态统计原理来选取测试因素建立正交表,从而根据正交表结果选取最佳测试方法.此外,该模型能自动产生测试向量并验证坐标旋转数字计算机的精度误差.仿真实验结果表明,该验证模型能大大减少验证次数,缩短验证周期,提高验证效率.

Orthogonal experiment design of EMI of security monitoring system in coal mines

Security monitoring system of coal mines is indispensable to ensure the safe and efficient production of colliery. Due to the special and narrow underground field of the coal mine, the electromagnetic interference can cause a series of misstatements and false positives on the monitoring system, which will severely hamper the safe production of coal industry. In this paper, first, the frequency characteristics of the interference source on the power line are extracted when equipment runs normally. Then the finite difference time domain method is introduced to analyze the effects of the electromagnetic interference parameters on the security monitoring signal line. And the interference voltage of the two terminal sides on the single line is taken as evaluating indexes. Finally, the electromagnetic interference parameters are optimized by orthogonal experimental design based on the MATLAB simulation on the normal operation of equipment.

Design and Implementation of DSP—Based IEEE802.16a OFDM System

A set of integrated Orthogonal Frequency Division Multiplexing(OFDM) modulation and demodulation system is proposed and its performance is verified in reference to the standard of IEEE802.16a for WMAN.The design,combined with analogy front-end,adopts Texas Instrument(TI) Company's 6701 evaluation module(EVM) to construct digital Intermediate Frequency(IF) platform.A DSP-based baseband processing software module is embedded into the system.The proposed scheme makes it possible for users to select various software modules to set up connections with different throughput performance.Taking into account the given hardware configuration,the design aims to optimize the algorithm modules to accomplish better system performance.

A Novel Two-step Frequency Offset Estimator for OFDM

Based on an orthogonal frequency division multiplexing(OFDM) training symbol with L identical parts, a novel carrier frequency offset (CFO) estimator is proposed for OFDM systems. The CFO is estimated in two steps, fine estimate and coarse estimate. In the first step, the fine estimation is performed based on the principle of minimum variance. However, the fine estimation has ambiguity since its estimate range is limited. In the second step, the coarse estimation is obtained, which results in a larger estimate range but less precision. Using the coarse estimation, the ambiguity of fine estimation is resolved. To fully use the correlation among L identical parts, the fine estimation resolved the ambiguity and the coarse estimation are optimally combined to obtain the final estimation. Furthermore, the estimation variance of the proposed method is derived. Simulation results demonstrate that the novel two-step estimator outperforms the conventional two-step estimator in terms of estimate performance and computational complexity.

Independent Component Analysis Based Blind Adaptive Interference Reduction and Symbol Recovery for OFDM Systems

To overcome the inter-carrier interference （ICI） of orthogonal frequency division multiplexing （OFDM） systems subject to unknown carrier frequency offset （CFO） and multipath, this paper develops a blind adaptive interference suppression scheme based on independent component analysis （ICA）. Taking into account statistical independence of subcarriers＇ signals of OFDM, the signal recovery mechanism is investigated to achieve the goal of blind equalization. The received OFDM signals can be considered as the mixed observation signals. The effect of CFO and multipath corresponds to the mixing matrix in the problem of blind source separation （BSS） framework. In this paper, the ICA- based OFDM system model is built, and the proposed ICA-based detector is exploited to extract source signals from the observation of a received mixture based on the assumption of statistical independence between the sources. The blind separation technique can increase spectral efficiency and provide robustness performance against erroneous parameter estimation problem. Theoretical analysis and simulation results show that compared with the conventional pilot-based scheme, the improved performance of OFDM systems is obtained by the proposed ICA-based detection technique.

A novel transmit scheme for OFDM systems with virtual subcarriers

Based on the double repetition structure in the frequency domain and the proper pulse shaping filter in the time domain, a novel transmit scheme for orthogonal frequency division multiplexing （OFDM） systems with virtual subcarriers is proposed. Our proposed scheme with comb-type pilots can provide feasibility to reduce the channel estimation errors at the band edges with conventional inverse fast Fourier transform （IFFT）-based channel estimator. Also this transmit scheme can provide additional diversity compared to the conventional OFDM scheme. Meanwhile, the receiver schemes corresponding to the proposed transmit scheme are presented to obtain the advantages mentioned above. All these efficacies are confirmed by simulations.

Maximum Likelihood Estimation of Clock Synchronization Error in OFDM System

The performance degradation of an orthogonal frequency division multiplexing (OFDM) systems due to clock synchronization error is analyzed and a pilot-aided maximum likelihood (ML) estimating method is proposed to correct it. The proposed algorithm enables clock synchronization error estimation from a pilot whose duration is only two symbol periods. The study shows that this method is simple and exact. The clock synchronization error can be corrected almost entirely.

A novel improved SLM scheme of the PAPR reduction technology in CO-OFDM systems

The higher peak-to-average power ratio(PAPR) is a major shortcoming of coherent optical orthogonal frequency division multiplexing(CO-OFDM) systems. Selective mapping(SLM) technology can effectively reduce the probability of high PAPR, but it has higher computational complexity, and requires additional bandwidth to transmit the side information, which will affect the transmission efficiency of the system. In response to these shortcomings, a novel improved SLM(NI-SLM) scheme with low complexity and without side information is proposed. Simulation results show that the proposed scheme can exponentially reduce the computational complexity, and the bit error rate(BER) performance can greatly approach the original signal. What's more, it shows the better PAPR reduction performance.