分区适应截值模型及其在交通标志识别中的应用

针对Ｈｏｐｆｉｅｌｄ网络模型在存储模式不满足０和１状态的均匀分布及数目对等的条件下存储容量及寻址能力下降的缺点，提出并用光束方向编码光学实现了三值（１，０，－１）互连的分区适应截值模型，并把这一模型应用到交通标志的识别中，结果表明该模型及光学系统有很好的稳定性。     

基于自适应径向基神经网络的类星体光谱自动识别方法

通过对光谱的研究来识别和认证类星体是天文学研究中的重要方法。文章提出了一种对类星体光谱进行自动识别的自适应径向基神经网络(RBFN)方法。该方法包括以下几个步骤: (1)先将训练样本归一化,再利用PCA变换进行降维,获得样本特征向量; (2)设计出K均值聚类算法与梯度下降法相结合的径向基神经网络结构的基本模型,再用SSE(sum of squares error)误差函数进行判断,对RBFN隐含层的神经元进行自动调节,直至满足给定误差阈值; (3)用训练得到的参数对用于测试的样本中的类星体光谱进行识别。该方法不但克服了经典RBFN算法选择隐层神经元数目的困难,而且还提高了对类星体识别的稳定性和正确率。研究结果对于大型光谱巡天所产生的海量数据的自动处理具有重要意义。     

一种自适应的Gabor图像特征抽取和权重选择的人脸识别方法

 为了克服光照、表情变化等因素对人脸识别的影响,本文提出了一种自适应的Gabor图像特征抽取和权重选择的人脸识别方法.该方法首先把每幅人脸图像经过Gabor小波变换后得到的40个不同尺度和方向下的图像都看作是独立的样本,再把不同人脸中的同一尺度和方向的变换结果进行特征重组,得到40个独立地新特征矩阵.为了增强对光照、表情变化的鲁棒性,每一新特征矩阵的识别贡献被本文所提出的自适应权重方法计算得到.其次,对每一新特征矩阵采用离散余弦变化进行降维,并采用了鉴别力量分析方法来选取最有鉴别力的离散余弦变换系数作为特征向量.最后,抽取线性鉴别分析特征进行识别.大量的实验证明了本文所提方法的有效性.     

Classification and recognition of transgenic product by terahertz spectroscopy and DSVM

The purpose of this paper is to construct a classification model that can identify class accurately and control imbalance. A novel adaptive decision support vector machine (DSVM) is proposed for the recognition of transgenic cotton seed based on terahertz spectroscopy (THz), which make the traditional support vector machine is ability of adaptive decision, and select optimal parameters by using particle swarm optimization (PSO). For the classification and recognition of the transgenic cotton seeds, firstly, the factor analysis (FA) is applied to reduce the dimension and extract the feature spectrum of original spectral information. Secondly, the feature spectrum is selected and fed into the model of DSVM to recognize the different transgenic cotton seeds. The experimental results show that the proposed method can effectively classify the different transgenic cotton seeds, and its recognition rate surpasses the comparative method evidently.     

基于指纹识别与LIN总线的汽车车门系统设计

为了改善传统汽车防盗系统安全性低的问题,提出了一种新型的汽车门门锁控制方法,基于LIN总线网络对车身控制和改进的指纹识别技术来实现。使用L9638作为LIN总线收发器,以ST792F150JDV1QC作为主控节点微控制器,实现LIN网络的通信功能和对网关的控制。针对提高车门指纹识别率问题,提出了Gabor滤波器指纹识别方法,使用八通道的Gabor滤波器对预处理的指纹图像进行滤波,提取指纹图像的指纹特征,使用欧氏距离方法对提取的指纹特征进行匹配。匹配结果经过LIN从节点控制器串行口接收,传递至总线,主节点控制器做出反应,实现门锁的开关。最终对系统实物进行搭建,进行实验测量系统的可行性。     

基于自适应无参经验小波变换和选择集成分类模型的运动想象

运动想象模式识别率的提高对脑机接口(BCI)技术的应用具有重要意义,本文采用自适应无参经验小波变换(APEWT)和选择集成分类模型相结合的方法提高脑电(EEG)信号的分类识别准确率.首先,通过APEWT将EEG信号分解成不同的模态;然后,使用最优模态重构后的信号计算其能量谱(ES)特征,使用最优模态分量计算其边际谱(MS)特征;最后,将不同时间段的ES特征和不同频段的MS特征输入到构建的选择集成分类模型中,从而得到其分类结果,并将该方法与其他4种组合方法进行比较.实验结果表明,本文方法具有较好分类准确率和实时性,其平均分类正确率高于其他4种方法,同时较近期使用相同数据的文献也有优势.本文为在线运动想象类BCI的应用提供了新的方法和思路.     

基于EMD-AWPP和HOSA-SVM算法的分布式光纤振动入侵信号的特征提取与识别

针对传统的信号处理方法无法有效区分不同振动入侵信号,提出一种基于EMD-AWPP和HOSA-SVM算法的振动信息特征提取与识别方法,用于解决分布式光纤振动入侵检测系统的高精度信号识别问题。处理不同振动类型时,该方法首先利用基于经验模态分解的自适应小波包处理算法,不仅对信号的低频部分进行了分解,而且对高频部分即信号的细节部分也进行了更好的时频局部化处理,改善了信号特征提取精度,减少传感信号异常值的影响; 其次采用高阶谱分析中的双谱和双相干谱,精确提取包含不同振动入侵信号类型的特征矢量; 最后在BPNN参比模型的基础上,用粒子群算法优化SVM的识别参数,使识别模型具有更强的自适应和自学习能力,克服了神经网络易陷入局部最优的不足之处,实现不同振动入侵信号的特征矢量识别。分析结果表明,针对不同类型的入侵源识别,该方法可以有效剔除随机噪声的影响,提取传感信息的特征矢量,降低异常值的影响,算法的预测类别与输出类别几乎一致,振动识别的精确率达到95%以上,识别效果明显强于BPNN网络的检测算法,提高了信息分析的准确性。     

Quantum Pattern Recognition with Probability of 100%

In recent years there has been an increasing focus on the quantum pattern recognition, especially quantum multi-pattern recognition in computer science. This paper presents a new quantum multi-pattern recognition method based on the improved Grover’s algorithm. This method not only details the process of quantum multi-pattern recognition using several unitary operators, but also introduces a new design scheme of initializing quantum state and quantum encoding on the pattern set. If the rate of the number of the recognized pattern on the total patterns is over 1/3, this new method can recognize multi-pattern simultaneously with the probability of 100%. Mathematic calculations and simulation results on the case show that the proposed method can accomplish multi-pattern recognition with the probability of 100%. However, the recognition probability of other pattern recognition methods is impossible to reach 1.     

Region-based SIFT approach to iris recognition

Traditional iris recognition systems transfer iris images to polar (or log-polar) coordinates and have performed very well on data that tends to have a centered gaze. The patterns of an iris are part of a 3-D structure that is captured as a two-dimensional (2-D) image and cooperative iris recognition systems are capable of correctly matching these 2-D representations of iris features. However, when the gaze of an eye changes with respect to the camera lens, many times the size, shape, and detail of iris patterns will change as well and cannot be matched to enrolled images using traditional methods. Additionally, the transformation of off-angle eyes to polar coordinates becomes much more challenging and noncooperative iris algorithms will require a different approach. The direct application of the scale-invariant feature transform (SIFT) method would not work well for iris recognition because it does not take advantage of the characteristics of iris patterns. We propose the region-based SIFT approach to iris recognition. This new method does not require polar transformation, affine transformation or highly accurate segmentation to perform iris recognition and is scale invariant. This method was tested on the iris challenge evaluation (ICE), WVU and IUPUI noncooperative databases and results show that the method is capable of cooperative and noncooperative iris recognition.     

EEG-Based Emotion Recognition by Exploiting Fused Network Entropy Measures of Complex Networks across Subjects

It is well known that there may be significant individual differences in physiological signal patterns for emotional responses. Emotion recognition based on electroencephalogram (EEG) signals is still a challenging task in the context of developing an individual-independent recognition method. In our paper, from the perspective of spatial topology and temporal information of brain emotional patterns in an EEG, we exploit complex networks to characterize EEG signals to effectively extract EEG information for emotion recognition. First, we exploit visibility graphs to construct complex networks from EEG signals. Then, two kinds of network entropy measures (nodal degree entropy and clustering coefficient entropy) are calculated. By applying the AUC method, the effective features are input into the SVM classifier to perform emotion recognition across subjects. The experiment results showed that, for the EEG signals of 62 channels, the features of 18 channels selected by AUC were significant (p < 0.005). For the classification of positive and negative emotions, the average recognition rate was 87.26%; for the classification of positive, negative, and neutral emotions, the average recognition rate was 68.44%. Our method improves mean accuracy by an average of 2.28% compared with other existing methods. Our results fully demonstrate that a more accurate recognition of emotional EEG signals can be achieved relative to the available relevant studies, indicating that our method can provide more generalizability in practical use.     

Control of a coupled map lattice model for vortex shedding in the wake of a cylinder

The flow behind a vibrating flexible cable at low Reynolds numbers can exhibit complex wake structures such as lace-like patterns, vortex dislocations and frequency cells. These structures have been observed in experiments and numerical simulations, and are predicted by a previously developed low-order coupled map lattice (CML). The discrete (in time and space) CML models consist of a series of diffusively coupled circle map oscillators along the cable span. Motivated by a desire to modify the complex wake patterns behind flexible vibrating cables we have studied the addition of control terms into the highly efficient CML models and explored the resulting dynamics. Proportional, adaptive proportional and discontinuous non-linear (DNL) control methods were used to derive the control laws. The first method employed occasional proportional feedback. The adaptive method used spatio-temporal feedback control. The DNL method used a discontinuous feedback linearization procedure, and the controller was designed for the resulting linearized system using eigenvalue assignment. These techniques were applied to a modeled vortex dislocation structure in the wake of a vibrating cable in uniform freestream flow. Parallel shedding patterns were achieved for a range of forcing frequency-forcing amplitude combinations studied to validate the control theory. The adaptive proportional and DNL methods were found to be more effective than the proportional control method due to the incorporation of a spatially varying feedback gain across the cylinder span. The DNL method was found to be the most efficient controller of the low-order CML model. The required control level across the cable span was correlated to the 1/1 lock-on behavior of the temporal circle map.     

可编程光学击中击不中变换及其在模式识别中的应用

提出了一种利用击中击不中变换实现具有边缘噪声的模式的形态识别方法并给出计算模拟结果。利用一个非相干光学相关器，构成了一个实时可编程光学击中击不变处理器。演示了字符识别过程并给出了实验结果。     

基于DMD的光谱分辨率自适应Hadamard变换成像光谱仪

数字微镜器件(DMD)可以灵活高效地实现各阶Hadamard编码模板,设计了一种基于DMD的光谱分辨率自适应的Hadamard变换成像光谱仪,分析了仪器的结构和工作原理及实现光谱分辨率自适应的方法。它可以根据目标景物和观测要求的不同而自适应地调整光谱成像的光谱分辨率,从而在光谱准确度和数据计算量之间做出合理的折衷,既满足了对目标分类识别的要求,又提高了数据传输和处理的速度。     

复杂背景下钢索图像的纹理分割与边界识别

 针对复杂背景下钢索图像难以准确分割的问题,提出一种基于纹理分析的钢索图像分割与边界识别方法.采用基于模糊Hough变换的纹理方向检测方法确定钢索走向,利用边缘方向密度直方图作为纹理特征,对与钢索纹理方向相应的边缘方向赋予不同权重,抑制纹理分割中背景的干扰,对钢丝绳图像进行聚类分割,采用检测平行直线的方法确定其边界,并根据算法参量对边界进行修正.在实验中,对比了边缘方向密度直方图特征与灰度共生矩阵、局部二值模式在钢索图像纹理分割中的结果与计算时间,结果表明边缘方向密度直方图特征计算速度快、受背景干扰小,分割准确率高.本文方法无须预先训练,受背景干扰小,可以准确地识别出钢索并确定其边界,能满足钢丝绳视觉检测的要求.     

Automated inspection of micro-defect recognition system for color filter

This study focused on micro-defect recognition and classification in color filters. First, six types of defects were examined, namely grain, black matrix hole (BMH), indium tin oxide (ITO) defect, missing edge and shape (MES), highlights, and particle. Orthogonal projection was applied to locate each pixel in a test image. Then, an image comparison was performed to mark similar blocks on the test image. The block that best resembled the template was chosen as the new template (or matching adaptive template). Afterwards, image subtraction was applied to subtract the pixels at the same location in each block of the test image from the matching adaptive template. The control limit law employed logic operation to separate the defect from the background region. The complete defect structure was obtained by the morphology method. Next, feature values, including defect gray value, red, green, and blue (RGB) color components, and aspect ratio were obtained as the classifier input. The experimental results showed that defect recognition could be completed as fast as 0.154 s using the proposed recognition system and software. In micro-defect classification, back-propagation neural network (BPNN) and minimum distance classifier (MDC) served as the defect classification decision theories for the five acquired feature values. To validate the proposed system, this study used 41 defects as training samples, and treated the feature values of 307 test samples as the BPNN classifier inputs. The total recognition rate was 93.7%. When an MDC was used, the total recognition rate was 96.8%, indicating that the MDC method is feasible in applying automatic optical inspection technology to classify micro-defects of color filters. The proposed system is proven to successfully improve the production yield and lower costs.     

用光学方法实现么正变换及一般线性变换(Ⅳ)——图形识别及投影算子

本文指出,图形识别实质上是图形空间中的表象变换(坐标系变换)。由于表象变换是通过么正变换实现的,用光学方法进行图形识别是比较合适的。本文着重讨论了如何只根据所需识别的标准图形组,构造光学系统,即用光学方法直接实现投影变换的方法。并进一步发展了文献[3]中的优化方法。 关键词：     

An optimisation design of adaptive illumination for a multi-reflective 3D scene

An illumination optimisation technique applied to multi-reflective 3-D machine vision based on a projector-camera system is introduced, in which the projector plays a key role to compensate for surface reflectance at each pixel to be inversely proportional to the brightness of the pixel under ambient light. The adaptive illumination technology was achieved by iterations emphasising different illumination intensities according to different surface orientations and requiring an accurate correspondence between the projector pixels and the camera pixels. In order to establish the most effective correspondence to prepare for subsequent adaptive illumination, 4 kinds of grating patterns, including sinusoidal, rectangular, triangular, and dual-frequency sinusoidal grating patterns, were projected and compared. The iterations were halted when an optimally lit scene was obtained; the further experiments under weak and strong light searched for the best method of illumination optimisation and confirmed the reliability of the adaptive illumination. The proposed optimisation design could run in real time and became a viable solution for industry.     

气液两相流波动信号的时频谱分析研究

为了研究气液两相流不同流型的动态特性,通过小波变换、希尔伯特-黄变换及自适应最优核三种时频方法对气液两相流动态差压信号进行处理.通过对时频谱的分析,可以清晰看出当流型从泡状流向弹状流、塞状流的转化过程中,信号的主要能量由15—35 Hz之间的频带向0—8 Hz频带转移,在弹状流时出现了两个谱峰.实验结果表明:希尔伯特-黄变换及自适应最优核方法的时频分辨率比小波分析高.基于自适应最优核方法的脊信息的提取,克服了模糊平面加窗效应的影响,对气液两相流动态信号表现出更高的时频分辨率,并增强了时频平面信息的可读性. 关键词： 气液两相流 流型识别 希尔伯特-黄变换 自适应最优核     

Recognition of multibreak patterns by 8-neighborhood-based General Hough Transform

In this paper, an 8-neighborhood-based Generalized Hough Transform (ENGHT) for the recognition of multibreak patterns is proposed, which is different from the Generalized Hough Transform (GHT). The difference between ENGHT and GHT appears in the process for Reference-table (R-table) configuration and for pattern recognition. The theoretic and experimental results show that ENGHT can be employed in the recognition of multibreak patterns with a high precision and a high recognition speed, and reduce the difficulty during the recognition that arose from the broken patterns to be processed by the conventional GHT. ENGHT can be employed in pattern recognition, especially in the multibreak pattern recognition.     

自适应分段时域质心特征在鱼类识别中的应用

提出了一种基于时域质心的时域自适应分段方法。该方法以时域质心为依据对信号的时域进行划分,在划分的各个子段内计算时域质心,并将其作为下一层划分的分割点。各个子段内的时域质心反映了信号的能量分布特性,可作为识别特征量。对三种常见的不同形状的鱼类进行了水池试验,提取自适应分段时域质心特征,并使用BP神经网络分类器成功进行了分类。结果表明:利用自适应分段时域质心特征可对不同形状的鱼类进行识别,且具有较高的识别率。