基于LabVIEW图像法车牌智能识别系统

车辆牌照自动识别是实现交通管理智能化的重要环节,设计中利用图像采集卡对经过的车辆车牌进行图像采集并传送至计算机,采用美国NI公司LabVIEW软件,实现图像预处理、图像去噪以及图像增强等功能;然后根据车牌颜色特征对其准确定位,采用阈值法分割车牌字符;最后由OCR函数来识别字符,识别结果保存至相应数据中,可以进行相应的违章、违规智能交通管理,经实验该系统成功实现车牌识别识别率达99%。     

基于实时样本采集的个性化手写汉字输入系统设计

手写汉字识别是手写汉字输入的基础。目前智能设备中的手写汉字输入法无法根据用户的汉字书写习惯,动态调整识别模型以提升手写汉字的正确识别率。通过对最新深度学习算法及训练模型的研究,提出了一种基于用户手写汉字样本实时采集的个性化手写汉字输入系统的设计方法。该方法将采集用户的手写汉字作为增量样本,通过对服务器端训练生成的手写汉字识别模型的再次训练,使识别模型能够更好地适应该用户的书写习惯,提升手写汉字输入系统的识别率。最后,在该理论方法的基础上,结合新设计的深度残差网络,进行了手写汉字识别的对比实验。实验结果显示,通过引入实时采集样本的再次训练,手写汉字识别模型的识别率有较大幅度的提升,能够更有效的满足用户在智能设备端对手写汉字输入系统的使用需求。     

基于Map-Reduce并行构架和AE的大数据指纹监控识别算法

传统的指纹识别监控方法在应用到以海量、在线和增量为特征的大数据环境中时,不能自动高效地进行识别,且识别率和识别精度都不高。由此设计了一种基于Map-Reduce并行框架和AE的大数据智能监控指纹识别算法。首先,描述了Hadoop和Map-Reduce实现并行计算的原理;然后,设计了基于AE和BP神经网络的识别模型,采用AE进行指纹图像自动特征提取,并提出了一种基于比较差异算法对AE进行参数初始化的训练算法,采用BP神经网络进行具体识别;最后,将识别过程分解为Map函数和Reduce函数,并对函数的具体功能进行了详细定义。在Map-Reduce并行框架下对FVC2004中的指纹数据库进行实验,实验结果表明了文中方法能自动、高效地进行指纹识别,与其他方法相比,具有识别效率高和识别精度高的优点,具有较大的优越性。     

未系安全带智能识别中安全带定位关键方法研究

安全带定位是实现机动车未系安全带智能识别的关键。针对道路监控图像特点,提出一种基于梯度变换的安全带定位方法。该方法在对卡口图像进行预处理的基础上,采用自适应阈值边缘检测算法以及积分投影方式定位车辆位置,设立车牌检测区域,以减少运算量,降低干扰,同时利用训练得到的Haar分类器识别车牌位置,通过逼近方式切取车辆右侧图像。最后采用梯度变换算法求得车窗各边缘坐标,实现车窗精准定位,并计算得到安全带位置。试验表明,该方法可实现安全带的准确定位,具有较好的实用性,为后续安全带识别奠定基础。     

基于BP神经网络的车牌字符识别算法研究

车牌字符识别是车牌识别系统中的关键环节。采用图像处理和神经网络相结合的方法设计新的车牌字符识别算法,先对分割出的车牌字符进行归一化处理,然后进行SOBEI．边缘检测和角点特征提取,最后输入BP神经网络进行训练、识别,其中BP神经网络模型属于改进型神经网络。通过一系列神经网络训练和仿真实验,车牌识别速度和正确率得到了明显的提高。     

基于SSIM的车牌识别算法研究

对车牌中的字符进行精确提取并识别是车牌识别系统面临的重要问题。在字符提取阶段,利用车牌图像特有的信息特征,提出以红色分量作为目标图像,依次进行水平和垂直投影完成字符分割,既可精确提取字符,又可矫正字符形态。将基于结构相似度SSIM(structure similarity)的算法应用于车牌识别,避免了前期对字符结构信息做大量的对比统计工作,省略了特征提取的步骤,降低了算法复杂度;利用角点检测方法对相似字符进行细化和区分,进一步完善了整体的识别性能。实验证明,该算法有较好的识别率。     

基于FPGA和DSP的车牌识别系统的研究

摘要：本文介绍了一种以DSP和FPGA为核心的车牌识别系统的设计方案,利用FPGA控制车牌图像数据的采集,经过图像格式转换之后传送到DSP中,进行车牌区域定位、字符分割、车牌识别等操作,最后将识别结果传送到LCD中显示。系统采用CCS（Code Composer Studio）集成开发环境,利用VC 语言编程实现车牌识别算法。通过嵌入式机器视觉库EMCV实现识别算法在DSP中的移植,并在LCD上创建工作窗口显示识别结果。实验结果表明该车牌识别系统的设计方案识别结果准确,功能稳定可靠。     

多光谱分离算法在目标识别中的研究

针对野外工作时机械扫描式的光学系统抗震性差、目标识别率低、实时性差等问题,设计了采用多光谱分离算法实现非扫描目标识别遥感系统。采用非扫描的M-Z干涉具提供空间光程差,由红外CCD采集干涉条纹信息,经CUP处理得到混合光谱,结合可见光视频图像提供的坐标系实现识别目标。其中采用遗传算法优化选择特征波长,然后由粗糙集分类提取未知目标谱的属性,取前1/3可信度的相应属性反演待测目标种类,相比传统算法减少约9倍的运算量。在不同天气、不同背景条件下做实验,得到系统在各种情况下的探测极限及识别概率。由实验数据可知,采用遗传算法和粗糙集分类相结合的多光谱分离算法可以快速、有效地识别未知目标的种类。     

仿选择性注意机制的语音情感识别算法

有效特征的选取一直都是语音情感识别算法的关键。为此,针对语音情感特征选择与构建的问题,一种仿选择性注意机制的语音情感识别算法被提出。考虑到语音信号的时频特性,算法首先计算语音信号的语谱图;其次,模仿选择性注意机制,计算语谱图的颜色、方向和亮度特征图,归一化后形成特征矩阵;然后,将特征矩阵重排列并进行PCA降维,形成情感识别特征向量;最后,利用改进的支持向量机分类方法进行语音情感识别。对愤怒、恐惧、高兴、悲伤和惊奇5种情感的识别实验显示,基于选择性注意的方法能够获得较好的识别效果,平均识别率为85.44%。相比于韵律特征和音质特征,语音情感识别率至少提高10%;相比于其它语谱特征,识别率提高7%左右。     

基于Delaunay网格和DLP的三维指纹识别系统设计

针对目前已有的嵌入式指纹识别系统存在的识别精度底和识别效率低的问题,设计了一种基于单DLP处理器的嵌入式三维指纹识别系统;首先,以TMS320VC5402为处理器核心,FPS200为图像采集芯片,EPM7128为外围逻辑设计了指纹识别系统硬件;然后,设计了基于Delaunay网格的上位机指纹识别软件,将指纹识别过程分为细节点DT网格构造、识别参考点获取、待识别指纹姿势校正和指纹匹配四个阶段进行实现;最后,采用指纹数据库FVC2004为测试数据库在上位机上进行仿真,测试结果表明:文中系统能较为精确地实现指纹识别,平均正确识别率高达98.1%,且与其他方法相比,具有FNMR和FMR值较低、正确识别率高和匹配时间少的优点。     

Self-organization of five species in a cyclic competition game

Cyclic competition game models, particularly the “rock–paper–scissors” model, play important roles in exploring the problem of multi-species coexistence in spatially ecological systems. We propose an extended “rock–paper–scissors” game to model cyclic interactions among five species, and find that two of the five can coexistent when biodiversity disappears, which is different from the “rock–paper–scissors” game. As the number of fingers is five, we named the new model the “fingers” game, where the thumb, forefinger, middle finger, ring finger, and little finger cyclically dominate their subsequent species and are dominated by their former species. We investigate the “fingers” model in two ways: direct simulations and nonlinear partial differential equations. An important finding is that the number of species in a cyclic competition game has an influence on the emergence of biodiversity. To be specific, the “rock–paper–scissors” model is in favor of maintaining biodiversity in comparison with the “fingers” model when the variables (population size, reproduction rate, selection rate, and migration rate) are the same. It is also shown that the mobility and reproduction rate can promote or jeopardize biodiversity.     

Experimental study on downstep in Chinese intonation

Downstep in pitch contour of Chinese Putonghua is examined using subtly designed sentences by controlling tone combinations. The results show both automatic and nonautomatic downstep phenomena exist in Chinese. In non-automatic downstep, low tones compress downwards the pitch range of the following syllables. and the main influence of downstep is on topline. Low tone not only lower the topline behind it, but also raise the high tones before it, the effects are compatible with each other. In automatic downstep, the topline of pitch contour in intonational phrase is presented as a linear downtrend, but it differs among speakers due to the effect of personal stress practice. In comparison with downstep phenomenon in other tone or non-tone languages, the downstep ratio in Chinese is not constant, and the domain of downstep is not limited within the adjacent tones.     

Contrastive analysis on tone between Putonghua and Taiwan Mandarin

The differences of the pitch and duration of Chinese syllables between Putonghua （PTH） and Taiwan Mandarin （TM） were studied. The speech materials to be used are not only isolated syllables, but also sentences. The results reveal that： For the isolated syllables, T1 and T2 in TM are influenced by Minnan dialect, therefore their pitch are lower than those in PTH. T3 is fall-rise in PTH, while it is fall in TM. Moreover, the syllable duration sequence for different tone is T3〉T2〉T1〉T4 in PTH, while it is T1〉T2〉T3〉T4 in TM. For the syllables in sentences, T2 is mid-rise in PTH, while it is mid-level in TM. And the T3 is longer than T4 but shorter than T1 or T2 in PTH, while it is the shortest in TM. Furthermore the effects of prosodic phrase boundary on duration for different tones are almost the same in PTH, but the lengthening part of T1 or T2 is longer than that of T3 or T4 in TM.     

Abelian-Higgs phase of SU(2) QCD and glueball energy

It is shown that SU(2) QCD admits an dual Abelian-Higgs phase, with a Higgs vacuum of a type-Ⅱ superconductor. This is done by using a connection decomposition for the gluon field and the random-direction approximation. Using a bag picture with soft wall, we presented a calculational procedure for the glueball energy based on the recent proof for wall-vortices [Nucl. Phys. B 741(2006)1].     

High efficiency filtering using two-dimensional photonic crystal coupled cavity structures

Transmission spectra of coupled cavity structures (CCSs) in two-dimensional (2D) photonic crystals (PCs) are investigated using a coupled mode theory, and an optical filter based on CCS is proposed. The performance of the filter is investigated using finite-difference time-domain (FDTD) method, and the results show that within a very short coupling distance of about 3λ, where λ is the wavelength of signal in vacuum, the incident signals with different frequencies are separated into different channels with a contrast ratio of 20 dB. The advantages of this kind of filter are small size and easily tunable operation frequencies.     

Signal generation in an isotropic medium in scanning electron acoustic microscope

Based on the research in Ref. [5][Materials Science and Engineering, 1989; A122： 57 63], an improved model of heat source is set up, the different modes of Lamb wave in an isotropic sample generated by a chopped electron beam at frequency f are obtained with integral transform and normal function expansion method, and the output signal of PZT coupled at the back surface of the sample is found out. The generation mechanism of SEAM （Scanning Electron Acoustic Microscopy） signal is discussed. It shows that the SEAM is a near field imaging technique with high spatial resolution and its best lateral spatial resolution is about 2√2α （α is the radius of the focused electron beam）. Some of experimental results of SEAM images are presented in the paper and it shows that the spatial resolution of SEAM is better than 0.5 μm and smaller than the thermal diffusion length of the sample. Therefore the character of near field imaging in SEAM is also proved experimentally.     

Real-time restoration of rotational blurred image using gradient-loading

The key to the restoration of rotational motion blurred image is how to restore the image under a low cost and to correct the irreversibility of the degradation function matrix.Based on the special qualities of degradation function matrix and precise deduction in space-domain, we present a new approach using gradient-loading for restoration of rotational blurred image.By easily adding a gradient operator, the irreversibility of the original matrix is corrected and can be applied for inverse filtering then.Gradientloading is the optimized approach which combines the advantages of both the approaches using constrained least square filtering and traditional diagonal-loading.Compared with the approach using least square filtering, its peak signal-to-noise ratio (PSNR) is improved from 3.18 to 6.46dB, while the computing time is reduced to 1/2-1/3.Experimental results demonstrate the effectiveness, noise-resistibility, robustness, and low complexity of this approach, which make it more suitable for real-time environment.     

All-Optical Logic Device Using Bent Nonlinear Tapered Y-Junction Waveguide Structure

We propose an all-optical logic device made of a bent tapered Y-junction waveguide with a Kerr-type nonlinear interface. It could provide an AND gate, OR gate, and exclusive-OR (XOR) gate. We could obtain different transmission results by adjusting the bending angle. The numerical simulation results show that the device functions as AND, OR, and XOR gates.     

Real-time, continuous-wave terahertz imaging by a pyroelectric camera

Real-time, continuous-wave terahertz (THz) imaging is demonstrated. A 1.89-THz optically-pumped farinfrared laser is used as the illumination source, and a 124 × 124 element room-temperature pyroelectric camera is adopted as the detector. With this setup, THz images through various wrapping materials are shown. The results show that this imaging system has the potential applications in real-time mail and security inspection.