New pattern recognition system in the e-nose for Chinese spirit identification

This paper presents a new pattern recognition system for Chinese spirit identification by using the polymer quartz piezoelectric crystal sensor based e-nose. The sensors are designed based on quartz crystal microbalance(QCM) principle,and they could capture different vibration frequency signal values for Chinese spirit identification. For each sensor in an8-channel sensor array, seven characteristic values of the original vibration frequency signal values, i.e., average value(A),root-mean-square value(RMS), shape factor value(S_f), crest factor value(C_f), impulse factor value(I_f), clearance factor value(CL_f), kurtosis factor value(K_v) are first extracted. Then the dimension of the characteristic values is reduced by the principle components analysis(PCA) method. Finally the back propagation(BP) neutral network algorithm is used to recognize Chinese spirits. The experimental results show that the recognition rate of six kinds of Chinese spirits is 93.33% and our proposed new pattern recognition system can identify Chinese spirits effectively.     

基于深度残差网络的脱机手写汉字识别研究

摘 要: 手写汉字识别是模式识别与机器学习的重要研究方向和应用领域。近年来,随着深度学习理论方法的完善、新技术的层出不穷,深度神经网络在图像识别分类、图像生成等典型应用中取得了突破性的进展,其中,深度残差网络作为最新的研究成果,已成功应用于手写数字识别、图片识别分类等多个领域。本文将研究深度残差网络在脱机孤立手写汉字识别中的应用方法,通过改进残差学习模块的单元结构,优化深度残差网络性能,同时通过对训练集的预处理,从数据层面实现训练生成模型性能的提升,最后设计实验,验证深度残差网络、End-to-End模式在脱机手写汉字识别中的可行性,分析、总结存在的问题及今后的研究方向。     

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。     

基于菌紫质膜联合子波变换的实时模式识别

菌紫质是一种结构和功能与高级动物视网的视生色素－－视紫红质极为相似的蛋白南，是一种优异的可逆光全息记录材料，子波变换是目标特征抽取和模式识别和有效方法。本文钭联合子波变换和菌紫质膜优异的光学全息记录特性结合起来，提出了一种全新的光学全息实时模式识别方案，给出了相应的实验结果。     

Color pattern recognition based on the joint fractional Fourier transform correlator

A new system of multi-channel single-output joint fractional Fourier transform correlator (JFRTC) for color pattern recognition is proposed based on the conventional system of multi-channel single-output joint transform correlator (JTC). The theoretical analysis and optical experiments are performed. With this method, one can obtain three correlation peaks at the output plane which show a pair of desired cross-correlation peaks and one auto-correlation peak. In comparison, the conventional system leads to more correlation peaks playing a noise role in color pattern recognition.     

一种光谱识别的新方法

提出了一种对光谱信号识别的新方法。针对光谱信号的特征 ,我们设计了基于径向基函数神经网络组成的统计混合模型 ,并构造了识别系统的代价函数。通过优化系统的代价函数 ,导出了类EM算法去估计混合模型的参数 ,从而构建对光谱特征识别的识别器。利用实际的拉曼光谱 ,对本文所提出的估计模型参数的算法与建立的光谱识别器进行了检验。我们还讨论了利用特征波长与相应的光谱强度 ,以及利用主分量分析组成输入特征矢量 ,及其这些输入特征矢量对光谱识别器应用的效果。实验结果表明 ,所提出的算法可以有效地估计模型参数 ,其建立的光谱识别模型具有较高的识别准确率。所提出的对光谱信号识别的方法通用性强 ,因此具有较为广阔的应用前景。     

汉语耳语音孤立字识别研究

耳语音识别有着广泛的应用前景,是一个全新的课题.但是由于耳语音本身的特点,如声级低、没有基频等,给耳语音识别研究带来了困难.本文根据耳语音信号发音模型,结合耳语音的声学特性,建立了一个汉语耳语音孤立字识别系统.由于耳语音信噪比低,必须对其进行语音增强处理,同时在识别系统中应用声调信息提高了识别性能.实验结果说明了MFCC结合幅值包络可作为汉语耳语音自动识别的特征参数,在小字库内用HMM模型识别得出的识别率为90.4%.     

一类新混沌系统的线性状态反馈控制

研究了一类新混沌系统的控制问题.利用Routh-Hurwitz准则对受控系统进行了稳定性分析，结合线性状态反馈方法理论上严格证明了达到控制目标反馈系数的选择原则.数值研究证明了该方法能够有效地控制混沌系统到失稳的平衡点或周期解，同时控制效果在弱噪声干扰下具有很强的鲁棒性. 关键词： 新混沌系统 线性状态反馈控制 Routh-Hurwitz准则 噪声     

A new data processing and calibration method for an eye-tracking device pronunciation system

In this paper, a new data processing and calibration method for a pronunciation system of an eye-tracking device is described. The eye-tracking device was created using both head mounted display (HMD) technology and remote operation capabilities. A pattern recognition computer program was used to distinguish the pupil position and calculate its coordinates.This system can be adapted to provide a digital speech function. A new method for processing the image of the eye in the PC-based system was also developed. With one video CCD camera and frame grabber analyzing a series of human pupil images while the subject is gazing at the screen, an auto-calibration algorithm is used to obtain the direction of the eye gaze in real time. The computers provide the speech sound according to the location where the eye gazes exceed 0.5 s. The availability of multipurpose in this eye-tracking system with very simple equipment will be reconfirmed for future advanced research.     

空间平移不变和角度受限下转动不变的特征识别与滤波器的设计

徐原能等人曾提出借助于circular harmonic expansion(圆谐波展开),设计空间平移不变和转动不变的滤波器系统,用于特征识别。本文在他们的基础上提出一种实现空间平移不变和在一定的角度范围内转动不变的特征识别方法,给出相应的滤波器设计具体公式。虽然转动不变的角度范围受到限制,但能够提高系统的信噪比,同时,不需要通过测量靶与待识别图象之间的相关函数的相位因子,来判定转动角度,而是代之直接测量相关函数的绝对值,由此推算转动角度。本文还列举一个具体的例子,来演示滤波器的设计方法,并与徐原能方案 关键词：     

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.     

基于BP—LVQ神经网络的纸币新旧识别算法研究

针对纸币图像的特点,在一个系统同时提取两组表示纸币新旧的图像特征,采用级联的组合方式,设计了一种新的BP—LVQ神经网络组合分类器对纸币进行新旧识别。经过实验对比验证,该方法有效提高15％的系统识别率,可应用于清分系统的纸币新旧识别。     

A flexible new three-dimensional measurement technique by projected fringe pattern

Aimed at the problems of inferior precision and bad maneuverability for three-dimensional (3D) measurement by projected fringe pattern, a flexible new 3D technique for performing system calibration and measuring was proposed. First, we analyzed the principle of conventional 3D measurement with projected fringe pattern, and pointed out the shortcoming of measurement system. Then, the CCD camera calibration technique is analyzed and we set up the perspective projection model which transforms the computer image coordinate to 3D world coordinate, and we get the coordinate of the CCD camera image lens. Third, the position of projection lens optical center can be obtained using the above model. At last, some experiment results presented show that this technique is more simple and robust in engineering than conventional measurement method.     

基于移动窗口相关系数光谱法的白酒品牌快速判别技术研究

为满足白酒品牌判别领域快速检测的迫切需求,提出了一种以紫外光谱为输入,移动窗口相关系数法为手段的白酒品牌快速判别新方法。研究以海之蓝酒作为对象品牌,实验结果显示各批次海之蓝酒在实验波长区间存在较高的吻合一致性,基于紫外光谱原始谱图的相关系数值和基于移动窗口光谱计算所得的相关系数值均达到0.99以上。由于其他洋河系列白酒与海之蓝酒均属于洋河酒厂生产,原料来源、生产制备工艺等存在较高的相似性,此时若仅使用紫外光谱原始谱图数据结合相似度算法进行品牌判别,则定量化的判别区分存在一定困难。移动窗口法可有效提高谱图细节差异,突出光谱在特征谱段的形状、强度、变化趋势差异,提高实验谱图在细节结构差异辨认分析能力。经过研究筛选,发现以移动窗口相关系数结合紫外光谱特征谱段(270～295 nm)可提高洋河系列酒间的谱图差异,实现这种有较高相似性白酒品牌的判别区分。此外,市售六种其他品牌白酒及乙醇作为对照组,进一步论证了该方案的可行性。该研究的创新之处在于提出了一种移动窗口相关系数光谱法进行白酒品牌快速判别分析的新思路,该方法同时具备速度快,操作简便等优点,具有较高的实际应用潜在价值,并可为其他食品的质量安全判别分析研究提供借鉴。     

Hamilton系统Mei对称性的一种新守恒量

研究Hamilton系统的Mei对称性直接导致的一种新守恒量. 给出Hamilton系统的Mei对称性的定义和判据方程，引入谐调函数，得到系统Mei对称性直接导致新守恒量的条件和形式，并给出应用算例. 结果表明， 谐调函数可根据寻找规范函数的需要适当选取, 从而使规范函数的寻求变得比较容易，而且由于谐调函数的选取具有多样性, 因此能够找到系统Mei对称性的更多的守恒量. 关键词： Hamilton系统 Mei对称性 新守恒量     

A new method for optical pattern recognition by correlation measurements

The method of computing correlation functions using a coherent optical analogue computer is well known. In this letter we describe a new method for the recognition of binary patterns. Usually the peak of maximum intensity in the correlation plane is taken as a “measure of similarity”. Instead of this “measure of similarity” we introduce a “measure of distance”. By the method proposed here the pattern to be recognized is detected by a zero in the intensity distribution in the correlation plane.     

A new type of conserved quantity of Mei symmetry for Lagrange system

This paper studies a new type of conserved quantity which is directly induced by Mei symmetry of the Lagrange system. Firstly, the definition and criterion of Mei symmetry for the Lagrange system are given. Secondly, a coordination function is introduced, and the conditions of existence of the new conserved quantity as well as its forms are proposed. Lastly, an illustrated example is given. The result indicates that the coordination function can be selected properly according to the demand for finding the gauge function, and thereby the gauge function can be found more easily. Furthermore, since the choice of the coordination function has multiformity, many more conserved quantities of Mei symmetry for the Lagrange system can be obtained.     

A new image processing method for evaluating the pupillary responses in a HMD-type eye-tracking device

In this paper, a new image processing method for evaluating the pupillary responses is described. This system is based on an eye-tracking system, using eye tracking and pattern recognition techniques with appropriate hardware and software. This pupillary response evaluation system includes a head-mounted display and a pinhole CCD camera. When the subject uses the head-mounted display to browse the screen of a computer, the CCD camera catches images of the subject's eye and transmits it to the computer. The program in the computer calculates the location and size of the pupil in the images. The pupil area calculation weighting function, visual fatigue detection, concentration detection, and eye/HMD slide compensation are included in the system. This allows the system to detect the pupil under different conditions. This system will know if anyone is using the system, then it evaluates the operating conditions of the subject, if the adjustment is correct, and guides the operator in using the system. The experimental results verified that this image system provides an economical and effective method for evaluating pupillary responses.     

水下目标被动测距的一种新方法:利用波导不变量提取目标距离信息

水声学中波导不变量的研究是近30年来引人注目的课题之一。水下目标辐射噪声的直达波和海面、海底反射波之间的干涉现象中隐含有下水目标的距离信息。提取这种距离信息就为水下目标的被动测距提供了一种新的途径。理论分析和实际海试都证明,甚至单水听器的LOFAR(Low Frequency Analysis Record)图都隐含着目标的距离和运动信息。本文给出利用波导不变量提取目标距离信息的理论推导,证明了在形成干涉条纹的外界条件具备时,利用多个水听器构成的基阵也能以较大增益提供目标距离信息。虽然组成基阵的每一水听器出现干涉条纹的条件是有差异的。这种差异在波束成形时可以加以利用和补偿。本文提出的理论和部分仿真、海试结果为水下目标被动测距和目标识别提供了一种新的途径。      

A new universality class in corpus of texts; A statistical physics study

Text can be regarded as a complex system. There are some methods in statistical physics which can be used to study this system. In this work, by means of statistical physics methods, we reveal new universal behaviors of texts associating with the fractality values of words in a text. The fractality measure indicates the importance of words in a text by considering distribution pattern of words throughout the text. We observed a power law relation between fractality of text and vocabulary size for texts and corpora. We also observed this behavior in studying biological data.