光学小波匹配滤波器与图形识别

小波变换弥补了傅里叶变换的不足，在信号及图像处理方面具有广阔的应用前景。文章深入讨论了小波变换匹配滤波器的理论基础及光学实现方法，并将其用于图形识别，给出了计算机模拟结果，与传统的匹配滤波器进行了比较，证实了小波变换在光学信息处理中的优越性。     

用多色菲涅耳全息滤波器作彩色图像特征识别

本文在用多色匹配滤波器(MSF)作彩色图像相关检测和用菲涅耳全息滤波器(FHF)作无透镜强度相关的基础上,提出一种把两者结合起来的技术,即用多色菲涅耳全息滤波器实现彩色图像的相关检测.     

用Joint变换实现特征识别

本文利用Joint变换的相关性质,首次分析了多特征信号的并行识别,实验证实了分析的正确.     

基于不变矩特征匹配的目标定位方法的实现

研究了弹道导弹在再入段的红外成像制导中存在几何失真及像质模糊的实时采集图像与基准地图的匹配定位问题。用不完全统计法证明经过平移、伸缩、旋转、加噪等多种复合变换下的图像的矩不变性,并选择6种经过消误差处理后的不变量作为弹道导弹成像匹配应用的特征量,采用Camberra距离最小法进行实时图像和基准图像的特征量匹配。实验部分验证了方法的正确性和可行性。     

用频域法实现圆环形目标物的计数

研究了用频域法实现圆环形目标物计数的方法。对空域中含有圆环形目标物的二值图像，经傅里叶变换后在频域内进行滤波，再经傅里叶反变换后取一阈值进行处理，则对应于圆环形目标物即变为一个实心体，然后对实心体进行计数即可实现对圆环形目标物的计数。对该方法进行了数学描述，并分别用模拟图像和实际图像进行了实验。结果表明，该方法消除噪声的效果好，对部分封闭圆的检测同样有效，且可用于目标物有粘连的情况。该方法对圆环形目标物的尺寸相差不大的情况更为有效。     

用双透镜Joint变换实现多特征识别

本文分析了多特征函数与函数阵列Joint变换的相关特性,提出用双透镜Joint变换实现多特征识别,并应用于变尺度.变方位特征识别.实验结果表明:分析是正确的,方法是可行的.     

用特征编码模板实现多模式分类识别的方法

以编码理论和神经网络优化算法为基础,提出了一种构造编码模板用于识别某一类模式集的方法,并以26个大写英文字母的识别作为实验进行了研究。     

Efficiency improvement of optical pattern recognition in ultra-fast all-optical intensity equalizer

To cope with a larger intensity fluctuation of optical signals in optical communication networks, we have proposed and demonstrated an ultra-fast all-optical intensity equalizer based on an optical pattern recognition technique. To achieve higher transfer efficiency for this equalizer, we designed and fabricated a highly-efficient optical amplitude-phase filter using a diffractive optical element for optical pattern recognition. The fabricated amplitude-phase filter provides a diffraction efficiency of 85%, and the all-optical intensity equalizer is demonstrated to have higher transfer efficiency, when this filter is used.     

振幅补偿匹配滤波中的旋转和尺度变化效应

本文研究了振幅补偿匹配滤波的旋转效应和尺度变化效应.计算机模拟的结果表明,这种滤波法对输入信号的旋转变化和尺度缩放变化,相对于传统的匹配滤波法及仅相位滤波法要敏感得多.文中还给出了这三和滤波下的相关峰值与旋转角度及尺度因子的关系曲线.     

光学相关器在自动目标识别中的应用

为使光学相关器更好地应用于自动目标识别领域,保证在较高识别速度下,具备良好的识别效果,介绍了应用于自动目标识别领域的联合变换相关器和匹配滤波相关器.针对匹配滤波相关器实现过程中的两项关键技术,在滤波器编码技术上提出了构造等相关峰综合鉴别函数以解决畸变不变识别问题.在相关峰识别技术上,针对阈值法存在的不足,提出利用BP神经网络对相关平面相关峰进行识别,采用划分感兴趣区域的方法减少了输入层神经元数目,简化了神经网络结构.借助搭建的光学相关器系统验证了该方法可对相关信号和噪声进行有效分类,提高了光学相关器的识别效果.     

Performance of an image-encrypted joint transform correlator for pattern recognition

An image-encrypted joint transform correlator for pattern recognition is proposed and evaluated. The correlator performs twofold correlation for recognizing the primary image and cipher, and the encrypting cipher acts to enhance or inhibit the correlation peak if the cipher is correctly input. By incorporating the original image with an appropriate cipher, the correlator achieves better pattern discrimination and is robust against distortion. Simulation results show that the pixel number and the cell size of cipher affect the characteristics of the correlator.     

一种用于光纤链路振动信号模式识别的规整化复合特征提取方法

相位光时域反射链路监测系统是一种利用光纤作为传感介质的传感系统, 能够监测、定位、识别入侵信号.模式识别模块是其重要组成部分, 实时智能区分安全扰动和危险入侵.本文提出一种用于光纤链路振动信号模式识别的复合特征提取方法.利用改进的双门限方法确定有效信号段的起止位置, 结合最大能量与最高信噪比挑选出采样周期内主要入侵扰动的特征段.综合利用特征段时域持续时间和小波包能量谱提取复合特征向量, 使用支持向量机进行模式识别.实验表明, 基于本文提出的规整化特征提取方法的模式识别准确率有了显著提高.     

药剂颜色标准的谱识别方法

提出一种药剂颜色标准的谱识别新方法 ,该方法突破了传统的三刺激值测色模式。提出了用光谱特征及非固定的背景模式 ,代替三刺激值及CIE测量模式 ,建立了良好的判别函数 ,适于药剂颜色标准的客观检验与定级 ,结果令人满意。     

Distortion-invariant pattern recognition using synthetic discriminant function based multiple phase-shifted-reference fringe-adjusted joint transform correlation

This paper proposes a novel pattern recognition system for invariance to noise and distortions. The technique first generates a synthetic discriminant function of the target image from its different distorted versions. It then takes four different phase-shifted versions of the reference image, which are individually joint transform correlated with the given input scene. Thus the proposed algorithm produces a single cross-correlation signal corresponding to each potential target. Also a fringe-adjusted filter is designed to generate a delta-like correlation peak with high discrimination between the signal and the noise. The pattern recognition system is also designed for the identification of multiple targets belonging to multiple reference objects simultaneously in a given input scene. The proposed technique is investigated using computer simulation including real-life images in different complex environments.     

非重叠背景噪声下的自适应维纳滤波模式识别方法

维纳滤波实现模式识别的关键问题是噪声知识的获得与估计。提出一种非重叠背景噪声的提取方法,首先将对噪声的粗略估计代入维纳滤波函数,得到相关峰。然后由相关峰的位置及参考图像的尺寸确定目标图像的位置和范围,从而提取出背景噪声图像。经过二次维纳滤波,得到改善的相关输出结果,实现了自适应过程。仿真结果表明在非重叠有色噪声环境下,与噪声估计法以及传统的维纳滤波方法相比,此方案具有较好的识别效果。     

基于光学子波变换的SDF匹配滤波器

提出了一种基于光学子波变换的综合判别函数匹配滤波器(WFSDF),将光学子波变换和综合判别函数相结合,只用一个简单的4f系统就能实现输入图像和SDF子波变换的相关运算。采用计算全息的方法制作复数匹配滤波器。计算机模拟仿真结果表明,与传统的SDF匹配滤波器相比,WFSDF匹配滤波器,锐化了相关峰,提高了识别率,同时简化了光路,能够实现多目标的畸不变识别。     

Existence problem of optical correlation based pattern recognition

The existence problem of optical correlation based pattern recognition, namely its range of validity and its limitation, is discussed in this paper conjointly with the function approximation theory of neural networks. The conclusion is that only if the sets to be recognized are linearly separable (which is rare) or the subsets, in which a segmental sample of the targets is involved, are linearly separable, can the classical 4f optical correlation system carry out the task of recognition inerrably. The recognition principle of a joint transform correlator is the same as that of a 4f system, and so is its range of validities. Based on the demonstration of the existence problem of optical correlation based pattern recognition an evaluation on some important problems that were studied in this field over the past 40 years is presented explicitly.