红外图像的目标识别技术研究
|
摘要
随着计算机技术的高速发展,红外图像目标识别技术得到了广泛关注和应用。利用红外CCD相机对目标进行图像采集,并对采集的图像进行识别是红外图像目标识别的基本任务。红外图像目标识别中最为关键的技术是预处理、特征量提取和分类器设计。
红外图像的预处理是红外图像目标识别的第一步,主要包括图像滤波,图像增强,阈值分割。本文使用综合滤波法进行了滤波处理,使用拉普拉斯变换和Butterworth高通滤波进行了图像增强处理,使用迭代法和最小类间方差法对图像进行了阈值分割处理,均取得了令人满意的效果。
红外图像特征提取的任务是获得可分性强、稳定性高的数学特征。本文从红外CCD的成像特性、目标的几何特征、图像矩、奇异值变换等角度分别讨论了红外目标图像特征提取的方法。首先,根据目标的几何特征量的要求,通过设计目标的最小外接矩形和选择合适的边缘提取算法等途径实现了对目标几何特征量的提取;进而,利用图像矩提取红外目标的区域形状的特征;再者,使用矩阵的SVD变换提取了目标的尺度奇异值特征,并针对目前存在的问题进行了部分的改进;最后,通过相关的实验,验证了这三种特征提取方法的有效性。
分类识别是红外图像目标识别的最后环节,设计性能优越的分类器及其算法是分类能否成功实现的关键。本文主要对最小距离分类器和基于最小错误率的Bayes分类器作了探讨和研究,两种分类方法都取得了良好的分类效果。
本文对预处理,特征提取和分类器设计中的各个算法进行了仿真,并验证了这些算法的可行性。
As the rapid development of computer technology, infrared image pattern recognition technology based on image processing gets more attentions and broad applications. Target recognition based on infrared CCD comera capturing imagines of the targets is a basic task of the image target recognition. Image preprocessing, feature extraction and classifiers design are most important technologies of image target recognition.
Preprocessing is the first step of image target recognition, which has three parties: noise reduction, image enhancement and image segment. In this thesis, the compositive filter method is applied to image filter, Laplace transformation and Butterworth high-pass method are applied to the image enhancement, and Otsu method and Iterative method are applied to the image segment. The preprocess results of these methods are effective.
The main mission of feature extraction is obtaining the mathematic features with high divisibility and stability. On the basis of the characteristics of infrared CCD comera, of the target’s shape, of the imagine quadrature and of the SVD transformation, these feature extraction methods are discussed. Firstly, geometry parameter fearures are extracted by means of designing the minimal rect of the target and choicing fit algorithm of edge detection. Secondly, several taeget’s scope shape features are extracted according to the second-order moment. Thirdly, SVD transformation features from target are extracted and some improvement is done according to the existing problem. Finally, these three feature extracting methods are proved effective through relevant experiments.
Classification-recognition is final step of image target recognition, so the prominent classifier and algorithms are key steps for classification. In this thesis, the least distance method and Bayes based on least of the rates about mistakes are designed, which are proved effective for targets’classification.
All methods of preprocessing, of feature extraction and of classier mentioned in this thesis are simulated, and were proved feasibily.
红外图像的预处理是红外图像目标识别的第一步,主要包括图像滤波,图像增强,阈值分割。本文使用综合滤波法进行了滤波处理,使用拉普拉斯变换和Butterworth高通滤波进行了图像增强处理,使用迭代法和最小类间方差法对图像进行了阈值分割处理,均取得了令人满意的效果。
红外图像特征提取的任务是获得可分性强、稳定性高的数学特征。本文从红外CCD的成像特性、目标的几何特征、图像矩、奇异值变换等角度分别讨论了红外目标图像特征提取的方法。首先,根据目标的几何特征量的要求,通过设计目标的最小外接矩形和选择合适的边缘提取算法等途径实现了对目标几何特征量的提取;进而,利用图像矩提取红外目标的区域形状的特征;再者,使用矩阵的SVD变换提取了目标的尺度奇异值特征,并针对目前存在的问题进行了部分的改进;最后,通过相关的实验,验证了这三种特征提取方法的有效性。
分类识别是红外图像目标识别的最后环节,设计性能优越的分类器及其算法是分类能否成功实现的关键。本文主要对最小距离分类器和基于最小错误率的Bayes分类器作了探讨和研究,两种分类方法都取得了良好的分类效果。
本文对预处理,特征提取和分类器设计中的各个算法进行了仿真,并验证了这些算法的可行性。
As the rapid development of computer technology, infrared image pattern recognition technology based on image processing gets more attentions and broad applications. Target recognition based on infrared CCD comera capturing imagines of the targets is a basic task of the image target recognition. Image preprocessing, feature extraction and classifiers design are most important technologies of image target recognition.
Preprocessing is the first step of image target recognition, which has three parties: noise reduction, image enhancement and image segment. In this thesis, the compositive filter method is applied to image filter, Laplace transformation and Butterworth high-pass method are applied to the image enhancement, and Otsu method and Iterative method are applied to the image segment. The preprocess results of these methods are effective.
The main mission of feature extraction is obtaining the mathematic features with high divisibility and stability. On the basis of the characteristics of infrared CCD comera, of the target’s shape, of the imagine quadrature and of the SVD transformation, these feature extraction methods are discussed. Firstly, geometry parameter fearures are extracted by means of designing the minimal rect of the target and choicing fit algorithm of edge detection. Secondly, several taeget’s scope shape features are extracted according to the second-order moment. Thirdly, SVD transformation features from target are extracted and some improvement is done according to the existing problem. Finally, these three feature extracting methods are proved effective through relevant experiments.
Classification-recognition is final step of image target recognition, so the prominent classifier and algorithms are key steps for classification. In this thesis, the least distance method and Bayes based on least of the rates about mistakes are designed, which are proved effective for targets’classification.
All methods of preprocessing, of feature extraction and of classier mentioned in this thesis are simulated, and were proved feasibily.
引文
[1]张宇波.声信号识别研究及在机械运行状态预测中的应用[D].长沙:湖南大学, 2004.
[2]卢瑞祥,牟轩沁,纪震.一种基于红外图像识别的自动消防监控系统[J].电子技术应用. 1998.2:7-8
[3]韩军利.红外图像目标识别技术研究[D].南京:南京理工大学. 2004.
[4]陈红伟,杨树谦.自动目标识别技术在飞航导弹上的应用[J].红外与激光工程.全国光电技术学术交流会论文集. 2004:822-824
[5]余静,游志胜.自动目标识别与跟踪技术研究综述[J].计算机应用研究. 2005.1:12-15
[6]王自勇,廖朝佩.红外图像自动目标识别技术进展[J].飞航导弹. 1996.7:41-48
[7]王甜.红外成像制导前视匹配目标识别方法研究[D].华中科技大学. 2006.
[8]边肇祺,张学工等.模式识别(第二版)[M].北京:清华大学出版社. 2000.1: 1-100.
[9]邹前进,冯亮,汪亚.红外图像空间噪声分析和预处理方法改进[J].应用光学. 2007.7: 426-430
[10]张秀凤,娄树理,张燕妮.红外图像模式识别的预处理研究[J].光电技术应用. 2006.10:58-60
[11]周刚.基于阈值划分的图像混合滤波器[D].新疆大学. 2007
[12]Rafael C Gonzalez, Richard E Woods. Digital ImageProcessing[M].北京:电子工业出版社. 2002:125-142
[13]何斌,马天予等. Visual C++数字图像处理(第二版)[M].北京:人民邮电出版社. 2002
[14]阴躲芳,李一民,王英妹,王林.图像阈值分割技术的研究[J].科技广场. 2008.3:136-137
[15]杨耿.运动目标的图像识别与跟踪研究[D].江苏大学. 2005.
[16]刘贯领.声目标识别方法研究[D].南京:南京理工大学. 2003.
[17]孙即祥,王晓华等.模式识别中的特征提取与计算机视觉不变量[D].北京:国防工业出版社. 2001
[18]李了了,邓善熙,丁兴号.基于大津法的图像分块二值化算法[J].微计算机信息. 2005.24:57-61
[19]魏伟波,芮筱亭.不变矩方法研究[J].火力与指挥控制. 2007.11.
[20]Hu.M. Visual Pattern Recognition By Moment Invariants[J]. IRE Trans. Inform. Theory, 1962.8: 179-187
[21]Y R Wong. Scene Matching With Invariant Moment[J]. Computer Graphics And Image Processing. 1978.8:16-24
[22]陈东,黄勇杰,沈振康.基于尺度奇异值变换的红外图像目标特征提取方法研究[J].红外与激光工程. 2001.6:157-159
[23]张忠诚,孟庆华,沈振康.红外目标特征分析[J].激光与红外. 1999.6:166-169
[24]郎锐.数字图像处理学[D].北京:北京希望电子出版社. 2003
[25]吴晓光,王涤琼,盛慧.一种获取图像区域最小外接矩形的算法及实现[J].计算机工程. 2004.6:124-126
[26]Y. L. Sheng, L. X. Shen. Orthogonal Fourier-Mellin moments for invariant pattern recognition〔J〕. J. Op t. Soc. AmA, 1994, 11 (6): 1748-1757.
[27]张春燕,翟素蓝,罗斌。基于形状不变矩的t-混合模型的图像分类[J]。计算机工程与应用. 2007.10:67-72
[28]M.R.Teague. Image analysis via the general theary of moment[J]. J.Opt.Soc. 1980,70 (8): 920-930
[29]C.H.Teh, R.T.Chin. On image analysis by the methods fo moment(J). IEEE Trans. Pattern Anal. Mach. Intell , 1988,10(4):496-513
[30]AKhotanzad, Y. H. Hong. Invariant image recognition by Zernike moments〔J〕. IEEE Trans. Pattern Anal. Mach. Intell, 1990, 12 (5):489-497.
[31]Y. L. Sheng, L. X. Shen. Orthogonal Fourier-Mellin moments for invariant pattern recognition〔J〕. J. Op t. Soc. AmA, 1994, 11 (6): 1748-1757.
[32]Z. L. Ping, R. G. Wu, Y. L. Sheng. Describing image with Chebyshev-Fourier Moments〔J〕. J. Op t. Soc. Am, 2002, 19(9): 1748-1754.
[33]GulengAmu, Surong Hasi, Xingyu Yang, Zilziang Ping. Image Analysis byPseudo-Jacobi (p = 4, q = 3)- Fourier Moments〔J〕. Applied Optics , 2004, 43 (10) : 2093 - 2101.
[34]李玉景,李林,李京.基于不变矩和支持向量机理论的舰船目标识别[J].科技信息. 2007.29:536-537
[35]Chen C.C Improved moment invariant for shape discrimination[J]. Pattern Recognition,1993,26(5):683-686
[36]杨建刚.人工神经网络实用教程[M].杭州:浙江大学出版社. 2001: 41-62,183-201.
[37]罗小桂,何雁.矩阵奇异值分解在计算技术中的应用[J].计算机与现代化. 2006.6: 67-68.
[38] Groutage D ,Bennink D. Feature sets for nonstationary signals.derivedfrom moments of the singular value decomposition of Cohen2Posch(positive time2 frequency) distributions[J]. Signal ProcessingIEEE Transactions on ,2000 ,48(5) :1498 - 1503.
[39]高仕龙.矩阵奇异值分解的图像性质及其应用[J].乐山师范学院学报. 2008.5:14-15
[40]杨淑莹.图像模式识别—VC++技术实现[M].北京:清华大学出版社,北京交通大学出版社. 2005
[41]吴逸飞译.模式识别——原理、方法及应用[M].北京:清华大学出版社. 2002: 1-55.
[42]聂伟荣.多传感器探测与控制网络技术——地面运动目标震动信号探测与识别[D].南京:南京理工大学, 2001.
[43]李月景等.图像识别技术及其应用[M].北京:机械工业出版社. 1985
[44]杨国青,陶常亮,郇正良. Bayes决策理论用于模式识别分类问题的进一步探讨.泰安师专学报[J]. 2002. 11: 6-8
[45]王军,王员云.基于Bayes决策的联机手写数字识别[J].福建电脑. 2008.3: 94-95
[46]章慎锋,杨淑莹,王厚雪.基于Bayes决策的手写体数字识别[J].天津理工大学学报. 2006. 2: 78-80
[47]肖斌.基于不变矩的图像几何变换不变性识别研究[D].陕西:陕西师范大学. 2007.4
[48]齐晓轩.基于BP神经网络的战场地面传感器侦察系统中目标识别算法的研究[D].沈阳:沈阳工学院. 2003.
[2]卢瑞祥,牟轩沁,纪震.一种基于红外图像识别的自动消防监控系统[J].电子技术应用. 1998.2:7-8
[3]韩军利.红外图像目标识别技术研究[D].南京:南京理工大学. 2004.
[4]陈红伟,杨树谦.自动目标识别技术在飞航导弹上的应用[J].红外与激光工程.全国光电技术学术交流会论文集. 2004:822-824
[5]余静,游志胜.自动目标识别与跟踪技术研究综述[J].计算机应用研究. 2005.1:12-15
[6]王自勇,廖朝佩.红外图像自动目标识别技术进展[J].飞航导弹. 1996.7:41-48
[7]王甜.红外成像制导前视匹配目标识别方法研究[D].华中科技大学. 2006.
[8]边肇祺,张学工等.模式识别(第二版)[M].北京:清华大学出版社. 2000.1: 1-100.
[9]邹前进,冯亮,汪亚.红外图像空间噪声分析和预处理方法改进[J].应用光学. 2007.7: 426-430
[10]张秀凤,娄树理,张燕妮.红外图像模式识别的预处理研究[J].光电技术应用. 2006.10:58-60
[11]周刚.基于阈值划分的图像混合滤波器[D].新疆大学. 2007
[12]Rafael C Gonzalez, Richard E Woods. Digital ImageProcessing[M].北京:电子工业出版社. 2002:125-142
[13]何斌,马天予等. Visual C++数字图像处理(第二版)[M].北京:人民邮电出版社. 2002
[14]阴躲芳,李一民,王英妹,王林.图像阈值分割技术的研究[J].科技广场. 2008.3:136-137
[15]杨耿.运动目标的图像识别与跟踪研究[D].江苏大学. 2005.
[16]刘贯领.声目标识别方法研究[D].南京:南京理工大学. 2003.
[17]孙即祥,王晓华等.模式识别中的特征提取与计算机视觉不变量[D].北京:国防工业出版社. 2001
[18]李了了,邓善熙,丁兴号.基于大津法的图像分块二值化算法[J].微计算机信息. 2005.24:57-61
[19]魏伟波,芮筱亭.不变矩方法研究[J].火力与指挥控制. 2007.11.
[20]Hu.M. Visual Pattern Recognition By Moment Invariants[J]. IRE Trans. Inform. Theory, 1962.8: 179-187
[21]Y R Wong. Scene Matching With Invariant Moment[J]. Computer Graphics And Image Processing. 1978.8:16-24
[22]陈东,黄勇杰,沈振康.基于尺度奇异值变换的红外图像目标特征提取方法研究[J].红外与激光工程. 2001.6:157-159
[23]张忠诚,孟庆华,沈振康.红外目标特征分析[J].激光与红外. 1999.6:166-169
[24]郎锐.数字图像处理学[D].北京:北京希望电子出版社. 2003
[25]吴晓光,王涤琼,盛慧.一种获取图像区域最小外接矩形的算法及实现[J].计算机工程. 2004.6:124-126
[26]Y. L. Sheng, L. X. Shen. Orthogonal Fourier-Mellin moments for invariant pattern recognition〔J〕. J. Op t. Soc. AmA, 1994, 11 (6): 1748-1757.
[27]张春燕,翟素蓝,罗斌。基于形状不变矩的t-混合模型的图像分类[J]。计算机工程与应用. 2007.10:67-72
[28]M.R.Teague. Image analysis via the general theary of moment[J]. J.Opt.Soc. 1980,70 (8): 920-930
[29]C.H.Teh, R.T.Chin. On image analysis by the methods fo moment(J). IEEE Trans. Pattern Anal. Mach. Intell , 1988,10(4):496-513
[30]AKhotanzad, Y. H. Hong. Invariant image recognition by Zernike moments〔J〕. IEEE Trans. Pattern Anal. Mach. Intell, 1990, 12 (5):489-497.
[31]Y. L. Sheng, L. X. Shen. Orthogonal Fourier-Mellin moments for invariant pattern recognition〔J〕. J. Op t. Soc. AmA, 1994, 11 (6): 1748-1757.
[32]Z. L. Ping, R. G. Wu, Y. L. Sheng. Describing image with Chebyshev-Fourier Moments〔J〕. J. Op t. Soc. Am, 2002, 19(9): 1748-1754.
[33]GulengAmu, Surong Hasi, Xingyu Yang, Zilziang Ping. Image Analysis byPseudo-Jacobi (p = 4, q = 3)- Fourier Moments〔J〕. Applied Optics , 2004, 43 (10) : 2093 - 2101.
[34]李玉景,李林,李京.基于不变矩和支持向量机理论的舰船目标识别[J].科技信息. 2007.29:536-537
[35]Chen C.C Improved moment invariant for shape discrimination[J]. Pattern Recognition,1993,26(5):683-686
[36]杨建刚.人工神经网络实用教程[M].杭州:浙江大学出版社. 2001: 41-62,183-201.
[37]罗小桂,何雁.矩阵奇异值分解在计算技术中的应用[J].计算机与现代化. 2006.6: 67-68.
[38] Groutage D ,Bennink D. Feature sets for nonstationary signals.derivedfrom moments of the singular value decomposition of Cohen2Posch(positive time2 frequency) distributions[J]. Signal ProcessingIEEE Transactions on ,2000 ,48(5) :1498 - 1503.
[39]高仕龙.矩阵奇异值分解的图像性质及其应用[J].乐山师范学院学报. 2008.5:14-15
[40]杨淑莹.图像模式识别—VC++技术实现[M].北京:清华大学出版社,北京交通大学出版社. 2005
[41]吴逸飞译.模式识别——原理、方法及应用[M].北京:清华大学出版社. 2002: 1-55.
[42]聂伟荣.多传感器探测与控制网络技术——地面运动目标震动信号探测与识别[D].南京:南京理工大学, 2001.
[43]李月景等.图像识别技术及其应用[M].北京:机械工业出版社. 1985
[44]杨国青,陶常亮,郇正良. Bayes决策理论用于模式识别分类问题的进一步探讨.泰安师专学报[J]. 2002. 11: 6-8
[45]王军,王员云.基于Bayes决策的联机手写数字识别[J].福建电脑. 2008.3: 94-95
[46]章慎锋,杨淑莹,王厚雪.基于Bayes决策的手写体数字识别[J].天津理工大学学报. 2006. 2: 78-80
[47]肖斌.基于不变矩的图像几何变换不变性识别研究[D].陕西:陕西师范大学. 2007.4
[48]齐晓轩.基于BP神经网络的战场地面传感器侦察系统中目标识别算法的研究[D].沈阳:沈阳工学院. 2003.