一种基于图像几何矩的非降维连续尺度解构模式
|
摘要
针对降维解构分析的多尺度图像处理容易造成高维信息的破碎与损失这一问题,建立了一种基于图像几何矩的非降维连续尺度解构模式,并在此模式下构建了10组非降维解构模型。模型以单通道图像为应用对象,以几何矩为算子,以不同尺度窗口下矩运算产生的特征值为基础对原始信息进行非降维解构。10组解构结果呈现均衡模式和增长模式两种形态。代入多尺度分割算法验证证明,基于图像几何矩的非降维连续解构分析可以提升多尺度分割精度,且增长模式下的解构信息对于图像分割更有利。
Dimensionality-reduction analyses are effective to deconstruct image completely,but also destructive for high-dimension information of image.Therefore,this study proposes a 2-dimension multiscale image deconstruction mode to solve this problem which is based on geometric moment.In this mode,10 groups of 2-dimension multiscale deconstruction models are built.Models use single channel image as target,use different geometric moments as operators,use different scales of windows as operand,and use moment calculation to get eigenvalues.Based on eigenvalues,multiscale models are built.According to statistics analysis,10 models are in two patterns,which are equilibrium pattern and growth pattern.Result shows that 2-dimension multiscale image deconstruction analysis could improve the accuracy of multiscale image segmentation,and growth pattern models are more effective than equilibrium pattern models.
Dimensionality-reduction analyses are effective to deconstruct image completely,but also destructive for high-dimension information of image.Therefore,this study proposes a 2-dimension multiscale image deconstruction mode to solve this problem which is based on geometric moment.In this mode,10 groups of 2-dimension multiscale deconstruction models are built.Models use single channel image as target,use different geometric moments as operators,use different scales of windows as operand,and use moment calculation to get eigenvalues.Based on eigenvalues,multiscale models are built.According to statistics analysis,10 models are in two patterns,which are equilibrium pattern and growth pattern.Result shows that 2-dimension multiscale image deconstruction analysis could improve the accuracy of multiscale image segmentation,and growth pattern models are more effective than equilibrium pattern models.
引文
[1]WANG Z,SIMONCELLI E P,BOVIK A C.Multiscale structural similarity for image quality assessment[C]∥The Thrity-Seventh Asilomar Conference on Signals,Systems&Computers,2003.IEEE,2003,2:1398-1402.
[2]王楠楠,李桂苓.符合人眼视觉特性的视频质量评价模型[J].中国图象图形学报,2001(6):15-19.
[3]曹建农,王平禄,董昱威.高分辨率遥感影像上居民地自动提取方法[J].武汉大学学报(信息科学版),2014,39(7):831-837.
[4]陈恩庆,王建波,齐林.一种新的多尺度边缘跟踪算法及其应用[J].计算机应用研究,2015,32(12):3801-3804.
[5]CANDSE J,DONOHO D L.Curvelets[R].Stanford:Department of Statistics,Stanford University,1999.
[6]DO M N,VETTERLI M.The contourlet transform:an efficient directional multiresolution image representation[J].IEEE Trans on Image Processing,2005,14(12):2091-2106.
[7]LIM W Q.The discrete shearlet transform:a new directional transform and compactly supported shearlet frames[J].IEEE Trans on Image Processing,2010,19(5):1166-1180.
[8]COHEN A,DAUBECHIES I.Non-separable bidimensional wavelet bases[J].Revista Matematica Iberoamericana,1993,9(1):51-137.
[9]JIA R Q.Multiresolution of LP Space[J].Math,1994,184:620-639.
[10]龙瑞麟.高维小波分析[M].北京:世界图书出版公司,1995.
[11]东野升云.多尺度分析与相似度的立体视频人类视觉评价模型的研究[D].长春:吉林大学,2011.
[12]关玉景.广义多尺度分析和等度多尺度分析及其应用[D].长春:吉林大学,2000.
[13]崔丽.N维广义拟实数进制小波&小波矩理论及其应用[D].长春:吉林大学,2004.
[14]周俊平.图像分割的多尺度方法研究[J].煤炭技术,2013,32(1):218-219.
[15]龙瑞霖.基于Krawtchouk矩的图像分析[D].南京:南京理工大学,2012.
[16]肖家庆,卢凌,李晟,等.基于矩不变量的图像目标识别[J].武汉理工大学学报(交通科学与工程版),2006(4):696-699.
[17]王海波,平子良.几种正交矩描述图像的性能比较[J].内蒙古师范大学学报(自然科学汉文版),2005(1):35-39.
[2]王楠楠,李桂苓.符合人眼视觉特性的视频质量评价模型[J].中国图象图形学报,2001(6):15-19.
[3]曹建农,王平禄,董昱威.高分辨率遥感影像上居民地自动提取方法[J].武汉大学学报(信息科学版),2014,39(7):831-837.
[4]陈恩庆,王建波,齐林.一种新的多尺度边缘跟踪算法及其应用[J].计算机应用研究,2015,32(12):3801-3804.
[5]CANDSE J,DONOHO D L.Curvelets[R].Stanford:Department of Statistics,Stanford University,1999.
[6]DO M N,VETTERLI M.The contourlet transform:an efficient directional multiresolution image representation[J].IEEE Trans on Image Processing,2005,14(12):2091-2106.
[7]LIM W Q.The discrete shearlet transform:a new directional transform and compactly supported shearlet frames[J].IEEE Trans on Image Processing,2010,19(5):1166-1180.
[8]COHEN A,DAUBECHIES I.Non-separable bidimensional wavelet bases[J].Revista Matematica Iberoamericana,1993,9(1):51-137.
[9]JIA R Q.Multiresolution of LP Space[J].Math,1994,184:620-639.
[10]龙瑞麟.高维小波分析[M].北京:世界图书出版公司,1995.
[11]东野升云.多尺度分析与相似度的立体视频人类视觉评价模型的研究[D].长春:吉林大学,2011.
[12]关玉景.广义多尺度分析和等度多尺度分析及其应用[D].长春:吉林大学,2000.
[13]崔丽.N维广义拟实数进制小波&小波矩理论及其应用[D].长春:吉林大学,2004.
[14]周俊平.图像分割的多尺度方法研究[J].煤炭技术,2013,32(1):218-219.
[15]龙瑞霖.基于Krawtchouk矩的图像分析[D].南京:南京理工大学,2012.
[16]肖家庆,卢凌,李晟,等.基于矩不变量的图像目标识别[J].武汉理工大学学报(交通科学与工程版),2006(4):696-699.
[17]王海波,平子良.几种正交矩描述图像的性能比较[J].内蒙古师范大学学报(自然科学汉文版),2005(1):35-39.