基于最大熵的目标分割和检测
|
摘要
论文研究图像分割包含两个简单子图象的合成图象,这两个简单子图象的先验知识是它们拥有全局最大熵。图象概率密度函数表明是准高斯型形式。估计概率密度函数的参数,然后将最大似然比检验法用于分割。采用迭代算法提高分割的准确性,扩展该方法用于任意概率密度函数的图象分割。
Segmentation of a composite image which contains two simple subimages is described. The a-priori knowledge about the two simple subimages is that they possess the maximum amount of entropy. The probability density functions(pdf s) of these image pixels are shown to be of the Quasi-gaussian form. Parameters for the pdf are estimated and then the maximum likelihood ratio test is applied to segmentation. An iterative algorithm is employed to improve the segmentation accuracy. Extension of this method to the segmentation of images with arbitrary pdf s is discussed.
Segmentation of a composite image which contains two simple subimages is described. The a-priori knowledge about the two simple subimages is that they possess the maximum amount of entropy. The probability density functions(pdf s) of these image pixels are shown to be of the Quasi-gaussian form. Parameters for the pdf are estimated and then the maximum likelihood ratio test is applied to segmentation. An iterative algorithm is employed to improve the segmentation accuracy. Extension of this method to the segmentation of images with arbitrary pdf s is discussed.
引文
[1]侯宏录,李宏.光电经纬仪测量飞行器三维坐标方法及误差分析[J].光电工程,2002,29(3):4-8.HOU Honglu,LI Hong.Measurement of Vehicle Three Dimension Coordinate with Photoelectrical Theodolite and Error Analysis[J].Opto-Electronic Engineering,2002,29(3):4-8.
[2]高世伟,郭雷,杨宁,等.一种新的粒子滤波目标跟踪算法[J].上海交通大学学报,2009,3(43):485-489.GAO Shiwei,GUO Lei,YANG Ning,et al.A New Particle Filter Object Tracking Algorithm[J].Journal of Shanghai Jiaotong University,2009,3(43):485-489.
[3]Yang Liu.A Novel Rate Control Scheme for Low Delay Video Communication of H.264/AVC Standard[J].IEEETrans Circuits Syst Video Technol,2007,17(1):68-78.
[4]Yin Min.A Rate Control Scheme for H.264 Video Under Low Bandwidth Channel[J].Journal of Zhejiang University:Science,2006:990-995.
[5]J.Cheng,Q.Liu,H.Lu.Texure classification using kernel independent component analysis[C]//Proc.of Int.Conf.on Pattern Recognition,2004:23-26.
[6]X.Liu,L.Cheng.Independent filters for texture classification[C]//Proc.of IEEE Int.Conf.on Image Processing,2002,3:24-28.
[7]X.Liu,L.Cheng.Independent spectral representations of images for recognition[J].Journal of the Optical Society of American,A,2003,20(7):1271-1282.
[8]X.Liu,D.Wang.Texture classification using spectral histograms[J].IEEE Trans.on Image Processing,2003,12(6):661-670.
[9]X.Liu,D.Wang.Texture classification using spectral histograms[J].IEEE Trans.on Image Processing,2003,12(6):661-670.
[10]B.C.Kuo,D.A.Landgrebe.Nonparametric weighted feature extraction for classification[J].IEEE Trans.Geosci.Remote Sens.,2004,42(5):1096-1105.
[2]高世伟,郭雷,杨宁,等.一种新的粒子滤波目标跟踪算法[J].上海交通大学学报,2009,3(43):485-489.GAO Shiwei,GUO Lei,YANG Ning,et al.A New Particle Filter Object Tracking Algorithm[J].Journal of Shanghai Jiaotong University,2009,3(43):485-489.
[3]Yang Liu.A Novel Rate Control Scheme for Low Delay Video Communication of H.264/AVC Standard[J].IEEETrans Circuits Syst Video Technol,2007,17(1):68-78.
[4]Yin Min.A Rate Control Scheme for H.264 Video Under Low Bandwidth Channel[J].Journal of Zhejiang University:Science,2006:990-995.
[5]J.Cheng,Q.Liu,H.Lu.Texure classification using kernel independent component analysis[C]//Proc.of Int.Conf.on Pattern Recognition,2004:23-26.
[6]X.Liu,L.Cheng.Independent filters for texture classification[C]//Proc.of IEEE Int.Conf.on Image Processing,2002,3:24-28.
[7]X.Liu,L.Cheng.Independent spectral representations of images for recognition[J].Journal of the Optical Society of American,A,2003,20(7):1271-1282.
[8]X.Liu,D.Wang.Texture classification using spectral histograms[J].IEEE Trans.on Image Processing,2003,12(6):661-670.
[9]X.Liu,D.Wang.Texture classification using spectral histograms[J].IEEE Trans.on Image Processing,2003,12(6):661-670.
[10]B.C.Kuo,D.A.Landgrebe.Nonparametric weighted feature extraction for classification[J].IEEE Trans.Geosci.Remote Sens.,2004,42(5):1096-1105.