2种非对称广义高斯分布模型的构造  

Construction of Two Asymmetric Generalized Gaussian Distribution Models

作　　者：汪太月 戴燕青 WANG Taiyue;DAI Yanqing(School of Science,Hubei University of Technology,Wuhan Hubei 430068;School of Mathematics and Statistics,Hubei Polytechnic University,Huangshi Hubei 435003)

机构地区：[1]湖北工业大学理学院,湖北武汉430068 [2]湖北理工学院数理学院,湖北黄石435003

出　　处：《湖北理工学院学报》2020年第4期51-55,共5页Journal of Hubei Polytechnic University

基　　金：国家自然科学基金项目(项目编号:61601417)。

摘　　要：广义高斯分布常被用于图像处理及视频信号的统计分析之中,非对称广义高斯分布突破了对称性的束缚,用于数据处理建模更为适宜.从高斯分布入手,分别对左右两侧的不同方差和不同概率密度的广义高斯分布进行了改进,从而构造出了2种非对称的广义高斯模型,且通过模拟实验实现了构造模型的效果.The generalized Gaussian distribution is often used in image processing and video signal statistical analysis, while the asymmetric generalized Gaussian distribution breaks through the constraint of symmetry, so it is more suitable for data processing modeling. In this paper, starting with the Gaussian distribution, the generalized Gaussian distribution with different variance and different probability density on the left and right sides is improved respectively, so as to construct the asymmetric generalized Gaussian density models, and the model effect is achieved through simulation experiments.

关 键 词：高斯分布 广义高斯分布 非对称广义高斯分布 模型构造 

分 类 号：O212.2[理学—概率论与数理统计]