基于变分方法的图像平滑和纹理提取  

Research on Image Smoothing and Texture Extraction Based on Variational Method

作　　者：吕小光 孙雪曼 朱国良 姜乐 卢盛泰 LüXiaoguang;SUN Xueman;ZHU guoliang;JIANG Le;LU Shengtai(School of Science,Jiangsu Ocean University,Lianyungang 222005,China;School of Computer Engineering,Jiangsu Ocean University,Lianyungang 222005,China;School of Electronic Engineering,Jiangsu Ocean University,Lianyungang 222005,China)

机构地区：[1]江苏海洋大学理学院,江苏连云港222005 [2]江苏海洋大学计算机工程学院,江苏连云港222005 [3]江苏海洋大学电子工程学院,江苏连云港222005

出　　处：《江苏海洋大学学报（自然科学版）》2021年第2期77-84,共8页Journal of Jiangsu Ocean University:Natural Science Edition

基　　金：江苏省自然科学基金—面上项目(BK20181483);连云港市“海燕计划”资助项目;连云港市“521高层次人才培养工程”资助项目;江苏海洋大学自然科学基金资助项目(Z2017004)。

摘　　要：一幅数字图像可以分解为光滑部分和纹理部分,光滑部分包含了图像结构的卡通信息,纹理部分包含了图像细节的边缘信息。图像平滑和纹理提取在图像处理、人工智能和模式识别等方面有着重要的应用。为了有效地平滑图像和提取纹理,建立了基于变分方法的图像平滑和纹理提取模型,采用交替最小化方法求解模型。在实验仿真中,通过与其他4种常用方法进行比较,验证了模型和算法的优越性,给出了模型及算法在图像合成、伪影去除和细节增强等方面的应用。A digital image can be decomposed into smoothing component and texture component,where the smoothing component contains the cartoon structural information and the texture component contains the edge information.Image smoothing and texture extracting play an important role in a wide range of applications such as image processing,artificial intelligence and pattern recognition.In order to smooth the image and extract the texture effectively,the model of image smoothing and texture extraction based on variational method is established,and the alternating minimization method is used to solve the model.In the experimental simulation,the superiority of the model and algorithm is verified by comparing with other four popular methods,and the application of the model and algorithm in image synthesis,artifact removal and detail enhancement is given.

关 键 词：变分 图像平滑 纹理提取 交替最小化 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]