基于广义回归神经网络的图像修复算法  

Image restoration algorithm based on generalized regression neural network

作　　者：王文霞[1] 王春红[1] 葛少磊 WANG Wen-xia WANG Chun-hong GE Shao - lei(Department of Computer Science and Technology, Yuncheng University, Yuncheng 044000,China School of Computer Science and Technology, Taiyuan University of Technology, Taiyuan 030024,China)

机构地区：[1]运城学院计算机科学与技术系,山西运城044000 [2]太原理工大学计算机科学与技术学院,山西太原030024

出　　处：《计算机工程与设计》2017年第11期3125-3130,共6页Computer Engineering and Design

基　　金：国家自然科学基金项目(11241005);山西省运城学院131人才专项基金项目(JG201634)

摘　　要：根据人类视觉感知的连通原理,提出一种基于广义回归神经网络的图像修复算法,该算法通过对图像数据进行回归分析确定缺失区域,具有简单和有效的特点。根据尺寸对缺失区域进行分离以及分类,将广义回归神经网络用于每个缺失区域,以便修复损坏的像素点。在文字去除、划痕去除以及噪声去除3个应用方面,对该算法的性能进行评估,采用客观衡量方式对修复图像的视觉质量进行评估。实验结果表明了该算法的有效性和可靠性。A new image restoration algorithm based on generalized regression neural network was proposed according to the human visual perception.It was simple and effective to determine the missing area by regression analysis of the image data.According to the size,the missing area was separated and classified.The generalized regression neural network was used for each missing area,to repair the damaged pixels.The performance of the proposed algorithm was evaluated in three applications including word removal,scratch removal and noise removal,and the visual quality of the restored image was evaluated using the objective measurement.Experimental results show that the proposed algorithm is effective and reliable.

关 键 词：图像修复 广义回归神经网络 文字去除 峰值信噪比 划痕去除 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]