基于主体区域保持的图像缩放算法  

An Novel Image Resizing Method Based on Important Area Maintain

作　　者：邹盼盼[1] 陆平[2,3] 朱恒亮[1] 董振江[3] 贾霞[3] 林晓[1,1] 马利庄[1] 

机构地区：[1]上海交通大学电子信息与电气工程学院数字媒体与数据重建实验室,上海200240 [2]东南大学信息科学与工程学院,江苏南京210000 [3]中兴通讯股份有限公司云计算及IT研究院,江苏南京210000

出　　处：《图学学报》2016年第2期230-236,共7页Journal of Graphics

基　　金：国家自然科学基金项目(U1304616,61133009,61502220);国家“973”重点基础研究发展规划项目(2011CB302200)

摘　　要：基于插值运算的缩放算法和经典的缝裁剪算法是两种常用的图像缩放算法,传统的缩放算法在缩放比例不一致的情况下其效果不佳,而缝裁剪算法在主体区域较大或者图像背景较为复杂时对图像的主体区域会造成一定破坏。针对以上问题,提出了一种基于主体区域保持的图像缩放算法,使用高斯差分对图像进行角点检测,利用角点产生凸包,根据凸包对图像进行主体区域检测,计算能量图并对位于主体区域像素点的能量给予相应的权重,根据权重的不同对主体区域进行不同程度的保护。实验结果表明,该算法能更好地保持图像主体区域。Scaling and seam carving are commonly used algorithm in image resizing, but scaling is not suitable for the case that the resize ratio is not uniformity, seam carving will also has a bad effect on important area in some cases. In this paper we proposed a novel algorithm which is based on important region maintain. Firstly, we make a corner detection based on difference of Gaussian, then Graham-scan algorithm is applied to do an important region detection, after that we will add different weights between important region and unimportant region. The experimental result shows our method is better than other image resizing algorithms in maintaining the important area.

关 键 词：缝裁剪 主体区域 角点检测 图像缩放 凸包 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]