湖北民族学院学报（自然科学版）2018，Vol.36Issue(2)：194-198,5.DOI:10.13501/j.cnki.42-1569/n.2018.06.018
一种改进的Seam Carving图像重定向算法
An Improved Seam Carving Image Retargeting Method
摘要
Abstract
The main idea of content-aware image resizing is performing different operations according to the importance of each pixel,and the goal is to preserve the important areas and reduce visual distortion. Seam carving is one of the main methods of image resizing. Traditional seam carving generates the impor-tance map by gradient,which can only indicate high energy near the edges of an object.We propose a no-vel method combining gradient map and saliency map to generate an importance map,which can protect the edge information and visual prominent object,and preserve the integrity of the important areas. The experimental results demonstrate that the proposed method can reduce the visual distortion and preserve the content and structure of the original image.关键词
重定向/seamcarving/梯度图/视觉显著度图Key words
resizing/seam carving/gradient map/saliency map分类
信息技术与安全科学引用本文复制引用
翟胜楠,孟凡佳,刘金华..一种改进的Seam Carving图像重定向算法[J].湖北民族学院学报（自然科学版）,2018,36(2):194-198,5.基金项目
国家自然科学基金项目(U1404623). （U1404623）
