崔嘉^{1, 2},
宋磊^{1, 2},
陆宏菊³,
唐明晰⁴,
戚萌^{1, 2,,}
1.山东师范大学信息科学与工程学院 济南 250358
2.山东师范大学智能信息计算与安全实验室 济南 250358
3.济南技师学院 济南 250031
4.香港理工大学设计学院 香港 999077
基金项目:国家自然科学基金(61902225, 61502285)，山东省青年基金(ZR2014FQ013)

详细信息

作者简介:崔嘉：男，1982年生，副教授，研究方向为图像处理、计算机辅助设计、智能设计
宋磊：男，1993年生，硕士生，研究方向为图像处理
陆宏菊：女，1982年生，讲师，研究方向为图像处理、语义分析
唐明晰：男，1956年生，教授，研究方向为智能设计、设计推理、计算机图形学
戚萌：女，1983年生，讲师，研究方向为虚拟现实、计算机图形学

通讯作者:戚萌　qimeng@sdnu.edu.cn

中图分类号:TP391.41

计量

文章访问数:292
HTML全文浏览量:156
PDF下载量:26
被引次数:0

出版历程

收稿日期:2019-12-30
修回日期:2020-10-23
网络出版日期:2020-12-11
刊出日期:2021-04-20

Continuous Seam Carving Algorithm Based on Just Noticeable Distortion Algorithm in Image Retargeting

Jia CUI^{1, 2},
Lei SONG^{1, 2},
Hongju LU³,
Mingxi TANG⁴,
Meng QI^{1, 2,,}
1. School of Information Science and Engineering, Shandong Normal University, Jinan 250358, China
2. Key Laboratory of Intelligent Information Processing, Shandong Normal University, Jinan 250358, China
3. College of Jinan Institute of Technology, Jinan 250031, China
4. School of Design, The Hong Kong Polytechnic University, Hong Kong 999077, China
Funds:The National Natural Science Foundation of China (61902225, 61502285), The Natural Science Foundation of Shandong Province (ZR2014FQ013)


摘要
摘要:图像缩放技术要求对图像缩放的同时保证重要信息不丢失且物体边缘不发生扭曲。近年来，Seam Carving及其改进算法得到了广泛的关注和研究。由于采用了离散式最小能量线迭代搜索策略，缩放信息无法在迭代过程中传递导致扭曲现象普遍存在。该文针对上述问题提出最小位移可视差(JND)检测算法，能够有效地检测每一次迭代中出现的潜在扭曲信息。能量权重$ {E}_{w} $能够将JND信息累加传递给后续的迭代过程，从而抑制缩放过程中的边缘扭曲现象。通过JND算法和能量权重，该文首次将离散的Seam Carving模型转变为连续缩放模型。最后，在公共数据集RetargetMe上与最新的图像缩放算法进行多组对比实验，验证了所提方法的有效性和先进性。
关键词:图像缩放/
Seam Carving/
最小位移可视差/
平均场近似
Abstract:Image retargeting technologies require important information preservation and less edge distortion during increasing/decreasing image size. The seam carving based algorithms, as the classic retargeting model, receive widespread attention in recent years. However, because of the discrete least energy seam searching strategy, the retargeting information can not be passed generation by generation, which causes retargeting distortions to prevail. To solve this problem, the Just Noticeable Distortion (JND) algorithm is proposed to detect the potential distribution of distortion information. Through the proposed energy weight E_w, the JND information can be passed to the following retargeting iteration for distortion reduction. According to the best knowledge, it is the first time to propose the seam carving algorithm in continuous way by the JND algorithm and energy weight, are the promising results also demonstrated compared with several new approaches at public database ‘Retarget Me’, qualitatively and quantitatively.
Key words:Image Retargeting/
Seam Carving/
Just Noticeable Distortion (JND)/
Mean Approximation



PDF全文下载地址:

https://jeit.ac.cn/article/exportPdf?id=8446c219-60d6-46d2-bac0-8b53561a2070