当前位置: X-MOL 学术IEEE Signal Process. Lett. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
A Fast Image Fusion with Discrete Cosine Transform
IEEE Signal Processing Letters ( IF 3.2 ) Pub Date : 2020-01-01 , DOI: 10.1109/lsp.2020.2999788
Monan Wang , Xiping Shang

A fast and effective image fusion method based on discrete cosine transform (DCT) is proposed. The fusion quality and computation time of current DCT-based image fusion methods largely depend on the selected block size, and the selection of the block size is very difficult in practice. A novel image fusion method based on DCT coefficient matrix features is proposed, which can effectively overcome the above difficulties. A DCT-based image fusion framework is proposed, which decomposes each source image into a base layer and a detail layer for image fusion. And optimize the calculation method of the base layer to better preserve the structure of the image. The effectiveness of the proposed method is verified by six image databases with more than 90 pairs of source images in total. The experimental results show that the proposed method can obtain the most effective results in terms of visual quality and objective evaluation of medical image fusion, and the fusion time is more efficient.

中文翻译:

具有离散余弦变换的快速图像融合

提出了一种基于离散余弦变换(DCT)的快速有效的图像融合方法。当前基于DCT的图像融合方法的融合质量和计算时间在很大程度上取决于所选择的块大小,块大小的选择在实践中非常困难。提出了一种新的基于DCT系数矩阵特征的图像融合方法,可以有效克服上述困难。提出了一种基于DCT的图像融合框架,将每个源图像分解为基础层和细节层进行图像融合。并优化了base layer的计算方式,更好的保留了图像的结构。所提出方法的有效性得到了六个图像数据库的验证,该数据库总共有 90 多对源图像。
更新日期:2020-01-01
down
wechat
bug