保险丝(电气)
图像融合
人工智能
计算机科学
图像(数学)
比例(比率)
曲率
算法
计算机视觉
复合图像滤波器
形态梯度
模式识别(心理学)
图像处理
特征检测(计算机视觉)
数学
物理
几何学
量子力学
电气工程
工程类
作者
Wei Tan,Jiajia Zhao,Xinkai Liang,Zhongshi Lv,Hanchen Lu
摘要
Remote sensing images obtained from a single sensor have the disadvantage of incomplete information in terms of either low spatial resolutions or low spectral resolutions. To overcome this disadvantage, a pansharpening algorithm based on weighted mean curvature filtering (WMCF) and dual-channel pulse-coupled neural network (PCNN) in multi-scale morphological gradient (MSMG) domain is proposed in this paper. Firstly, the PAN image is decomposed into three parts, a small-scale image, a large-scale image, and a base image through a WMCF-based decomposition model. Then, a PCNN fusion strategy modulated by MSMG is used to fuse the base image and each band of the MS image. Finally, each fused bands are combined to obtain the final fused image. Experiments in four datasets demonstrate that the proposed algorithm obtains the best performance in most cases.
科研通智能强力驱动
Strongly Powered by AbleSci AI