图像融合
计算机视觉
人工智能
计算机科学
杠杆(统计)
融合
遥感
保险丝(电气)
自编码
传感器融合
图像(数学)
合成孔径雷达
转化(遗传学)
图像质量
噪音(视频)
图像纹理
光学图像
图像传感器
遥感应用
模式识别(心理学)
图像处理
图像配准
像素
作者
Benyu Ji,Yun Zhang,Hongyue Gao,Zhiguo Liang,Yankun Huang,Hang Ren,Qinglong Hua,Yaohua Li
出处
期刊:
[Institute of Electrical and Electronics Engineers]
日期:2025-08-03
卷期号:: 8878-8882
被引量:2
标识
DOI:10.1109/igarss55030.2025.11243381
摘要
This paper proposes an optical and SAR image fusion algorithm based on the Masked Autoencoder (MAE) to leverage the complementary strengths of both data types. Using a pre-trained MAE encoder, the algorithm extracts contour and texture features from optical and SAR images, fuses them separately, and combines the results. A fast IHS transformation restores color in the final fused image. Experiments demonstrate that the method preserves optical image details while integrating SAR structural information, reducing noise and improving fusion quality. Compared to traditional methods, the proposed approach excels in visual quality and information retention, offering a promising solution for remote sensing image fusion.
科研通智能强力驱动
Strongly Powered by AbleSci AI