基于AI的遥感影像条带噪声滤波方法研究  

Research on AI Based Band Noise Filtering Method for Remote Sensing Images

作　　者：江海宝 皮原征 JIANG Haibao;PI Yuanzheng(Guangdong Provincial Institute of Land and Resources Surveying and Mapping,Guangzhou 510700,China)

机构地区：[1]广东省国土资源测绘院,广东广州510700

出　　处：《电声技术》2024年第12期57-59,共3页Audio Engineering

摘　　要：研究基于人工智能(Artificial Intelligence,AI)的遥感影像条带噪声滤波方法,旨在提高条带噪声去除的精准度与效率。通过卷积神经网络(Convolutional Neural Networks,CNN)提取噪声特征,并结合生成对抗网络分离噪声与信号,利用强化学习优化滤波参数。实验表明,该方法在峰值信噪比、结构相似性等指标上均优于传统方法,特别是在不同噪声强度下保持了较高的图像质量。In order to improve the accuracy and efficiency of stripe noise removal,the stripe noise filtering method of remote sensing images based on Artificial Intelligence(AI) is studied.The noise features are extracted by Convolutional Neural Networks(CNN),and the noise and signal are separated by combining with the generated countermeasure network,the filtering parameters are optimized by reinforcement learning.Experiments show that this method is superior to traditional methods in terms of peak signal-to-noise ratio and structural similarity,especially in maintaining high image quality under different noise intensities.

关 键 词：遥感影像 条带噪声 深度学习 

分 类 号：TN7[电子电信—电路与系统] TP39[自动化与计算机技术—计算机应用技术]