摘要
提出了一种用于图像分类的变分模型,该模型结合正则化过程,可以较好地保持图像边缘信息,同时可以用于图像恢复。利用模拟和真实SAR图像的分类仿真试验表明,基于变分法的极化SAR图像分类方法不仅能够实现SAR图像的正确分类,克服SAR图像中相干斑噪声的影响,并且算法快速,易于实现。
In this paper, a variational model is presented for image classification combining with an edge-preserving regulation process. Simulation results on both synthetic and real SAR images show that the proposed method can be used to do SAR image classification correctly in despite of the influence of the speckle noise. Furthermore, this method is fast and easy to implement.
出处
《电波科学学报》
EI
CSCD
北大核心
2008年第4期736-739,共4页
Chinese Journal of Radio Science
关键词
合成孔径雷达
极化
图像分类
变分模型
synthetic aperture radar
polarimetric
image classify
variational model