论文部分内容阅读
SAR图像的变化检测技术是其应用于动态监测的核心,不同尺度的SAR图像会表示出不同细节程度的图像特征,为了得到更精确有效的变化检测结果,必须在图像处理过程中引入尺度概念。本文以EKLD法和比值法为主对比研究了图像尺度空间和样本窗口尺度对变换检测方法的影响,在此基础上提出一种多尺度渐进的变化检测方法,实现了对SAR图像变化检测快速而精确的处理。通过实验表明该方法在准确率和算法效率上具有优势,有一定的实用价值。
SAR image change detection technology is the core of its application to dynamic monitoring. SAR images of different scales show different details of image features. In order to obtain more accurate and effective change detection results, the concept of scale must be introduced in image processing. In this paper, we mainly study the influence of image scale space and sample window scale on the transform detection method based on the EKLD method and the ratio method. On this basis, a multi-scale progressive change detection method is proposed to detect the change of the SAR image quickly, Precise handling. Experiments show that this method has some advantages in accuracy and efficiency, and it has some practical value.