论文部分内容阅读
为了提高图像变化检测的准确率,缩短图像变化检测的检测时间,文章提出了一种新的合成孔径雷达(SAR)图像的变化检测算法。首先对实验图像进行对数变换,把图像中的乘性噪声转化成加性噪声。然后利用非线性各项异性扩散方程(PM扩散方程)的去噪算法对变换后的图像进行去噪,在对数域利用差值法获取差异图像。考虑到差异图像统计特性复杂,难以建立准确的模型,文章利用FCM算法对差异图像进行聚类,得到变化检测结果。实验结果表明,文章提出的检测算法具有较高的准确率和较短的检测的时间。
In order to improve the accuracy of image change detection and shorten the detection time of image change detection, a new change detection algorithm of Synthetic Aperture Radar (SAR) image is proposed. First of all, the experimental image is transformed logarithmically to convert the multiplicative noise in the image into additive noise. Then, denoising algorithm based on non-linear anisotropic diffusion equation (PM diffusion equation) is used to denoise the transformed image, and the difference image is obtained by difference method in logarithmic domain. Taking into account the complex statistical characteristics of the difference images, it is difficult to establish an accurate model. The article uses the FCM algorithm to cluster the difference images to obtain the change detection results. The experimental results show that the proposed algorithm has higher accuracy and shorter detection time.