论文部分内容阅读
提出了一种基于多分辨率小波分析和高斯 -马尔可夫随机场理论的边缘图像融合方法。在图像的多分辨率小波分析的基础上 ,基于高斯 -马尔可失随机场理论 ,在不同图像的对应区域 ,用回归分析的方法分别提取一组统计参数 ,这些参量表征了图像的局部结构特征 ,计算其相似性测度 ,最后由输入图像及其特征的相似性矩阵生成融合后的边缘图像。实验证明 ,这种融合方法具有较强的适应性和可靠性 ,即使在图像信噪比较低的情况下 ,也能取得较好的融合效果。
An edge image fusion method based on multi-resolution wavelet analysis and Gaussian-Markov random field theory is proposed. Based on the multi-resolution wavelet analysis of images, a set of statistical parameters are respectively extracted from the corresponding regions of different images by means of regression analysis based on Gaussian-Markov random field theory. These parameters represent the local structural features of the image , The similarity measure is calculated, and finally the fused edge image is generated from the similarity matrix of the input image and its features. Experiments show that this fusion method has strong adaptability and reliability, even in the case of low signal-to-noise ratio, but also to obtain a better fusion effect.