摘要
针对红外与可见光图像的配准问题,在传统的基于图像边缘相关匹配配准的基础上,提出一种改进的方法。该方法首先手动选取待配准图像与参考图像中包含同一目标的子图像;接着利用Canny算子提取子图像的边缘和子图像的边缘相关性,求取出子图像间的仿射变换模型参数;再进一步修正该参数,得到原图像之间的仿射变换模型;然后对待配准图像进行仿射变换及双线性插值,从而实现图像配准。实验结果证明,该方法较之传统的利用图像边缘相关性的配准方法,提高了图像配准的速度,具有较高的实时性和应用价值。
A new method based on the correlation of sub-image and affine transformation is presented in order to realize the registration of infrared image and visual image. It has improved traditional image registration method based on the edge correlation of images. In the method,firstly,croping image manually and getting the sub- image from the reference image and image prepared for registration. The sub - image contains the same target of the scene. Secondly,extracting the edges of sub - image by Canny operator,and calculating parameters of affine transformation model between sub- images by edge correlation. Thirdly,further amending the parameters to get affine transformation model between original images,and carrying out affine transformation and bilinear interpolation in the image prepared for registration to realize image registration. The results prove that the method has a faster image registration speed compared to traditional method based on edge correlation,and have better practical value.
出处
《现代电子技术》
2009年第10期104-106,共3页
Modern Electronics Technique
关键词
子图像
边缘检测
相关匹配
图像配准
sub - image
edge detection
correlation matching
image registration
