X射线成像可以获得待测物体内部清晰的组织结构，但是X射线无法对物体边缘精准成像。为了得到待测物体更完整的图像，需要引入可见光成像技术，将两种图像融合。本文首先用交互式配准方法及仿射变换对采集到的可见光图像与X射线图像进行配准，然后对可见光图像进行边缘检测提取图像边缘，再采用像素加权平均进行融合，分别计算融合前后的图像的熵并比较。得出融合图像的熵比可见光及X射线图像的熵值小，说明可见光图像的边缘代替X射线图像的边缘导致融合图像灰度值减小。
X-ray imaging provides a clear
organizational structure inside the object to be tested, but X-rays do not
accurately image the edge of the object. In order to obtain a more complete
image of the object to be measured, visible light imaging technology needs to
be introduced to fuse the two images. This paper first uses interactive
registration method and affine transformation to match the captured visible
light image with the X-ray image, then uses the visible light image edge
detection to extract the image edge, then uses the pixel-weighted average
fusion, calculates the entropy of the image before and after fusion and
compares it respectively. It is obtained that the entropy of fusion image is
smaller than that of visible light and X-ray image, which indicates that the
edge of visible light image fusion X-ray image causes the grayscale value of
fusion image to decrease.
