Normalized cross-correlation Normalization cross correlation (NCC)

NCC as its name, is used to describe the degree of correlation of two goals, that can be used to describe the similarity between the target. NCC will generally be used for image matching, i.e. a search for the highest matching image corresponding region of NCC as a small area known, and then align the entire image.

Suppose two images , then the NCC can be calculated as follows:

Which represents a point multiplication.

For example, the following two images, we want to align

enter description here

 

enter description here

First, the selected area has two images as one image in templete

enter description here

And select the region including templete another image

enter description here
The final results are aligned

enter description here

 

Guess you like

Origin blog.csdn.net/qq_41598072/article/details/91441098