Reverse projection of histograms

1. What is the role of reverse projection?
Reverse projection is used to find the most matched point or area of a specific image (usually smaller or only 1 pixel, which is called a template image) in an input image (usually larger, that is, the positioning template image appears at the position of the input image.
2. How to find (work) reverse projection )?
The search method is to continuously cut image blocks with the same size as the template image in the input image, and compare them with the template image by histogram comparison.

suppose we have a 100x100 input image and a 10x10 template image. The search process is as follows:
(1) cut a temporary image (0, 0) to (10, 10) from the upper left corner of the input image;
(2) generate the histogram of the temporary image;
(3) use the histogram of the temporary image and the histogram of the template image for comparison. The comparison result is C;
(4) histogram comparison result C is the pixel value at the result image (0, 0);
(5) cut the temporary image from (0, 1) to (10, 11), compare the histogram, and record the result image;
(6) repeated (1 )~ (5) step until the bottom right corner of the input image.

 

(This image is referenced fromHttp://www.opencv.org.cn)
3. What is the result of reverse projection?
The results of reverse projection include the histogram comparison results starting from each input image pixel. We can regard it as a two-dimensional floating point array, a two-dimensional matrix, or a single-channel floating point image.
4. What is special?
If the input image is as big as the template image, reverse projection is equivalent to histogram comparison. If the input image is smaller than the template image, you can directly strike ~~.