Please considersing correctly locate image without the same size

Nobody has claimed this yet.

The is one CRITICAL issue, which I think has a severe negative impact on the use of the library:
Let's say, you want to find a small image from a big image, and there is really the same image in the big one when checking both through human's eyes, but the part in the big image is actually bigger or smaller than the small image.

Now the library really cannot find the specified image in the above situation in the current version, so people have to repeatedly and endlessly crop the same images whenever using the library in different devices. As a consequence, many people stop using the library just because of the problem and I believe that we can see obvious soaring in the library's use after addressing the issue.

However, according to my knowledge on OpenCV, its method named matchTemplate can be used to solve the problem by resizing one of both images first and I have noticed that SikuliX1 had also made some progress in the aspect such as the following:
https://github.com/RaiMan/SikuliX1/commit/2192b435d0c7aaa00bf0352b5f58cbc206bbd285

So, would you mind fix the problem as soon as possible?@rainmanwy

Start by reviewing the library’s current image-search behavior and compare it with OpenCV’s matchTemplate approach and the linked SikuliX1 commit. Determine how differently sized versions of the same image should be matched, then verify that the existing Robot Framework image-search flow finds those cases without requiring repeated crops.

Written by the indexing model from the issue text.