sikuli-driver team mailing list archive

[RaiMan <question237289@xxxxxxxxxxxxxxxxxxxxx>]

Question #237289 on Sikuli changed:
https://answers.launchpad.net/sikuli/+question/237289

    Status: Open => Answered

RaiMan proposed the following answer:
yes seems to be some problem towards native level. (works on my system
with regular 1.0.1)

So i come back to my recommendation: make a tiny test program in C++ to test and debug the functions of VisionProxy.dll (see below about the principal structure)
highgui of openCV has some high-level functions to read and write image files and create CV::Mat from them.

- I have one more doubt. 
The image handling towards native level is done in class Finder, that feeds the FindInput in vision.cpp
- the screen capture goes to source (the image to search in)
- the image to search goes to target
both are converted to CV::Mat at different places

finder.cpp is the topmost level for the find operation and returns the
results in find_results.

currently we have some mix in the sense, that target images might be
read from the image file in vision.cpp.

Beginning with version 1.1 all images will only be given to the native
level as CV::Mat (which will make the use of cv_highgui obsolete), since
images will be cached in memory.

-- 
You received this question notification because you are a member of
Sikuli Drivers, which is an answer contact for Sikuli.