[request] matching should have more parameters to influence accuracy and robustness

[RaiMan]

Hello,

We are using SikuliX-1.1.0-Beta4 for automated GUI testing and we are very happy with its capabilities and features. We are using a default similarity level of 0.98, which is high enough to detect subtle gui differences. However, we've noticed, that when the compared images are black, there can be very big visual differences (as percent of the image) and still they are matched with very high similarity - sometimes exceeding our default level of 0.98!

The following two black images, for example, are matched with 0.982491: http://imgur.com/a/IjsDk
The following two gray images are not matched with 0.961927: http://imgur.com/a/9Yd9z

Is this expected behaviour? If so, do you have any idea how we can work-around this?

[RaiMan (raimund-hocke)]

more on this see related question.