Skip to content

New feature request : Find image in image #110

@neoarc

Description

@neoarc

A short and descriptive title

  • Find image in image

A detailed description of the feature

  • If we have 2 image objects, we can test imageA is part of imageB.
  • If imageA contains imageB, returns relative coordinates (or absolute coords)

A few typical use case scenarios

  • See this pseudo code
    Screen.grab(imageA, ...whole screen...)
    Screen.grab(imageB, ...image of some button...) // bad-way, i will request this issue separately
    var coord = new Coord;
    if (imageA.findImage(imageB, coord) == true) {
    robot.Mouse().click(coord); // click found image
    }

How you expect the feature to work

  • Must be very fast
  • Don't require to turn Aero off

Any other useful information

  • This implementation provides the easiest way to automate UI automation, making it easier for end users who do not know programming to easily access them.
  • Comparing image buffers directly to pixels by pixel is much faster than expected.
  • This is the method used by AutoHotkey, NaMacroJS (My own javascript platform similar to robot-js).

PS. Sorry, I'm finally robot-js user, not robot native. But I'm a c ++ developer.

Metadata

Metadata

Assignees

No one assigned

    Labels

    RequestA cool feature request for the library

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions