Reputation: 165
I need to write a text recognition (from image) application. The main idea is that while my application is running, I may have the need to transform some text of an image into manageble text. So, in this case with the mouse I need to draw a square around the area that i need to capture and the software must convert the content of the extracted picture into text.
I solved the problem of image recognition. I also find a very easy way to capture from screen.
What I need to do now is be able to select with the mouse the interesting area that is over an other running opplication (for example over a webpage or over an image opened in Paint). That must be like the screencapture on windows7, you create a selection of the screen and this is saved like a picture.
By looking around, I didn't find anything and I don't knwo where to start.
Many thanks
Upvotes: 0
Views: 738
Reputation: 5137
You can achieve that using either a 'tricky' easy way or a real but difficult approach.
The tricky way
Screen Recorder applications usually use this approach: Whenever user wants to select an area, you display a full screen border less Window with 0% opacity, then user attempts to select the screen area, and he is actually selecting your Window area, so you can receive mouse events and display/draw a rectangular shape to show the selection area to the user.
In this approach, the program needs to know when to display the Window and when to hide it. This can be done by for example defining Hotkeys for capturing:
The real way
Using this way, you need to set a message hook in order to receive mouse and keyboard events while user is interacting with desktop not your program. This is not an easy to accomplish approach and I recommend you the first one.
Upvotes: 1