Question -- how could I go about programming something to visually "look" around the screen input of the computer it's running on.
Example -- the program runs, switches to an open internet browser, and then "watches" a pre-defined area of pixels, comparing pictures from one frame to the next, to determine if some trigger has been pulled. Let's say a small pre-defined box has been set around the area where the progress icon shows the page loading (when window is maximized). Is there some type of class or function that natively implements something like this?
And to further the question, if there is a function/class like this (in any language), is there a way to "spider crawl" through the screen contents, "looking" for that image?
A Digital Eye
Started by Coldblackice, Mar 21 2009 07:33 PM
4 replies to this topic
#1
Posted 21 March 2009 - 07:33 PM
|
|
|
#2
Posted 23 March 2009 - 06:21 AM
It depends on the OS you are using.
#3
Guest_Jordan_*
Posted 23 March 2009 - 03:29 PM
Guest_Jordan_*
Would you be watching the whole browser window or just a section? What if they moved the window out of your section of pixels?
#4
Posted 23 March 2009 - 10:57 PM
this function is OS specific, however most provide both a method of capturing screen input and getting the absolute position of a relative position inside a window. If you want to do anything complex, you'll need to get into image segmentation which would not be easy. Howver if you do try, I recomend the histogram approach (ends up fast and fairly accurate)
Edit: god **** iPod keyboard
Edit: god **** iPod keyboard
#5
Posted 24 March 2009 - 01:47 PM
TkTech said:
this function is OS specific, however most provide both a method of capturing screen input and getting the absolute position of a relative position inside a window. If you want to do anything complex, you'll need to get into image segmentation which would not be easy. Howver if you do try, I recomend the histogram approach (ends up fast and fairly accurate)
Edit: god **** iPod keyboard
Edit: god **** iPod keyboard
hmm... in response to watching a pre-determined section of screen vs. watching a specific window -- either/or. I don't need to use this program on anyone elses computer other than my own, so the browser window would always be maximized and thus wouldn't have to worry about it moving around.
Where do I find documentation or how-to on using these OS specific functions? I don't know how complex I would need it. Do you know much about the functions? Can they compare a selected screen capture to an image (say a jpg) and then make a judgement on correlation?


Sign In
Create Account

Back to top









