OCR text identifing in scar?

how would I go upon using scar to search for the font on a webpage , and return what it typed in plain text on the debug window?

suppose the page had Hello World in time new roman, but it was an image. How would i go about using SCAR to identify what the text says and show it in the debug window?

HTML:

program FontTest;
const

  fontname      = 'Tahoma';
  size          = 16;
  bold          = false;
  italic        = false;
  underline     = false;
  strikethrough = false;
var
  font, bitmap, sizex, sizey: integer;
begin
  //Load the font into memory
  font := LoadCharsFromFont2(fontname, size, bold, italic, underline, strikethrough);
 
  //Generate a bitmap mask from the previously loaded font
  bitmap := CreateBitmapMaskFromText('Hello', font);
 end.
 

This is the code so far I made, how do I identify the text on a website?

 

GetPage('');

Will return everything on that page has a string. You can use Between(''); to shrink it down to what yoy want.

 

thanks, however GetPage only returns HTTP webpage contents as string. However, the text i want to find is in an image.

 

Code:

SafeCopyCanvas(GetClientCanvas, GetBitmapCanvas(Bitmap), 0, 0, 0, 0, 0, 0, 0, 0);

 

thanks, but as me being a scar noob, im not sure how to implement this on to my existing script.

Would I have to make a a bitmap of each letter to make

SafeCopyCanvas(GetClientCanvas, GetBitmapCanvas(Bitmap), 0, 0, 0, 0, 0, 0, 0, 0);

work?

ohyeah, if I didn't mention this, the Hello World image changes words every time I goto that webpage. Like sometime it will say hello world, sometime hi, etc.

 

No, set the coordinates of where to pick the picture off the canvas and then the coordinates to copy that onto the bitmap.

You will have to make a bitmap of each phrase and then compare the two bitmaps to see if they are the same. So. . . Make them into an array and then use For etc Do etc if (etc = etc) then break; etc.