A3LGDIPlus / converting bitmaps to arrays

[Oldschool]

There is an open source program called Openscrape written C/C++ that uses an "in memory" OCR technique of interest. Particularly they are using an Alpha Channel separation technique to drop an image background, and obtain a clean foreground.

The color cube specifications (as in the C transform) are used to separate the foreground from the background pixels.

// Load entire foreground pixel array into memory
      ::ZeroMemory(&bmi, sizeof(BITMAPINFO));
      bmi.bmiHeader.biSize       = sizeof(BITMAPINFOHEADER);
      bmi.bmiHeader.biWidth     = width;
      bmi.bmiHeader.biHeight       = -height;
      bmi.bmiHeader.biBitCount   = 32;
      bmi.bmiHeader.biPlanes       = 1;
      bmi.bmiHeader.biCompression  = BI_BITFIELDS;
      bmi.bmiHeader.biSizeImage = width * height * 4;

      pBits = ::new BYTE[bmi.bmiHeader.biSizeImage];
      ::ZeroMemory(pBits, bmi.bmiHeader.biSizeImage * sizeof(BYTE));

      hbm = (HBITMAP) GetCurrentObject(hdc, OBJ_BITMAP);
      ::GetDIBits(hdc, hbm, 0, height, pBits, &bmi, DIB_RGB_COLORS);

      for (x = 0; x < width; x++) {
         for (y = 0; y < height; y++) {
            alpha = pBits[y*width*4 + x*4 + 3];
            red = pBits[y*width*4 + x*4 + 2];
            green = pBits[y*width*4 + x*4 + 1];
            blue = pBits[y*width*4 + x*4 + 0];

I'm pretty sure this can be done with A3L, just not entirely sure how at this point.

Maybe ImageMagick will be required...

http://www.autoitscript.com/forum/index.ph...52178&st=15

Does anyone have any ideas on how to do this in AutoIt?

[Zedna]

I'm pretty sure this can be done with A3L, just not entirely sure how at this point.

Maybe ImageMagick will be required...

http://www.autoitscript.com/forum/index.ph...52178&st=15

Does anyone have any ideas on how to do this in AutoIt?

Just note: All A3L (Auto3Library) functionality was ported to standard UDF in AutoIt 3.2.10