asked on

biWidth wrong in tagBITMAPINFOHEADER when using twain

I use the twain driver to scan an image. Then I read the raw image data i get a wrong width for the following dpi-resolutions:
96 dpi -> the width is to low by 3
100 dpi -> the width is to low by 2

I have attached an image to show this.

On higher Resolutions (150, 300, 360dpi) it always works.

What did I wrong? Is there a differnet way to get the hight and width of an hDIB?

The strange thing is it that these resolutions work perfect in other programms (e.g. IrfanView)

Please Help

Greedings Sonja

bool CScanDialog::PDFAddPicturePage(HDIB hDIB)
{
  HPDF_Page  page;
  HPDF_Image image;
 
  BYTE * BGRimage;
  BGRimage = (BYTE *) GlobalLock(hDIB);
 
  /*
    typedef struct tagBITMAPINFOHEADER{
      DWORD  biSize; 
      LONG   biWidth; 
      LONG   biHeight; 
      WORD   biPlanes; 
      WORD   biBitCount; 
      DWORD  biCompression; 
      DWORD  biSizeImage; 
      LONG   biXPelsPerMeter; 
      LONG   biYPelsPerMeter; 
      DWORD  biClrUsed; 
      DWORD  biClrImportant; 
    } BITMAPINFOHEADER, *PBITMAPINFOHEADER; 
  */
 
  long img_width    = * (BGRimage + 1 * sizeof(DWORD) + 3); img_width *= 0x100;
       img_width   += * (BGRimage + 1 * sizeof(DWORD) + 2); img_width *= 0x100;
       img_width   += * (BGRimage + 1 * sizeof(DWORD) + 1); img_width *= 0x100;
       img_width   += * (BGRimage + 1 * sizeof(DWORD)    );
  long img_height   = * (BGRimage + 1 * sizeof(DWORD) + 1 * sizeof(long) +3);img_height *= 0x100;
       img_height  += * (BGRimage + 1 * sizeof(DWORD) + 1 * sizeof(long) +2);img_height *= 0x100;
       img_height  += * (BGRimage + 1 * sizeof(DWORD) + 1 * sizeof(long) +1);img_height *= 0x100;
       img_height  += * (BGRimage + 1 * sizeof(DWORD) + 1 * sizeof(long) +0);
  DWORD img_header_length  = * (BGRimage +1); img_header_length *= 0x100;
        img_header_length += * (BGRimage);
 
  BGRimage += img_header_length; // Offset of the header
  BGRimage += (0xFF * 4);        // Offset of the color table
  /*Here some  code that writes the image to a pdf file*/
}

Open in new window

Scan.JPG

jkr

In the above code, there's no reference to any DPI values nor are you setting 'biWidth'. Could you show the complete code and explain the calculations in the above snippet?

Sonja_M

ASKER

Hi jkr!

To do a scan you simply have to set some parameters for the scanner and then start the scan. The DPI-Value is stored in

#define SCANNER_DPI 96

The code above is run after the scan. I get a handle to a hDIB-Image and try to use this data. The follwing is not possible, because the compile can not convert void to tagBITMAPINFOHEADER:

tagBITMAPINFOHEADER * MyDibHead = BGRimage;

So what I did is to read the raw data in line 25-28 from biWidth. Later on I do not need the dpi anymore. I just need the pixels, but I only ant to use around 100dpi then scanning.

I hope that got clearer.

Greedings

Sonja

SOLUTION

jkr

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial

ASKER CERTIFIED SOLUTION

alb66

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial

Sonja_M

ASKER

Thank you very much, you both helped me a lot.

@jkr: First I tryed to cast the header - but i alway got a "can not cast"-error. Solution 1 worked (i thought I tryed this already).

@alb66: Where did you got this from? It was the solution - of course - now it works perfect. I have a white extra line now sometimes but that does not bother me. I was more afraid that a differend scanner sends me the picture with one pixel more.

Here is my final code (got much shorter):

bool CScanDialog::PDFAddPicturePage(HDIB hDIB)
{
BYTE * BGRimage;
BGRimage = (BYTE *) GlobalLock(hDIB);

BITMAPINFOHEADER * MyDibHead = (BITMAPINFOHEADER *) BGRimage;

long img_height = MyDibHead->biHeight;
long img_width = (((MyDibHead->biWidth + 3) / 4) * 4);

BGRimage += MyDibHead->biSize;
/*Here some code that writes the image to a pdf file*/
}

alb66

For example, you can find that information in the BITMAPINFO documentation:
http://msdn.microsoft.com/en-us/library/ms532284(VS.85).aspx

<<<<< A DIB consists of two distinct parts: a BITMAPINFO structure describing the dimensions and colors of the bitmap, and an array of bytes defining the pixels of the bitmap. The bits in the array are packed together, but each scan line must be padded with zeroes to end on a LONG data-type boundary. >>>>>