Bitmap Header Types

The bitmap has four basic header types:

• BITMAPCOREHEADER
• BITMAPINFOHEADER
• BITMAPV4HEADER
• BITMAPV5HEADER

The four types of bitmap headers are differentiated by the Size member, which is the first DWORD in each of the structures.

The BITMAPV5HEADER structure is an extended BITMAPV4HEADER structure, which is an extended BITMAPINFOHEADER structure. However, the BITMAPINFOHEADER and BITMAPCOREHEADER have only the Size member in common with other bitmap header structures.

The BITMAPCOREHEADER and BITMAPV4HEADER formats have been superseded by BITMAPINFOHEADER and BITMAPV5HEADERformats, respectively. The BITMAPCOREHEADER and BITMAPV4HEADER formats are presented for completeness and backward compatibility.

The format for a DIB is the following (for more information, see Bitmap Storage ):

• a BITMAPFILEHEADER structure
• either a BITMAPCOREHEADER, a BITMAPINFOHEADER, a BITMAPV4HEADER, or a BITMAPV5HEADER structure.
• an optional color table, which is either a set of RGBQUAD structures or a set of RGBTRIPLE structures.
• the bitmap data
• optional Profile data

A color table describes how pixel values correspond to RGB color values. RGB is a model for describing colors that are produced by emitting light.

Profile data refers to either the profile file name (linked profile) or the actual profile bits (embedded profile). The file format places the profile data at the end of the file. The profile data is placed just after the color table (if present). However, if the function receives a packed DIB, the profile data comes after the bitmap bits, like in the file format.

Profile data will only exist for BITMAPV5HEADER structures where bV5CSType is PROFILE_LINKED or PROFILE_EMBEDDED. For functions that receive packed DIBs, the profile data comes after the bitmap data.

A palettized device is any device that uses palettes to assign colors. The classic example of a palettized device is a display running in 8 bit color depth (that is, 256 colors). The display in this mode uses a small color table to assign colors to a bitmap. The colors in a bitmap are assigned to the closest color in the palette that the device is using. The palettized device does not create an optimal palette for displaying the bitmap; it simply uses whatever is in the current palette. Applications are responsible for creating a palette and selecting it into the system. In general, 16-, 24-, and 32-bits-per-pixel (bpp) bitmaps do not contain color tables (a.k.a. optimal palettes for the bitmap); the application is responsible for generating a optimal palette in this case. However, 16-, 24-, and 32-bpp bitmaps can contain such optimal color tables for displaying on palettized devices; in this case the application just needs to create a palette based off the color table present in the bitmap file.

Bitmaps that are 1, 4, or 8 bpp must have a color table with a maximum size based on the bpp. The maximum size for 1, 4, and 8 bpp bitmaps is 2 to the power of the bpp. Thus, a 1 bpp bitmap has a maximum of two colors, the 4 bpp bitmap has a maximum of 16 colors, and the 8 bpp bitmap has a maximum of 256 colors.

Bitmaps that are 16-, 24-, or 32-bpp do not require color tables, but may have them to specify colors for palettized devices. If a color table is present for 16-, 24-, or 32-bpp bitmap, the biClrUsed member specifies the size of the color table and the color table must have that many colors in it. If biClrUsed is zero, there is no color table.

The red, green, and blue bitfield masks for BI_BITFIELD bitmaps immediately follow the BITMAPINFOHEADER, BITMAPV4HEADER, and BITMAPV5HEADER structures. The BITMAPV4HEADER and BITMAPV5HEADER structures contain additional members for red, green, and blue masks as follows.

When the biCompression member of BITMAPINFOHEADER is set to BI_BITFIELDS and the function receives an argument of type LPBITMAPINFO, the color masks will immediately follow the header. The color table, if present, will follow the color masks. BITMAPCOREHEADER bitmaps do not support color masks.

By default, bitmap data is bottom-up in its format. Bottom-up means that the first scan line in the bitmap data is the last scan line to be displayed. For example, the 0th pixel of the 0th scan line of the bitmap data of a 10-pixel by 10-pixel bitmap will be the 0th pixel of the 9th scan line of the displayed or printed image. Run-length encoded (RLE) format bitmaps and BITMAPCOREHEADER bitmaps cannot be top-down bitmaps. The scan lines are DWORD aligned, except for RLE-compressed bitmaps. They must be padded for scan line widths, in bytes, that are not evenly divisible by four, except for RLE compressed bitmaps. For example, a 10- by 10-pixel 24-bpp bitmap will have two padding bytes at the end of each scan line.