bmp.htm

刚刚看到本站有Visual C++数字图象处理（人民邮电出版社）的电子书

    structure has the following form:<br>
    <br>
    <br>
    <br>
    typedef struct icondir {<br>
    word idreserved;<br>
    word idtype;<br>
    word idcount;<br>
    icondirentry identries[1];<br>
    } iconheader;<br>
    <br>
    following are the members in the icondir structure:<br>
    <br>
    idreserved reserved; must be zero. <br>
    idtype specifies the resource type. this member is set to 1. <br>
    idcount specifies the number of entries in the directory. <br>
    identries specifies an array of icondirentry structures containing<br>
    information about individual icons. the idcount member specifies the number<br>
    of structures in the array.<br>
    <br>
    the icondirentry structure specifies the dimensions and color format for an<br>
    icon. the structure has the following form:<br>
    <br>
    <br>
    <br>
    struct icondirectoryentry {<br>
    byte bwidth;<br>
    byte bheight;<br>
    byte bcolorcount;<br>
    byte breserved;<br>
    word wplanes;<br>
    word wbitcount;<br>
    dword dwbytesinres;<br>
    dword dwimageoffset;<br>
    };<br>
    <br>
    following are the members in the icondirentry structure: <br>
    <br>
    bwidth specifies the width of the icon, in pixels. acceptable values<br>
    are 16, 32, and 64.<br>
    <br>
    bheight specifies the height of the icon, in pixels. acceptable<br>
    values are 16, 32, and 64.<br>
    <br>
    bcolorcount specifies the number of colors in the icon. acceptable values<br>
    are 2, 8, and 16.<br>
    <br>
    breserved reserved; must be zero. <br>
    wplanes specifies the number of color planes in the icon bitmap. <br>
    wbitcount specifies the number of bits in the icon bitmap. <br>
    dwbytesinres specifies the size of the resource, in bytes. <br>
    dwimageoffset specifies the offset, in bytes, from the beginning of the<br>
    file to the icon image.<br>
    <br>
    icon image<br>
    <br>
    each icon-resource file contains one icon image for each image identified in<br>
    the icon directory. an icon image consists of an icon-image header, a color<br>
    table, an xor mask, and an and mask. the icon image has the following form:<br>
    <br>
    <br>
    <br>
    bitmapinfoheader icheader;<br>
    rgbquad iccolors[];<br>
    byte icxor[];<br>
    byte icand[];<br>
    <br>
    the icon-image header, defined as a bitmapinfoheader structure, specifies the<br>
    dimensions and color format of the icon bitmap. only the bisize through<br>
    bibitcount members and the bisizeimage member are used. all other members<br>
    (such as bicompression and biclrimportant) must be set to zero.<br>
    <br>
    the color table, defined as an array of rgbquad structures, specifies the<br>
    colors used in the xor mask. as with the color table in a bitmap file, the<br>
    bibitcount member in the icon-image header determines the number of elements<br>
    in the array. for more information about the color table, see section 1.1,<br>
    &quot;bitmap-file formats.&quot;<br>
    <br>
    the xor mask, immediately following the color table, is an array of byte<br>
    values representing consecutive rows of a bitmap. the bitmap defines the<br>
    basic shape and color of the icon image. as with the bitmap bits in a bitmap<br>
    file, the bitmap data in an icon-resource file is organized in scan lines,<br>
    with each byte representing one or more pixels, as defined by the color<br>
    format. for more information about these bitmap bits, see section 1.1,<br>
    &quot;bitmap-file formats.&quot;<br>
    <br>
    the and mask, immediately following the xor mask, is an array of byte values,<br>
    representing a monochrome bitmap with the same width and height as the xor<br>
    mask. the array is organized in scan lines, with each byte representing 8<br>
    pixels.<br>
    <br>
    when windows draws an icon, it uses the and and xor masks to combine the icon<br>
    image with the pixels already on the display surface. windows first applies<br>
    the and mask by using a bitwise and operation; this preserves or removes<br>
    existing pixel color. windows then applies the xor mask by using a bitwise<br>
    xor operation. this sets the final color for each pixel.<br>
    <br>
    the following illustration shows the xor and and masks that create a<br>
    monochrome icon (measuring 8 pixels by 8 pixels) in the form of an uppercase<br>
    k:<br>
    <br>
    windows icon selection<br>
    <br>
    windows detects the resolution of the current display and matches it against<br>
    the width and height specified for each version of the icon image. if windows<br>
    determines that there is an exact match between an icon image and the current<br>
    device, it uses the matching image. otherwise, it selects the closest match<br>
    and stretches the image to the proper size.<br>
    <br>
    if an icon-resource file contains more than one image for a particular<br>
    resolution, windows uses the icon image that most closely matches the color<br>
    capabilities of the current display. if no image matches the device<br>
    capabilities exactly, windows selects the image that has the greatest number<br>
    of colors without exceeding the number of display colors. if all images<br>
    exceed the color capabilities of the current display, windows uses the icon<br>
    image with the least number of colors.<br>
    <br>
    <br>
    <br>
    cursor-resource file format<br>
    <br>
    a cursor-resource file contains image data for cursors used by windows<br>
    applications. the file consists of a cursor directory identifying the number<br>
    and types of cursor images in the file, plus one or more cursor images. the<br>
    default filename extension for a cursor-resource file is .cur.<br>
    <br>
    cursor directory<br>
    <br>
    each cursor-resource file starts with a cursor directory. the cursor<br>
    directory, defined as a cursordir structure, specifies the number of cursors<br>
    in the file and the dimensions and color format of each cursor image. the<br>
    cursordir structure has the following form:<br>
    <br>
    <br>
    typedef struct _cursordir {<br>
    word cdreserved;<br>
    word cdtype;<br>
    word cdcount;<br>
    cursordirentry cdentries[];<br>
    } cursordir;<br>
    <br>
    following are the members in the cursordir structure: <br>
    <br>
    cdreserved reserved; must be zero. <br>
    cdtype specifies the resource type. this member must be set to 2. <br>
    cdcount specifies the number of cursors in the file. <br>
    cdentries specifies an array of cursordirentry structures containing<br>
    information about individual cursors. the cdcount member specifies the number<br>
    of structures in the array.<br>
    <br>
    a cursordirentry structure specifies the dimensions and color format of a<br>
    cursor image. the structure has the following form:<br>
    <br>
    <br>
    <br>
    typedef struct _cursordirentry {<br>
    byte bwidth;<br>
    byte bheight;<br>
    byte bcolorcount;<br>
    byte breserved;<br>
    word wxhotspot;<br>
    word wyhotspot;<br>
    dword lbytesinres;<br>
    dword dwimageoffset;<br>
    } cursordirentry;<br>
    <br>
    following are the members in the cursordirentry structure: <br>
    <br>
    bwidth specifies the width of the cursor, in pixels. <br>
    bheight specifies the height of the cursor, in pixels. <br>
    bcolorcount reserved; must be zero. <br>
    breserved reserved; must be zero.<br>
    wxhotspot specifies the x-coordinate, in pixels, of the hot spot. <br>
    wyhotspot specifies the y-coordinate, in pixels, of the hot spot. <br>
    lbytesinres specifies the size of the resource, in bytes. <br>
    dwimageoffset specifies the offset, in bytes, from the start of the file to<br>
    the cursor image.<br>
    <br>
    cursor image<br>
    <br>
    each cursor-resource file contains one cursor image for each image identified<br>
    in the cursor directory. a cursor image consists of a cursor-image header, a<br>
    color table, an xor mask, and an and mask. the cursor image has the following<br>
    form:<br>
    <br>
    <br>
    <br>
    bitmapinfoheader crheader;<br>
    rgbquad crcolors[];<br>
    byte crxor[];<br>
    byte crand[];<br>
    <br>
    the cursor hot spot is a single pixel in the cursor bitmap that windows uses<br>
    to track the cursor. the crxhotspot and cryhotspot members specify the x- and<br>
    y-coordinates of the cursor hot spot. these coordinates are 16-bit integers.<br>
    <br>
    the cursor-image header, defined as a bitmapinfoheader structure, specifies<br>
    the dimensions and color format of the cursor bitmap. only the bisize through<br>
    bibitcount members and the bisizeimage member are used. the biheight member<br>
    specifies the combined height of the xor and and masks for the cursor. this<br>
    value is twice the height of the xor mask. the biplanes and bibitcount<br>
    members must be 1. all other members (such as bicompression and<br>
    biclrimportant) must be set to zero.<br>
    <br>
    the color table, defined as an array of rgbquad structures, specifies the<br>
    colors used in the xor mask. for a cursor image, the table contains exactly<br>
    two structures, since the bibitcount member in the cursor-image header is<br>
    always 1.<br>
    <br>
    the xor mask, immediately following the color table, is an array of byte<br>
    values representing consecutive rows of a bitmap. the bitmap defines the<br>
    basic shape and color of the cursor image. as with the bitmap bits in a<br>
    bitmap file, the bitmap data in a cursor-resource file is organized in scan<br>
    lines, with each byte representing one or more pixels, as defined by the<br>
    color format. for more information about these bitmap bits, see section 1.1,<br>
    &quot;bitmap-file formats.&quot;<br>
    <br>
    the and mask, immediately following the xor mask, is an array of byte values<br>
    representing a monochrome bitmap with the same width and height as the xor<br>
    mask. the array is organized in scan lines, with each byte representing 8<br>
    pixels.<br>
    <br>
    when windows draws a cursor, it uses the and and xor masks to combine the<br>
    cursor image with the pixels already on the display surface. windows first<br>
    applies the and mask by using a bitwise and operation; this preserves or<br>
    removes existing pixel color. window then applies the xor mask by using a<br>
    bitwise xor operation. this sets the final color for each pixel.<br>
    <br>
    the following illustration shows the xor and the and masks that create a<br>
    cursor (measuring 8 pixels by 8 pixels) in the form of an arrow:<br>
    <br>
    following are the bit-mask values necessary to produce black, white,<br>
    inverted, and transparent results:<br>
    <br>
    pixel result and maskxor mask<br>
    <br>
    black 0 0 <br>
    white 0 1 <br>
    transparent 1 0 <br>
    inverted1 1 <br>
    <br>
    windows cursor selection<br>
    <br>
    if a cursor-resource file contains more than one cursor image, windows<br>
    determines the best match for a particular display by examining the width and<br>
    height of the cursor images.<br>
    <br>
    <br>
    ==============================================================================<br>
    <br>
    <br>
    bitmapfileheader (3.0)<br>
    <br>
    <br>
    <br>
    typedef struct tagbitmapfileheader { /* bmfh */<br>
    uint bftype;<br>
    dword bfsize;<br>
    uint bfreserved1;<br>
    uint bfreserved2;<br>
    dword bfoffbits;<br>
    } bitmapfileheader;<br>
    <br>
    the bitmapfileheader structure contains information about the type, size, and<br>