save 1bit windows-bitmap
hello,
does anyone have information on how to save a 1Bit Windows-Bitmap file? (binary image)
especially what do i write in the RGBQUAD structure for the red green and blue values?
thanks!
It's pretty straightforward actually.
There are two RGBQUAD entries (palette), two times the usual 4 bytes: Blue, Green, Red, filler.
The image data contains 8 pixels per byte. Do not forget the padding of the lines to multiples of 4 bytes.
Here's my BMP save routine which works for all bit depths. It doesn't do any RLE however:
ImageData is a class for holding an image in memory. GetRowData gets a pointer to the start of a line, everything else should be pretty clear from the names.
There are two RGBQUAD entries (palette), two times the usual 4 bytes: Blue, Green, Red, filler.
The image data contains 8 pixels per byte. Do not forget the padding of the lines to multiples of 4 bytes.
Here's my BMP save routine which works for all bit depths. It doesn't do any RLE however:
bool Save( const GR::tChar* szFileName, GR::Graphic::ImageData* pData ){ GR::IO::FileStream MyFile; if ( !MyFile.Open( szFileName, IIOStream::OT_WRITE_ONLY ) ) { return false; } // BitmapFileHeader MyFile.WriteU16( 0x4d42 ); MyFile.WriteU32( 0 ); // Size of Bitmap File MyFile.WriteU32( 0 ); MyFile.WriteU32( 0 ); // Offset to Bitmap bits // BitmapInfoHeader MyFile.WriteU32( 40 ); // size of BITMAPINFOHEADER MyFile.WriteU32( pData->Width() ); MyFile.WriteU32( pData->Height() ); MyFile.WriteU16( 1 ); int iBits = pData->BitsProPixel(); if ( iBits == 15 ) { iBits = 16; } MyFile.WriteU16( iBits ); MyFile.WriteU32( 0 ); // BI_RGB MyFile.WriteU32( 0 ); // sizeof Image MyFile.WriteU32( 0 ); MyFile.WriteU32( 0 ); MyFile.WriteU32( 0 ); MyFile.WriteU32( 0 ); if ( pData->BitsProPixel() <= 8 ) { // Palette for ( size_t i = 0; i < pData->Palette()->Entries(); ++i ) { MyFile.WriteU8( pData->Palette()->Blue( i ) ); MyFile.WriteU8( pData->Palette()->Green( i ) ); MyFile.WriteU8( pData->Palette()->Red( i ) ); MyFile.WriteU8( 0 ); } } for ( int j = pData->Height() - 1; j >= 0; --j ) { MyFile.WriteBlock( pData->GetRowData( j ), pData->BytesPerLine() ); int iPadding = ( pData->BytesPerLine() % 4 ); while ( iPadding % 4 ) { MyFile.WriteU8( 0 ); iPadding++; } } MyFile.Flush(); GR::u32 dwSize = (GR::u32)MyFile.GetSize(); MyFile.SetPosition( 2, IIOStream::PT_SET ); MyFile.WriteU32( dwSize ); MyFile.SetPosition( 4, IIOStream::PT_CURRENT ); if ( pData->Palette() ) { MyFile.WriteU32( 14 + 40 + (GR::u32)pData->Palette()->Entries() * 4 ); } else { MyFile.WriteU32( 14 + 40 ); } MyFile.Close(); return true;}
ImageData is a class for holding an image in memory. GetRowData gets a pointer to the start of a line, everything else should be pretty clear from the names.
thank you!
so although it is only a 1bit image, it still needs to save as much as a full RGBA image? (4bytes per pixel?)
thanks!
so although it is only a 1bit image, it still needs to save as much as a full RGBA image? (4bytes per pixel?)
thanks!
Quote:Original post by ehmdjiiNo, it's 1 bit per pixel. So 8 pixels in 1 byte.
so although it is only a 1bit image, it still needs to save as much as a full RGBA image? (4bytes per pixel?)
However, the BMP file specification states that each scanline must end on a multiple of 4 bytes. So if your image is 10px wide, that's 2 bytes, so you'd need to have 2 bytes of padding at the end of each row in the file.
Quote:Original post by ehmdjii
thank you!
so although it is only a 1bit image, it still needs to save as much as a full RGBA image? (4bytes per pixel?)
thanks!
What Evil Steve said. Plus, the two RGBQUADs i mentioned are for the palette. Since the image data only allows for one bit per pixel you need to specify what color actually is used (there's not only black and white). This is the so-called palette.
This allows you to use less memory for the image. The image data only carries indices into that palette.
This topic is closed to new replies.
Advertisement
Popular Topics
Advertisement