image toolkit box Victor Library

Victor Image Processing Library

Image File Formats


The Victor Image Processing Library is an image toolkit for creating applications that read and write raster image file formats and can convert images from one format to another. Images can represent documents, black & white photos, color photos, and scanned images. Victor is small, fast, and easy to use. More Victor info.

File Types:

BMP | DIB | EPS | GIF | JFIF| JPEG| PCX | PNG | TGA | TIFF | BIF | other

Compression schemes:

The Victor Library gives your application the ability to convert an image from any file type to any other file type . With color reduction, color promotion, bit conversion, scale to gray, and other techniques you can alter pixel depth (bits per pixel) and palette size.

BMP is the Microsoft Windows format for bitmapped images of 1-, 4-, 8-, 16-, 24-, or 32-bits pixel depth. The pixel data can be stored raw or RLE compressed.

Specification: Microsoft Windows Programmer's Reference

DIB stands for device independent bitmap, this is an alias for BMP, above.


EPS stands for Encapsulated PostScript. This format is used for embedding a grayscale or color image in a PostScript document or for printing on a PostScript printer.

Specification: #5002, Encapsulated PostScript File Format Specification Version 3.0 Adobe Systems.

GIF stands for Graphic Interchange Format. This format was developed by Compuserve. GIF images may have 1- to 8-bits per pixel. Animated GIF files contain multiple images, each contained as an individual frame.

Compressed GIF files use the LZW compression. LZW compression and decompression are completely enabled in the Victor Image Processing Library.



JPEG stands for Joint Photographic Experts Group, and is a compression standard rather that a file format. Images can be JPEG-compressed and stored in *.JPG or *.TIF files. The Victor Library loads and saves both file types.

JPEG is used for 8-bit and 24-bit images. The JPEG compression scheme is highly efficient but is so at the expense of image quality. This is a "lossy" compression which means that image data are always discarded during compression.

The JPEG compression follows this sequence:

  1. Converts RGB data to the YCC color space. Y, luminence, is similar to value in HSV colorspace. C and C are chrominance values that hold the color information.
  2. Usually subsamples the CC data (by averaging adjacent pixels) so the amount of color data is one quarter of the original.
  3. Sends the YCC data through a DCT (discrete cosine transform) to generate coefficients in the frequency domain.
  4. Quantitizes the DCT coefficients. A quality factor of 100 uses a divisor of one, a factor of 1 uses a divisor of 50. This sets a lot of the coeffient values to zero.
  5. The quantizied coefficients are Huffman compressed and the compressed data are written to the file.


JPEG Specification: "Digital Compression and Coding of Continuous-tone Still Images, Part 1, Requirements and Guidelines," ISO/IEC DIS 10918-1. Available from ANSI sales (212) 642-4900. JFIF file specification: Literature Dept., C-Cube Microsystems, 399A W Trimble Rd, San Jose CA 95131, (408) 944-6300.

The Independent JPEG Group archive on also contains an on-line copy of the JFIF specification and additional JPEG information as: and .

See also: Application Note on JPEG compression

PCX was developed by Zsoft for the PC Paintbrush program. Due to its inefficient compression scheme this format has lost its former popularity.

Specification: Zsoft (no current address or phone number, Zsoft was absorbed by Word Star, which was absorbed by SoftKey). Many graphics textbooks contain a description of the PCX format.

PNG stands for Portable Network Graphics. This format is intended to be a patent-free replacement for LZW-compressed GIF. It uses the LZ77 compression rather than LZW. Supported pixel depth is 1- to 48-bit.

Specification: PNG FTP archive site --

  • tgainfo
  • loadtga
  • loadtgawithalpha
  • savetga

TGA is the format developed by Truevision for the Targa video framegrabber. It supports 8-, 16-, 24-, and 32-bit pixel depth, compressed or uncompressed.

Specification: Truevision

TIFF stands for Tagged Image File Format. TIFF was developed jointly by Aldus (now Adobe), Hewlett-Packard, and Microsoft as a flexible method to store scanned images of virtually any pixel type. The most common are 1-, 8-, and 24-bit pixel depth. A TIFF file that contains more than one image is called "multipage" TIFF.

For document storage the TIFF format includes the efficient Group 3 and very efficient Group 4 compression schemes. For photographic images PackBits (a run length encoding), JPEG, and LZW compressions are optional. TIFF is probably the most widely used of all file formats.

Specification: Adobe Developers (206) 628-6593

BIF is 8-bit pixel brightness data stored as one byte per pixel. Brightness values are assumed to range from 0 (black) to 255 (white). The image data is stored in rows where the first row of data corresponds to the top of the image. There is no header information or palette data stored in the file.

To successfully use a raw data file you must know the image width and allocate an image of that width to hold the data.

The common file type used for raw data is BIF, which stands for binary image file.

Specification: Catenary Systems


Other file formats

The Victor Library reads all popular raster image file formats. New file formats are added as their general use expands. That is, when images in a new format become generally available Victor will support that format.

Vector image formats are not supported.


Victor Image Processing Library

Victor Image Processing Library homepage

Copyright © 1995 - 2016 Catenary Systems Inc. All rights reserved. Victor Image Processing Library is a trademark of Catenary Systems.