Wavelet compression is a technique used to compress video and audio signals. It works best for compressing high-contrast images and short-duration audio. As video and audio files are often very large, compression is important because it allows information to be shared over a network or sent through email more quickly than the uncompressed file could be. Wavelet compression uses a process called wavelet transforms to gather the necessary information about the file. The resulting file can be either lossless, meaning it is as good as the original, or lossy, meaning some information has been lost.
In order to compress, for example, a photograph using wavelet compression, the entire image is treated as a series of wavelets. Wavelets are changes from pixel to pixel as measured by the deviation of an individual pixel from zero. The distance of the deviation is recorded as a coefficient, a whole number that measures some property or characteristic. In this case, the coefficient measures the color of the pixel. The process of measuring and recording the coefficients of the pixels is called a wavelet transform.
At this stage in the wavelet compression, no compression has yet taken place. The image has just been translated into a form that the computer can work with. Instead of a color image, the computer now has a series of numbers. The next step in wavelet compression is a process called decomposition.
During decomposition, the coefficients of adjacent pixels are averaged. Averaging produces a simplified version of the wave, making the description of the image smaller. This process is repeated until the image is fully compressed. The more times the process is repeated, the smaller the data file can be and the quicker the file will be to transmit.
The final size of the compressed file depends not just on the size of the original file, but also on the desired quality of the final product. Sometimes as a file is decompressed the information that the computer deems unnecessary is thrown away to allow the file to get smaller. This is known as lossy compression, and it results in a file that is not quite as good as the original.
Wavelet compression can also use a lossless method, in which no information is thrown away. This results in a compressed file that, when decompressed, is exactly the same quality as the original file. The trade-off is that a file compressed with a lossless method will be larger than one compressed with a lossy method.