TABLE OF CONTENTS Most modern image compression methods are either progressive or optionally so. Progressive compression is an attractive choice when compressed images are transmitted over a communications line and are decompressed and viewed in real time. When such an image is received and is decompressed, the decoder can very quickly display the entire image in a low-quality format, and improve the display quality as more and more of the image is being received and decompressed. A user watching the image developing on the screen can normally recognize most of the image features after only 5 10% of it has been decompressed.
This should be compared to raster-scan image compression. When an image is raster scanned and compressed, a user normally cannot tell much about the image when only 5 10% of it has been decompressed and displayed. Images are supposed to be viewed by humans, which is why progressive compression makes sense even in cases where it is slower or less efficient than nonprogressive.
Perhaps a good way to think of progressive image compression is to imagine that the encoder compresses the most important image information first, then compresses less important information and appends it to the compressed stream, and so on. This explains why all progressive image compression methods have a natural lossy option; simply stop compressing at a certain point. The user can control the amount of loss by means of a parameter that tells the encoder how soon to stop the progressive encoding process. The sooner encoding...
