A comparison of one- and two-dimensional vector quantization of images

The authors discuss results for compressing images using both 2-D and 1-D vector quantization (VQ). Typically, an image is scanned in a line-by-line basis. The 2-D technique is based on a window that spans several adjacent lines. The 1-D technique is based on a window that spans a portion of a single line. The advantages of the 1-D scheme are considerable reduction in implementation complexity, reduced storage requirement, and a faster encoder response. Problems encountered in the 1-D technique are discussed and compared to the problems encountered in the 2-D technique. Methods to improve the perceptual quality for 1-D coding are given