A high resolution laser scanner or recorder, used in the printing industry, will address a large number of pixels on a page, typically 108 - 101° pixels. The data rate will generally be about 10 Mbit/s. Processing, transmitting and storing such quantities of data gives emphasis on data compression techniques. Further, in many instances the compacted data can be more conveniently processed than the original pixel or run length format data. An example of this is the scaling of outline coded fonts and logotypes, and gray level coded photographs. We discuss in this paper the requirements imposed on codes useful in compacting laser scanned images. Primary features of interest are: compression ratio and its dependence on the type of image, processing capacity and speed required in coding and decoding, possibility of processing the coded data, and effects on the quality of the image. Some further points concern noise reduction, unfaithful coding techniques, origin of the coded images, and merging of data of different code structure. The problems created by the lack of standardization of graphics industry image coding are commented on and a comparison is made to telefacsimile codes. The coding problem is also addressed from the point of view of the extensive research carried out in the fields of image processing and image analysis. The research in high resolution image coding is seen to be at an early stage compared to the level of research in the fields of coding of television pictures, photograph images or low resolution facsimile.

Raimo Nurmi

Papers

Coding Of Data For Laser Recorders