Browse Prior Art Database

Byte or Multi-Byte Coded Form for Representation of Non-Coded Information

IP.com Disclosure Number: IPCOM000043631D
Original Publication Date: 1984-Sep-01
Included in the Prior Art Database: 2005-Feb-05
Document File: 4 page(s) / 48K

Publishing Venue

IBM

Related People

Fisher, RC: AUTHOR

Abstract

While encoding systems for the representation of alphanumeric data are commonplace, graphic images have not been amenable to comparative methods of encoding. This article describes a method to allow compressed-image data to exhibit the same byte-oriented data structure as coded alphanumeric data. EBCDIC and ASCII are well-known code schemes for the representation of alphanumeric character data. Non-coded information (NCI), however, being of an unstructured form, has until now not lent itself to a coded form which can assimilate the characteristics of ASCII and EBCDIC -- viz., coexist within the same data structures as do ASCII and EBCDIC.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 39% of the total text.

Page 1 of 4

Byte or Multi-Byte Coded Form for Representation of Non-Coded Information

While encoding systems for the representation of alphanumeric data are commonplace, graphic images have not been amenable to comparative methods of encoding. This article describes a method to allow compressed-image data to exhibit the same byte-oriented data structure as coded alphanumeric data. EBCDIC and ASCII are well-known code schemes for the representation of alphanumeric character data. Non-coded information (NCI), however, being of an unstructured form, has until now not lent itself to a coded form which can assimilate the characteristics of ASCII and EBCDIC -- viz., coexist within the same data structures as do ASCII and EBCDIC. In the past, attempts at coding NCI have concentrated on versions of run-length encoding, both in single- and two-dimensional form, largely to minimize the amount of data to be stored and transmitted. A quite common form, the Huffman run-length encoding system, yields a variable bit length to represent various runs of binary 11 or 0/ information, and while this results in a compaction of the data, it has a form which does not readily lend itself to existence in a data structure based on a byte or multi-byte form of expression. In the preceding article, a single-dimension compression algorithm for compaction of NCI data is described. It yields a one or two-byte representation of binary data. The concepts presented therein have formed the basis for the invention presented in this description, i.e., a coded form of representation of image data which easily exists in standard information-processing data structures. Because of its byte or multi-byte nature, it readily lends itself to a form of data entry which is hexadecimal, yet describes binary 1 and 0/ data of run lengths varying over a wide range. Let us consider an Image Catalog, such as that shown in Fig. l, as consisting of a 32K-byte memory organized as 16K x 16 bits. The first 128 words are pointers containing the starting addresses of up to 128 separate images. The pointers themselves are composed of 16 bits, sufficient to address up to 64K bytes of memory. Memory address position 0/0/0/0/ in hex always contains the next available starting address, so that, initially, addresses 1 to 127 each contain hex 0000, and address hex 0000 contains address location 128. If an image is written into the catalog containing 100 words, then address location hex 0001 would contain the binary number 128, and address location hex 0000 would contain the number 228 as the next available location. Addresses between 128 and 227 would comprise in hex notation a description of the first image. Fig. 2 shows a representation of the first image to be described in hexadecimal form and placed in the image catalog. (It should be noted that images can be of any size up to a single one occupying the entire page.) The image in Fig. 2 comprises the area required by 8 equivalent text characters with a p...