Tile encoding in image printing

Abstract

A method for digitally printing a digital image by reference to a succession of code words representing blocks of pixels of the original image. The image is originally subdivided into uniform size blocks of pixels. Each of the blocks of pixels is given a unique identifier unless it is identical to any of the previously scanned blocks. Identical groups are given the same identifier. In printing the digital image, the identifiers are scanned in sequence and the corresponding pixel blocks are successively stored in buffer memories for modulating the printer.

Description

The present invention relates to block encoding and, in particular, to the application of block encoding in the printing of digital images.

In laser printing or scanning characters onto a CRT display, it is well known to modulate the scanner in response to a code word representing each character. Typically, a seven bit code word or other non-binary code word represents a character to be printed and a decoder responds to the code word to address a bit map in suitable memory to modulate the printer or scanner to provide a succession or zeros and ones to print the character image.

It is also known to store information necessary for displaying characters in various sizes and fonts, although the size of the storage often becomes excessive. To reduce the amount of data required for a character output, a technique known as "run length coding" was developed to compress the data by storing the length data of the black and white areas of the characters as shown in U.S. Pat. No. 3,582,905. These systems, however, only apply to predetermined characters such as the scanning of alphanumeric characters. It would be desirable, therefore, to extend the concept of character recognition and bit mapping to the printing of stored digital images.

It is also known in the prior art to be able to reproduce and transmit high detail pictoral data such as in facsimile systems. Such systems generall require either a very wide bandwidth or a relatively long time to transmit or recreate the image. It is also known, as taught in U.S. Pat. No. 3,215,773, that there is a high degree of redundancy present in most images and that savings can be achieved by taking advantage of this redundancy. In particular, U.S. Pat. No. 3,215,773 teaches the scanning of an image in contiguous small areas and representing the signal derived from each area by a code signal having a complexity determined by the detail present in that area.

It is also known as described in U.S. Pat. No. 4,013,828 to scan and process an image in picture element groups of predetermined size. As each picture element group is scanned, the pattern represented by the corresponding dithered image bits is compared to a dictonary of patterns stored in a first memory. If the pattern is not among those in the memory, it is assigned an associated code word and entered into the memory. In addition, both the pattern and its code word are transmitted to the remote location where they are stored in a second memory, and individual cells of a cell group corresonding to the scan picture elements are energized in accordance with the pattern. If, on the other hand, a scan pattern is the same as one already stored in the first memory, only the code word associated with that pattern is transmitted to the remote location. The code word is used at the remote location to retrieve the associated pattern from the second memory when it is displayed at the corresponding cell group of the display panel.

A difficulty with these prior art systems, however, is that they are not adaptable to a variety of digital images having various degrees of redundancy and capable of being represented by variable length code words. It is, therefore, an object of the present invention to extend the block encoding concept of a one byte or less code word to a variable number of bits to represent by identifiers or code words various portions of a digital image. It is another object of the present invention to be able to print an image in the same manner as printing a character without any major changes in the printing logic. A still further object of the present invention is to be able to compress data on a storage element such as a disk in order to run a display at a higher rate. Further advantages of the present invention will become apparent as the following description proceeds, and the features characterizing the invention will be pointed out with particulrity in the claims annexed to and forming a part of this specification.

Briefly, the present invention is a method for digitally printing a digital image by reference to a succession of code words representing blocks of pixels of the original image. The image is originally subdivided into uniform size blocks of pixels. Each of the blocks of pixels is given a unique identifier unless it is identical to any of the previously scanned blocks. Identical groups are given the same identifier. In printing the digital image, the identifiers are scanned in sequence and the corresponding pixel blocks are successively stored in buffer memories for modulating the printer.

For a better understanding of the present invention, reference may be had to the accompanying drawings wherein the same reference numerals have been applied to like parts and wherein:

FIG. 1 is a general block diagram of the system incorporating the present invention;

FIG. 2 is an illustrative digital image broken down into uniform sub-blocks;

FIG. 3 is an illustration of unique identifiers or character numbers and associated bit patterns or tile sets; and

FIG. 4 is an illustration of a tile stream or scan line record in accordance with the present invention for displaying a digital image.

With reference to FIG. 1, there is generally shown a general block diagram of a system incorporating the present invention. In particular, there is shown a memory generally shown at 10, a disc 12 storing a digital image to be displayed, a processor 14 communicating with the memory 10 and the disc 12, and a display 16. The memory 10 is further broken down into various locations including the operating system 18, the tile encoding procedure 20, a tile set 22, a tile stream 24 and output buffers 26 and 28 communicating with the display 16.

FIG. 2 illustrates a typical digital image stored on disc 12. In accordance with the present invention, the digital image is segregated into tiles or blocks, each block containing N by M bits or pixels. For purposes of illustration, it will be assumed that there are 64 blocks and that each of the tiles or blocks 1 through 64 shown in FIG. 2 comprises an 8.times.8 matrix of bits or pixels. It should be noted, however, that any arbitrary matrix size N by M is suitable, for example 16.times.16 bit blocks, 8.times.16, 6.times.8, or 13.times.9 bit blocks. Assuming there is a digital image stored on disc 12 and it is desired to segregate the image into 8.times.8 tiles or blocks of pixels, the first step is to compare each 8.times.8 bit block with all previous blocks or tiles to establish unique characters or identifiers related to the blocks.

With reference to FIG. 2, the first 8.times.8 bit block, number 1, is automatically assigned an identifier 0, since there are no 8.times.8 bit blocks preceding block number 1. Thus, the unique identifier 0 is identified with bit block number 1. It should be noted, however, that the identifier zero is often reserved for a bit block of all zeroes. Next, bit block 2 is compared to bit block 1 in the processor 14. Assuming that bit block 2 is identical to bit block 1, that is the bit pattern of block 1 is identical to the pit pattern of block 2, no unique identifier is associated with bit block 2. Bit block 2 is therefore represented by the identified 0. Next, bit block 3 is compared to bit block 1 and bit block 2. In this case, assuming that bit block 3 is different from bit blocks 1 and 2, a unique identifier number 1 is associated with bit block 3.

In this manner, a scratch pad memory of identifiers and related first occurrences of a particular bit pattern or tile set is established. With reference to FIG. 3, there is shown a table of sequential identifiers and also a table or tile set of related 8.times.8 blocks of pixels. The identifier 0 corresponds to tile set 1 and the idetifier number 1 corresponds with tile set 3 since tile set 2 is identical to tile set 1.

With continued reference to FIG. 3, the next unique identifier number 2 is associated with tile set or bit block 8. In other words, in the hypothetical example, tile set 8 is different from any of the previous 7 tile sets. This also implies that the comparison of tile sets 4, 5, 6 and 7 has determined that tile sets 4, 5, 6 and 7 are identical to either tile set 1 or tile set 3. In a similar manner, each of the 64 tile sets of FIG. 2 is successively compared with all previous tile sets. FIG. 3 illustrates the comparison of each bit pattern with all previous bit patterns. That is, tile set 13 is different from any of the previous 12 tile sets and is given the unique identifier 3, tile set 15 is different from the previous 14 tile sets and is given the identifier 4, down to tile set 52 which is different from all previous 51 tile sets and is given the unique identifier 18.

No further unique identifiers being shown after identified 18 means that tile sets 53 through 64 are identical to one of the tile sets 1 through 52. Each of the tile sets corresponding to a unique identifier is then stored in a separate memory location, tile set memory 22. It should be noted that in the example given, there are 18 unique identifiers. In a typical digital image, however, the number of unique identifiers is often found to be greater than 256, thus necessitating an identifier code greater than 8 bits or 1 byte.

One further memory table is established, the tile stream 24 shown in FIG. 1. The tile stream 24 is a record of the identifier corresponding to each of the 8.times.8 bit blocks of the image in FIG. 2. Thus, the unique identifier for block 1 is 0 as illustrated in FIG. 3. In the tile stream, a zero (0) will be inserted in the location corresponding to block 2 to signify that the tile set corresponding to block 2 is in fact the same tile set as for block 1. That is, the tile stream is a stream of identifiers to identify the tile set or bit pattern relating to each of the 64 blocks.

Thus, the first two identifiers of the tile stream will be zeros. Since the block number 3 is different from blocks 1 and 2, and therefore has a unique identifier 1, a one (1) will be inserted in sequence following the two zeros in the tile stream. Assuming that block 4 is identical to block 3, block 4 will receive the same identifier as block 3, i.e. identifier number 1. In a similar manner, assuming that block 5 is identical to block 1 and blocks 6 and 7 are identical to block 3, a first scan line of the tile stream will be (001100112) as illustrated in FIG. 4.

Block 8, as illustrated in FIG. 3, has the unique identifier 2 and thus the 2 is inserted in sequence in the tile stream to correspond to the last 8.times.8 bit block or block 8 in the first line of the image of FIG. 2. Therefore, for each of the 64 blocks of FIG. 2, a corresponding identifier is provided in the tile stream. The identifer in the sequence of tile streams is either unique representing that the block is not identical to any of the other previous blocks or is a previous identifier representing that the particular block is identical to another previous.

It should be noted that for each block of FIG. 2 there is a corresponding identifier in FIG. 4. The first identifier in the second scan line is a 0 indicating that block 9 is identical to block 1. In a similar manner, block 10 is identical to block 3, and blocks 11 and 12 are identical to block 8. Block 13 is different from any previous block and therefore receives the unique identifier 3. Block 14 is identical to block 13 and therefore also receives identifier 3 and blocks 15 and 16 are different from all previous blocks and identical, and therefore receive the identifier 4.

There are 18 different tile sets. The remaining tile sets are identical to at least one of the other tile sets. These 18 tile sets are stored in the tile set memory location 22. The tile stream is stored in the tile stream location 24. It should be noted that in all digital images there are generally found to be redundnacies. By taking advantage of these redundancies, it is possible to by only storing a tile set and a tile stream to greatly decrease the amount of storage location required by relating each segregated block of the image to an identifier, In general, most images probably cannot be identifed with less than 256 identifiers but clearly a two byte identifer would be sufficient. In accordance with the present invention, by listing the number of unique identifiers required, it is possible to determine the number of bits necessary to completely relate identifiers to each bit block of the image.

With the tile sets stored in location 22 and the tile stream in location 24, the system is ready to display or print the image. Before printing, however, it is necessary to establish a buffer size to accomodate the bit patterns, usually either 8 or 16 bits. The height of the bit pattern is the number of bits in the vertical direction, in the illustrated example, 8 bits.

The first step in printing is threfore to read the first tile stream or record that has been stored on disc 12 in a scratch pad area into a temporary buffer as illustrated as temporary buffer 30. This is the first scan line of the image, that is (00110112). For each identifier in the temporary buffer 30, the next step is to load the corresponding tile set into the first buffer. Preferably, the first buffer is of sufficient size to hold the first scan line of the image. That is, the first scan line will be the tile sets corresponding to image blocks 1 through 8. Thus, in the particular example given, buffer 26 as well as buffer number 28 will be of sufficent size to accomodate eight 8.times.8 pixel blocks.

After the first record has been loaded into buffer 26, the first buffer contents modulate the printer or display to display the first scan line of the image on the display 16. While printing out the contents of the buffer 26, the second record or scan line (01223344), as illustrated in FIG. 4, is read from the disc into the temporary buffer 30. The tile sets corresponding to the second scan line are loaded into buffer 28. When buffer 26 has been read out to modulate the display, the contents of buffer 28 began modulating the display. In this manner, the entire image is displayed by sequentially displaying the tile sets corresponding to each of the scan lines illustrated in FIG. 4, until the tile sets corresponding to the eighth scan line (13, 8, 5, 2, 0, 1, 0, 0) has been displayed.

The addressing of the tile sets is straightforward. If the beginning address of block 1 of the tile set is 0000 is binary, then the beginning address of block 3 corresponding to identifier 1 is 1000 is binary or 8, since each of the tile sets is an 8.times.8 block. The next block in the tile set is block 8 corresponding to identifier 2 and therefore is located at beginning address 10000 is binary or 16. Therefore, to find the beginning address of a particular tile set it is only necessary to multiply the identifier corresponding to that tile set by 8 to determine the beginning address of the particular block. It should be noted that there are alternate means to address a particular tile set, for example, a pointer to the position in the word is kept. The Appendix is a preferred embodiment of the method of block encoding for displaying digital images. It should be noted that the results of this process is to create a compressed image on a disk and to allow the display to be operated at a higher bandwidth than the disk.

While there has been illustrated and described what is at present considered to be a preferred embodiment of the present invention, it will be appreciated that numerous changes and modifications are likely to occur to those skilled in the art, and it is intended in the appended claims to cover all those changes and modifications which fall within the true spirit and scope of the present invention. ##SPC1##

Claims

1. A method for displaying a digital image comprising the steps of

(a) segregating the digital image into a plurality of blocks of pixels,

(b) assigning a first identifier associated with the first block of pixels,

(c) comparing the second block to the first block and assigning a second identifier with the block if the blocks are different and assigning the first identifier to the second block if the blocks are identical,

(d) comparing the third block to the first and second blocks and assigning a third identifier associated with the third block if the third block is different from the first and second blocks and the first and second blocks are different and assigning the associated identifier of either the first or second block to the third block if the third block is identical to either the first or second block,

(e) repeating the sequence for said plurality of blocks,

(f) preparing a block set for each nonidentical block, correlating the block set with the associated identifier,

(g) preparing a block set stream of identifiers associated with each successive block of the image, and

(h) displaying the image by scanning the identifiers in the block set stream and loading the associated block sets into a display buffer.

2. In an image display system for displaying a digital image, said image being segregated into a plurality of bit blocks, each nonidentical bit block being associated with an identifier, the system including a first buffer, a second and a bit block stream storage, the bit block stream storage containing the identifier corresponding to each subsequent bit block of the digital image, the method of displaying the digital image comprising the steps of

scanning the bit block stream for each identifier associated with each sequential bit block of the digital image,

loading a first portion of the bit blocks corresponding to the identifiers in the first buffer,

initiating display of the bit blocks in the first buffer,

loading the bit blocks corresponding to a second portion of the bit block stream in the second buffer,

displaying the bit blocks in the second buffer after completion of the display of the bit blocks in the first buffer, and

repeating the sequence for each portion of the bit block stream whereby the entire digital image is displayed.

3. The method of claim 2 wherein each bit block is an M by N pixel matrix, M and N being powers of 2, M being the number of columns and N being the number of rows and wherein the starting address of each bit block is NX wherein X is the identifier associated with each bit block.

4. The method of claim 3 wherein each bit block is an 8.times.8 pixel matrix and each bit blocks is addressed in multiples of 8.

5. A method for displaying a digital image comprising the steps of

(a) segregating the digital image into a plurality of blocks of pixels,

(b) assigning unique identifiers associated with non-identical blocks of piexels,

(c) preparing a block set for each nonidentical block, correlating the block set with the associated identifier,

(d) preparing a block set stream of identifiers associated with each successive block of the image, and

(e) displaying the image by scanning the identifiers in the block set stream and loading the associated block sets into a display buffer.

6. In an image display system for displaying a digital image, said image being segregated into a plurlity of bit blocks, each nonidentical bit block being associated with an identifier, the system including a first buffer, a second buffer and a bit block stream storage, the bit block stream storage containing the identifier corresponding to each subsequent bit block of the digital image, the method of displaying the digital image comprising the steps of

scanning the bit block stream for each identifier associated with each sequential bit block the digital image,

loading a first portion of the bit blocks corresonding to the identifiers in a buffer,

initiating display of the bit blocks in the buffer, and

repeating the sequence for each portion of the bit block stream whereby the entire digital image is displayed.

7. In an image display system for displaying a digital image, said image being segregated into a plurality of bit blocks, each nonidientical bit block being associated with an identifier, the system including a first buffer, a second buffer and a bit block stream storage, each bit block being an M by N pixel matrix, M being the number of columns and N being the number of rows and wherein the starting address of each bit block is NX wherein X is the identifier associated with each bit block, the bit block stream storage containing the identifier corresponding to each subsequent bit block of the digital image, the method of displaying the digital image comprising the steps of

scanning the bit block stram for each identifier associated with each sequential bit block the digital image,

housing a first portion of the bit blocks corresponding to the identifiers in the first buffer,

initiating display of the bit blocks in the first buffer,

loading the bit blocks corresponding to a second portion of the bit block stream in the second buffer,

displaying the bit blocks in the second buffer after completion of the display of the bit blocks in the first buffer, and

repeating the seqence for each portion of the bit block stream whereby the entire digital image is displayed.