Method for Compressing Character Image Data

The present invention relates to a font system utilizing a font, for example, a numeral, character or symbol which is displayed on a screen of a ?. There is a printer, for example, a laser printer for printing one page as a unit, particularly the invention is directed to a font data compressing method which stores by compressing each byte of a font into a specific compression code.

Generally, a font system utilizing a font, as shown in FIG. 1, which includes a processor 1 for controlling the whole operation of the system, a font memory 2 for storing a font for example with respect to each numeral, character and symbol or the like, a frame buffer 3 for storing a font of amount of one page by the control of said processor 1, and a video signal producing means 4 for outputting the font stored in said frame buffer 3 as a video signal of serial form with synchronizing to a video clock signal by the control of said processor 1.

The font memory 2 may, for example a ROM, be stored semi-permanently with the font, or it may down loaded with the font from an auxiliary device such as a diskette such as a RAM.

According to the font system constructed as above, a processor 1 seeks the font to be output from the font memory 2 and stores it in a corresponding location of the frame buffer 3. In the font memory 2 there is stored with a same font with the font of last output outputting by storing in the frame buffer 3. That is to say, in a font memory 2 there is stored with a bit map font.

Thus, when a bit map font of the quantity of one page is stored in the frame buffer 3, a video signal producing means 4 outputs the bit map font stored in the frame buffer 3 as a serial video signal by the control of the processor 1 by synchronizing to a video signal clock.

A font system described is simple in processing and becomes relatively speedy in processing speed because of storing the bit map font of lastly outputting form as it is. There is a problem, that in selection of the font because many memories are required for storing one font in case where the number of dots which is the bit map size of a font is large since each separate font should be stored by each dimension. This is a very serious problem in case of processing a Korean character "Hangeul" and a Chinese character "Kanji" in a system requiring a high resolution such as a laser printer.

An object of the present invention is to provide a compressing method of a font capable of decreasing a utilizing capacity of a memory by storing each bytes of said font data by compressing without storing font data as they are.

An object of the present invention is achieved by adopting a previous byte among the font data as a reference byte and a present byte as a comparative byte, and comparing those two bytes, and according to said compared result, storing into a memory by assigning a specific compression code.

FIG. 1 shows a block diagram illustrating a conventional font system.

FIG. 2 is a block diagram showing a compressing process of a font data according to a compressing method of the present invention.

FIG. 3 is a schematic diagram showing the compressing sequence of a font data according to the compressing method of the present invention.

FIG. 4 is a table of a compression code assigning in response to the compared result according to the compressing method of the present invention.

FIGS. 5A, 5B, 5C, 5D, 5E, 5F, 5G, 5H, 5I, 5J, 5K and 5L are schematic diagrams showing the comparative examples according to the compressing method of the present invention.

FIG. 6 is a flow chart illustrating the compressing method of the present invention.

FIG. 7 is an examplary diagram of the font for applying the compressing method of the present invention.

FIG. 2 is a block diagram showing a process for compressing the bit map font data according to the compressing method of the present invention. As illustrated in the drawing, a bit map font is stored in a bit map font file 12 and is compared according to a compression program 11 by separating into a reference byte and a comparative byte, and a specific compression code is assigned according to the compared result and stored in a compression code file 13. FIG. 3 is an explanatory diagram showing a sequence for compressing a font data, in which a compression of a font is carried out by dividing it into a column unit, and scanning from the top toward the bottom in a same column. A present byte of a font to be compressed is compared by adopting the byte immediately before as a reference byte. A first byte of each column is compared with 00H (in here, H represents a hexadecimal digit) as a reference byte because the first byte has no byte immediately before to be a reference. A specific compression code is assigned to each byte according to said compared result. When the compression of a font is carried out, it is stored respectively by dividing into a compressed data and a non-compressed data.

FIG. 4 shows a table giving a specific compression code in response to the compared result of 12 kinds compared a reference byte with a comparative byte in accordance with the compressing method of the present invention, whose compression code is, after seeking an accumulative degree with respect to the 12 kinds by placing a Ming-style printing type font of Hangeul 40.times.40 dots as one set, given in optimum by a Hoffmann method from said statistical data.

This compression code assigning method will be explained in detail with reference to the comparative exemplary diagrams of FIGS. 5A to 5L.

At first, as shown in FIG. 5A, when a reference byte and a comparative byte are same, as in case where a reference byte is 60H and a comparative byte which is a present byte to be compressed is 60H, a compression code "0" is assigned. As shown in FIG. 5B, in case where a comparative byte is not same with a reference byte and said byte is 00H (WHITE), a compression code "1100" is assigned. As shown in FIG. 5C, in case where a comparative byte is not same with a reference byte and said byte is FFH (BLACK), a compression code "11010" is assigned. As shown in FIG. 5D, when it becomes same with a comparative byte in case that a reference byte is shifted by one bit to left (SL) as in case where a reference byte is 01H and comparative byte is 02H, a compression code "111111" is assigned. As shown in FIG. 5E, when the reference byte is the same with a comparative byte in case that a reference byte is shifted by one bit to right (SR) as in case where a reference byte is 80H and a comparative byte is 40H, a compression code "11110" is assigned. As shown in FIG. 5F, when the reference byte becomes the same with a comparative byte in case that a reference byte is shifted by one bit to left thereafter it is ANDed with a reference byte (SL-AND) as in case where a reference byte is 03H and a comparative byte is 02H, a compression code "11011" is assigned. As shown in FIG. 5F, when it becomes same with a comparative byte in case that a reference byte is shifted by one bit to left thereafter it is ORed with a reference byte (SL-OR) as in case where a reference byte is 03H and a comparative byte is 07H, a compression code "10110" is assigned. As shown in FIG. 5H, when it becomes same with a comparative byte in case that a reference byte is shifted by one bit to right thereafter it is ANDed again with a reference byte (SR-AND) in case where a reference byte is COH and a comparative byte is 40H, a compression code "111110" is assigned. As shown in FIG. 5I, when it becomes same with a comparative byte in case that a reference byte is shifted by one bit to right thereafter it is ORed again with a reference byte (SR-OR) as in case where a reference byte is 80H and a comparative byte is COH, a compression code "10111" is assigned. As shown in FIG. 5J, when it becomes same with a comparative byte by reversing a reference byte after i-th bit (REV 1), that is to say, by reversing a reference byte after third bit as in case where a reference byte is 00H and comparative byte is 1FH, a compression code "1010" and "011" which designates third bit being l-th bit are assigned. As shown in FIG. 5K, when it becomes same with a comparative byte in case that a reference byte after i-th bit is reversed thereafter a byte after j-th bit is reversed again, that is to say, a reference byte after third byte is reversed thereafter a byte after 6-th bit is reversed again (REV 2) in case where a reference byte is 00H and a comparative byte is 1CH, a compression code "100", and "011" and "110" designating "6" which is i-th bit and "3" which is j-th bit are assigned. As shown in FIG. 5L, when any one among above eleven kinds does not correspond to therebetween as in case a reference byte is 10H and a comparative byte is 78H, a compression bit "1110" and its comparative byte 78H are assigned.

FIG. 6 is a signal flow chart showing a method for compressing a font data according to a compressing method of the present invention, in which firstly 00H as a basic byte is established, and the first byte of a column is read out and it is established as a comparative byte thereafter they are compared. At this moment, a compression code "0" is stored to a buffer 1 when a comparative byte is same with a reference byte, a compression code "1100" is stored to a buffer 1 when a comparative byte is 00H, that is to say, a comparative byte is all white, a compression code "11010" is stored to a buffer 1 when a comparative byte is FFH, that is all black, a compression code "111111" is stored to a buffer 1 when it is same with a comparative byte in case that a reference byte is shifted by one bit to left (SL), a compression code "11110" is stored to a buffer 1 when it is same with a comparative byte in case that a reference byte is shifted by one bit to right (SR), a compression code "11011" is stored to a buffer 1 when it is same with a comparative byte in case that a reference byte is shifted by one bit to left thereafter it is ANDed again with a reference byte (SL-AND), a compression code "10110" is stored to a buffer 1 when it is same with a comparative byte in case that a reference byte is shifted by one bit to left thereafter it is ORed again with a reference byte (SL-OR), a compression code "111110" is stored to a buffer 1 when it is same with a comparative byte in case that a reference byte is shifted by one bit to right thereafter it is ANDed again with a reference byte (SR-AND), a compression code "10111" is stored to a buffer 1 when it is same with a comparative byte in case that a reference byte is shifted by one bit to right thereafter it is ORed again with a reference byte (SL-OR); when it is same with a comparative byte in case that a reference byte after i-th bit is reversed (REV 1), then a compression code "1010" is stored to a buffer 1 at the same time said "1" is stored with three bits; when it is same with a comparative byte in case that a reference byte after i-th byte is reversed thereafter a byte after j-th byte is reversed again (REV 2), then a compression code "100" is stored to a buffer 1 at the same time said "l" and "j" are sequentially stored with three bits respectively; when it is the case that is does not correspond to the case of aforementioned eleven kinds (NEW), then a compression code "1110" is stored to a buffer 1 at the same time its comparative byte is stored to a buffer 2.

Thereafter, said comparative byte is established as a reference byte, and the next byte is read out and it is established as a comparative byte thereafter aforementioned comparing operation is carried out repeatedly, and then a compression code is stored to a buffer 1 at the same time a comparative byte which is not compressed is stored to a buffer 2.

An applied example of a compressing method of the present invention as described above will be explained with reference to a case of FIG. 7 hereinafter.

FIG. 7 is an exemplary diagram of a font with respect to a Hangeul character " ", in which when a font with regard to a column 1 is compressed by said compressing method of the present invention, it is shown in table 1 as below.

That is to say, it is assigned with each compression code "0" because a reference byte and a comparative byte are the same from first byte to 7th byte of column 1, when 8th byte reverses a byte after 3rd bit of 7th byte which is a reference byte and reverses again a byte after 6th bit, it becomes the same with 8th byte which is its comparative byte, therefore a compression code "100" is assigned at the same time non-compression codes "011", "101" which represent their 3rd bit and 6th bit are assigned; and when 9th byte shifts the 8th byte which is a reference byte by one bit to 10 right, it becomes same with the 9th byte which is its comparative byte, therefore a compression code "11110" is assigned. When all bytes of column 1 are compressed by these method, the compression codes and the non-compression codes are assigned as shown in table 1, and then these compression codes are stored in a buffer 1 and the non-compression codes are stored in a buffer 2.

Wherein, the compression codes become to 61 bits and the non-compression codes become to 9 bits thereby there are 70 bits in total. Therefore, they become to take merely 27.3% relative to 256 bits (=32.times.8 bits) which are the total bit number of column 1 that is not compressed.

As described above in detail, according to the present invention each of the bytes of a font data are compressed and then stored, the capacity for storing them in the memory becomes decreased and accordingly a large quantity of font data can be stored therein, and there is also an effect that a capacity of the memory can be saved up to about 75% in the case of Hangeul Ming-style printing type of 40.times.40 dots.