GB2073987A

GB2073987A - Memory table

Info

Publication number: GB2073987A
Application number: GB8039576A
Authority: GB
Original assignee: Dainippon Screen Manufacturing Co Ltd
Current assignee: Dainippon Screen Manufacturing Co Ltd
Priority date: 1979-12-21
Filing date: 1980-12-10
Publication date: 1981-10-21
Also published as: DE3047695A1; FR2473236B1; FR2473236A1; JPH0151829B2; GB2073987B; DE3047695C2; JPS5688138A

Abstract

A method and device for preparing a memory table for color control conditions for use in a picture reproducing machine having a table index memory are disclosed. Groups of turning point data of straight line segment approximation functions of basic characteristics curves are stored in a memory, and then one of the groups of the turning point data are read out of the memory by a data selector. Then, data between the two adjacent turning points are processed by adding or subtracting a correction factor (Figures 4, 5 not shown) along straight line segments, each connecting the two adjacent turning points, in a function generator which outputs the group of the turning point data and the processed data in order of their address numbers into the table index memory, thereby obtaining a memory table having a desired characteristics curve. <IMAGE>

Description

SPECIFICATION Memory Table This invention relates to a method and device for preparing a memory table for color control conditions such as a density range correction, a color correction, a gradation correction, and the like, for use in a picture reproducing machine such as a color scanner and a color-facsimile.

In a picture reproducing machine such as a color scanner and a color facsimile, which performs the process in a digital manner, a memory table is used instead of an analog circuit of a conventional analog machine, which performs a non-linear processing, in order to carry out it readily in a real-time processing.

However, conversion charactersitics of the memory table are varied depending on a color tone of an original picture, a gradation curve, and the like, and thus are rewritten to the desired characteristics per each original picture.

In a conventional method, a plurality of tables prepared in advance are stored in a magnetic memory together with their corresponding indexes, and then, as occasion demands, the desired table is read out of the magnetic memory to a memory of a scanning circuit of the color scanner.

However, one memory table is required to each color channel separated and further one memory table is also required to each of a density range correction, a color correction, a gradation correction, and so forth.

Such a lot of memory tables desired requires a memory having a large capacity and accordingly, in practice, it is difficult to perform this method. Further, even if it is possible to prepare the memory tables in advance, it takes rather long time to select the desired tables by increasing the access time and thus to set up the color control conditions of the color scanner.

It is an object of the present invention to provide a method for preparing a memory table for color control conditions for use in a picture reproducing machine such as a color scanner and a color facsimile, free from the above described defects, which is quick, stable and economical, and which reduces a lot of memory capacity.

It is also an object of the present invention to provide a machine for preparing a memory table for color control conditions for use in a picture reproducing machine such as a color scanner and a color facsimile, free from the above described defects, which is quick, stable and economical, and which reduces a lot of memory capacity.

According to the present invention there is provided a method for preparing a memory table for color control conditions for use in a picture reproducing machine, comprising the staps of (a) storing groups of turning point data of straight line segment approximation functions of basic characteristics curves into a first memory, (b) reading one of the groups of the turning point data of the first memory, (c) processing data between the two adjacent turning points along straight line segments, each connecting the two adjacent turning points, and (d) storing the group of the turning point data and the processed data between the two adjacent turning points in order to their address numbers into a second memory, thereby obtaining a memory table having a desired characteristic curve.

According to the present invention there is also provided a machine for preparing a memory table for color control conditions for use in a picture reproducing machine having a table index memory, comprising (a) a memory for storing groups of turning point data of straight line segment approximation functions of basic characterstics curves, (b) a data selector which selects one of the groups of the turning point data read out of the memory, and (c) a function generator which processes data between two adjacent turning points from the turning point data read out of the memory, along straight line segments, each connecting the two adjacent turning points, and which outputs the group of the turning point data and the processed data between the two adjacent turning points in order of their address numbers to the table index memory of the picture reproducing machine, in which a memory table having a desired characteristics curve is prepared.

In order that the present invention may be better understood, preferred embodiments thereof will be described with reference to the accompanying drawings, in which: Figure 1 is a block diagram of a means for performing a method for preparing a memory table according to the present invention, together with a table index memory of a table indexer of a picture reproducing machine; Figure 2 shows a straight line segment graph for explaining a preparation of a table according to an approximation method:: Figure 3 shows a relation between a characteristics curve of a basic table and a characteristics curve of a desired table; Figure 4 shows a correction value curve stored in a correction table; Figure 5shows three kinds of correction value curves stored in correction tables; Figure 6 shows examples of a characteristics curve of a basic table used in the means shown in Figure 1; Figure 7 is a block diagram of another means for performing another method for preparing a memory table according to the present invention, together with a table index memory of a table indexer of a picture reproducing machine; and Figure 8 shows characteristics curves of basic tables used in the means shown in Figure 7.

Referring now to the drawing there is shown in Figure 1 one embodiment of a table preparing means 8 for color control conditions such as a density range correction, a color correction, a graduation correction, and so forth, according to the present invention, together with a table index memory 1 of a table indexer of a color scanner.

The address line 2 of the table index memory 1 is connected to an input line 4 of an input picture signal Dx via a line switching means 3 such as a three state buffer circuit. The output data line 5 of the table index memory 1 is connected to an output line 6 of an output picture signal Dywhich is processed by a table, via a line switching means 7 which is turned in synchronization with the line switching means 3 so that the address line 2 may be coupled to the input line 4 when the output data line 5 is connected to the output line 6.

When the table is rewritten with the desired data, the address line 2 of the table index memory 1 is connected to an address signal line 9 of a table preparing means 8, and the table index memory 1 is enabled in the writing mode.

A proper number of basic tables having basic conversion characteristics curves which are selected from all the conversion characteristics curves required to the memory tables, are selected. Then, the data along basic conversion characteristics curve of each basic table are compressed or skipped to obtain a group of turning point data by a conventional method according to a straight line segment approximation function whose segments are connected at the turning points. Thus the obtained groups of the turning point data are stored in a turning point data memory 10 of the table preparing means 8 prior to the operation.

The turning point data memory 8 includes a plurality of areas 11, each storing each group of the turning point data of each basic table. One group of the turning point data are read out of the corresponding area 11 of the memory 10 and are sent to a function generator 12 which outputs straight line segment approximation function data.

As shown in Figure 2, the function generator 12 receives the group of the turning point data (x0,y0), (xl,yl), ~~~~~~, and (xn,yn), and the data between the two adjacent turning points are obtained by determining a primary function f1, f2 , i.e. a straight line segment, passing through the two adjacent turning points, that is, the data along the straight line segments are operated.

For example, when the address number A is varied from 0 to 256 in the direction of the X axis, an output code D between the two adjacent turning points in the direction of the Y axis is calculated according to the following formula.

When x0 < A < x1, D = f1 = a1A + y0, and therefore a = Y2-Yo x,~xO When x1 < A < x2, D = f2 = a2A + yr, and thus a2 = y2-y1 x2-x1 When xn-1 < A < xn, D = fn = anA + yn-1, and therefore an = Yn-Yn-1 Xn-Xn-1 Each address number A and the corresponding output code D which is obtained by the above calculation, may be sent to the table index memory 1 as the data of the coordinates point.

The address number A output from the function generator 12 is sent to the address signal line 9, and addresses the table index memory 1 and a correction memory 13. The code D output from the function generator 12 is sent to an input terminal of an add-subtractor 14. A correction code d output from the correction memory 13 is fed to another input terminal of the add-subtractor 14. Both the codes D and d are added or subtracted each other in the add-subtractor 14 to obtain a corrected code which is sent to the table index memory 1.

Consequently, when the address numbers A are addressed from 0 to 256, the function generator 12 outputs the codes D corresponding to the address numbers A, and the corrected codes obtained by adding or subtracting the correction codes d to or from the codes D, are sent to the table index memory 1 and are stored therein, thereby obtaining a memory table having a desired characteristic curve.

In Figure 3 there are shown a basic characteristics curve 15 output from the function generator 12, shown by a solid line, and a corrected characteristics curve 16 output from the add-subtractor 14, shown by a broken line. The difference between the two curves 15 and 16 corresponds to the correction code d. The correction curve of the correction codes d is stored in a correction table, as shown in Figure 4.

The correction memory 13 is divided into a plurality of blocks 17 for storing the correction tables of different kinds of correction characteristics curves, as shown in Figure 5, and each block 17 is selected by a table selector 18 connected to the correction memory 13 depending on the desired correction characteristics.

As shown in Figure 5, the correction characteristics curves have different maximum extreme values in their centers (b), left hand side (a), and right hand side (c).

The correction characteristics curves shown in Figures 5a and 5c are symmetrical, and hence, if the addressing operation is carried out in the reverse direction, i.e. from 256 to 0, the correction characteristics shown in Figure Scare obtained from those shown in Figure 5a. Therefore, it is enough to prepare the correction characteristics curves shown in Figures 5a and 5b.

Further, in each correction table, the maximum extreme value of the correction characteristics curves can be expressed by means of four bits, and the correction characteristics curves are fixed. Hence, the correction memory 13 may be a type of readout only memory (ROM).

On the other hand, for example, regarding the gradation correction ten basic gradation correction curves are prepared, as shown in Figure 6, and they are stored in the areas 11 of the turning point data memory 10 prior to the operation.

Further, in Figures 5a, 5b and 5c are shown four correction characteristics curves, and, if the addition (D+d) and the subtraction (D-d) are carried out from the code D and the correction code d per one correction characteristics curve, altogether 24 correction tables are substantially to be prepared.

Accordingly, 240 memory tables corrected are prepared from 10 basic tables and 24 correction tables, and thus there will be altogether 250 memory tables of 240 memory tables corrected and 10 basic tables.

Further, if each basic table includes approximately 16 turning points and thus a group of 16 turning point data, in practice, the conversion characteristics curve can be reproduced sufficiently, and accordingly one area 11 for storing the group of 16 turning point data can be at least 16 bytes of rather small capacity.

In this way, the capacity of the turning point data memory 10 may be 16 bytes x 10 = 160 bytes, and that of the correction memory 13 may be 4 bits x 256 addresses x 8 tables of Figures 5a and 5b = 1024 x 8 bits = 1024 bytes. Hence, the table preparing means 8 requires altogether 1184 bytes. That is, the table preparing means 8 having 1184 bytes is substantially capable of storing the table index memory 1 having 256 bytes multiplied by 250 memory tables, i.e. 64000 bytes of 64K bytes. In other words, the table preparing means 8 substantially corresponds to a device having 64K bytes of capacity.

From the above description of its readily understood that the turning point data memory 10 can be a very small capacity, and may readily be composed of small number of integrated circuit elements such as a read only memory (ROM), a read-write non-volatile memory, and the like.

The selection of the area 11 of the turning point data memory 10 is performed by an area selector 19 which is provided with marks or numbers corresponding to the basic characteristics curves shown in Figure 6.

One of the 250 memory tables obtained as described above is indexed by a combination of the selection numbers of the table selector 18 and the area selector 19 and the data indicating the addition or the subtraction of the correction code d to or from the code D.

In Figure 7 there is shown another embodiment ofthetable preparing means 8 for the color control conditions according to the present invention, together with the table index memory 1. The table preparing means 8 shown in Figure 7 has almost the same construction as the one shown in Figure 1, except that a subtractor 26, a weighting circuit 27 and a weighting factor sellter 28 instead of the correction memory 13 and the table selector 18 which are shown in Figure 1.

In this embodiment, for example, four basic characteristics curves 21 - 24 of four basic tables shown in Figure 8 are selected from the desired conversion characteristics curves. Each characteristics curves 21,22, 23 or 24 is deviated to the maximum extent with respect to a straight line 20 of a formula y=x. The groups of the turning point data of the four basic tables are stored in the areas 11 of the turning point data memory 10 in advance, as abovementioned.

A characteristics curve which is deviated less than the basic curves 21 - 24 can be prepared from the basic curves. For example, a characteristics curve 25 shown by a broken line in Figure 8 is prepared by using the turning point data of the curve 21 as follows.

The function generator 12 outputs the output codes D of the points along the curve 21, when the address numbers A are addressed, as described above. On this occasion, the output code D and the address number A are fed to the subtractor 26 wherein (D-A) is calculated to obtain a difference code d1 which is to be sent to the weighting circuit 27. In the weighting circuit 27 the difference code d1 or (D-A) is divided by a weighting factor m which is generated by the weighting factor settler 28, to obtain a coded2. The code d2 output from the weighting circuit 27 and the output code D from the function generator 12 are sent to the add-subtractor 14 and the subtraction (D-d2) is performed therein to obtain a corrected code d3 to be sent to the table index memory 1, thereby obtaining a memory table corresponding to the curve 25 in the table index memory 1.

The address number A equal the output code D of the line 20 represented by the formula y=x, and therefore the difference code d, output from the subtractor 26 equals a deviation AD between the curve 21 and the line 20.

The weighting factors m settled in the weighting factor settler 28, for example, may be $, 81,82 and #, and by using these factors eight different characteristics curves including the curve 21 are obtained.

Now, each basic table for the basic curve 21, 22, 23 or 24 requires the memory capacity of 16 bytes, as described above, and, even if 20 basic tables are prepared, the turning point data memory 10 requires only the capacity of 320 bytes. In this embodiment, therefore, the table preparing means 8 including the turning point data memory having 320 bytes of capacity if capable of preparing 160 memory tables.

Then, when the weighting circuit 27 of Figure 7 is arranged after the correction memory 13 of Figure 1, the capacity for obtaining the correction characteristics curves shown in Figure 5 are largely reduced. That is, in this embodiment, only the two correction characteristics curves, each having the maximum value, of the correction tables shown in Figures 5a and 5b are prepared in advance, and require the capacity of 256 bytes, while the turning point data memory 10 of Figure 1 requires the capacity of 160 bytes. Thus, altogether 416 bytes are required.

In this case, if the weighting factors m are determined to -, 4, 4 and this embodiment can prepare 250 memory tables, which is the same number as the embodiment shown in Figure 1. It is apparent that the capacity required is largely reduced from 1184 bytes to 416 bytes.

Further, according to the present invention, by a combination of the embodiments shown in Figures 1 and 7, a complicated characteristics curve may be prepared and a large number of memory tables may be prepared by using a memory having a small capacity.

Although the present invention has been described in terms of preferred embodiments thereof illustrated in the accompanying drawings, however, various changes and modifications can be made by those skilled in the art without departing from the scope of the present invention.

Claims

1. A method for preparing a memory table for color control conditions for use in a picture reproducing machine, comprising the steps of: (a) storing groups of turning point data of straight line segment approximation functions of basic characteristics curves into a first memory; (b) reading one of the groups of the turning point data out of the first memory; (c) processing data between the two adjacent turning points along straight line segments, each connecting the two adjacent turning points; and (d) storing the group of the turning point data and the processed data between the two adjacent turning points in order of their address numbers into a second memory, thereby obtaining a memory table having a desired characteristics curve.

2. A method as defined in claim 1, wherein the group of the turning point data and the processed data between the two adjacent turning points are corrected by adding or subtracting a correction factor in order of the address numbers to obtain a group of corrected data which are stored into the second memory, thereby obtaining a memory table having a desired characteristics curve.

3. A device for preparing a memory table for color control conditions for use in a picture reproducing machine having a table index memory, comprising :.

(a) a memory for storing groups of turning point data of straight line segment approximation functions of basic characteristics curves; (b) a data selector which selects one of the groups of the turning point data read out of the memory; and (c) a function generator which processes data between two adjacent turning points from the turning point data read out of the memory, along straight line segments, each connecting the two adjacent turning points, and which outputs the group of the turning point data and the processed data between the two adjacent turning points in order of their address numbers to the table index memory of the picture reproducing machine, in which a memory table having a desired characteristics curve is prepared.

4. A device as defined in claim 3, further comprising a data correction means which corrects the group of the turning point data and the processed data between the two adjacent turning points by adding or subtracting a correction factor in order of their address numbers to obtain a group of corrected data which are sent to the table index memory of the picture reproducing machine.

5. A device as defined in claim 4, wherein the data correction means comprises a correction memory storing correction factors; a factor selector which selects the correction factor read out of the correction memory; and an add-subtractor which performs an addition or a subtraction of the data sent from the function generator and the correction factor to obtain the group of the corrected data.

6. A device as defined in claim 4, wherein the data correction means comprises a subtractor which subtracts the address number from the data sent from the function generator to output a difference data; a weighting circuit which divides the difference data sent from the subtractor by a weighting factor predetermined; and an add-subtractor which performs an addition or a subtraction of the data sent from the function generator and the weighted data sent from the weighting circuit to obtain the group of the corrected data.

7. A method of preparing a memory table substantially as hereinbefore described with reference to and as illustrated in Figures 1 to 6 or 7 and 8 of the accompanying drawings.

8. A device for preparing a memory table substantially as hereinbefore described with reference to and as illustrated in Figures 1 to 6 or 7 and 8 of the accompanying drawings.