JPH04188952A - Color picture communication system - Google Patents

Abstract

PURPOSE:To improve a communicating efficiency, and to operate a faithful color reproduction by operating a communication by an N-valued data corresponding to an equal color space between communicating devices, and converting and processing the data into the specific color space of its own by the device at a reception side. CONSTITUTION:RGB color multilevel coded data read by a color scanner 1 are converted into the standard equal color space by a color converting part 2. Then, the color data L, (a), and (b) are N-valued so that the density can be held in a certain constant area by a density holding type N-valuing part 3, and turned into data L2, (a)2, and (b)2. Then, the data are compression-encoded by a compressing part 4, data L3, (a)3, and (b)3 are transmitted to a communication controlling part 5, and then the transmission is operated. The received standard color space data are converted into the specific color space of its own by the device at the reception side in order to reproduce a picture. Thus, the communicating efficiency can be improved, and the faithful color reproduction can be operated.

Description

[Detailed description of the invention] [Industrial application field]

The present invention relates to a color image communication system, for example, a color image communication system for performing color image communication between different types of devices such as color scanners, color printers, and computers with color displays.

[Conventional technology]

Color image communication between different devices refers to the communication of color images between a color scanner and a color printer, a color scanner and a computer with a color display, or a computer and a color printer, and also includes transmission communication using PC communication and facsimile. It can be done. The data read by a color scanner that reads color images is
For example, there are many multivalued data such as 8 bits for each color of RGB. On the other hand, some color printers are capable of printing only binary data, such as 1 bit each of CMY, as seen in color printers of the thermal melt transfer method and the inkjet method. Conventionally, color image communication between different models has been performed as follows. Color image communication between a color scanner that reads multivalued data and a color printer that outputs binary data is processed as shown in FIG. Multivalued color data read by the color scanner 101 is binarized by a binarization unit 102. Then, in order to improve communication efficiency, the data is compressed by the compression unit 103 and transmitted. The receiving side receives the compressed data,
By decompressing it in the decompressing unit 106, it becomes binary color data,
The configuration or procedure is to output it to a binary color printer 107. In addition to the above-mentioned color printers, there are also color printers that can print out multi-valued data, such as thermal dye sublimation color printers and electrophotographic color printers. Even when outputting to these color printers, processing is performed as shown in FIG. 6 in order to improve communication efficiency. That is, the data is processed in the order of the binarization section 102, the compression section 103, the decompression section 1.06, and the multi-value conversion section 108. The multivalue converting unit 108 converts the resulting binary color into multivalue color data, and outputs it to the multivalue color printer 107.

[Problem to be solved by the invention]

However, in the conventional example described above, the color density is not binarized in a certain fixed area, but is binarized using a simple method, and furthermore, color scanners and color printers each have their own color reproduction range. Since color space conversion is not taken into consideration, there are the following drawbacks. (1) Color density is not preserved in a certain area,
This results in an image with poor color balance. (2) It is not possible to faithfully reproduce the colors of the original. The present invention has been made in view of the prior art, and it is an object of the present invention to provide a color image communication system that has high communication efficiency and can reproduce colors with high fidelity.

[Means to solve the problem]

In order to solve this problem, the color image communication system of the present invention has the configuration shown below. That is, in a color image communication system for communicating color images between different types of devices, the sending device includes a first color conversion means that converts the color image data of the sending side cargo into data corresponding to a uniform color space. and N-value converting means for converting the converted data into N-values so as to preserve density within a predetermined area, and the receiving device converts the received data into multi-value color data corresponding to the uniform color space. and a second color conversion means that converts the multi-value color data obtained by the multi-value color data into data corresponding to a color space specific to the receiving device.

[Effect]

In such a configuration of the present invention, between the transmitting and receiving devices,
Exchange data that corresponds to the standard color space. In particular, the receiving device converts the received standard color space data into its own unique color space to reproduce an image.

【Example】

Embodiments of the present invention will be described in detail below with reference to the accompanying drawings. FIG. 1 shows a block configuration diagram relating to color image communication between different models according to an embodiment. The color multi-value data of 8 bits for each RGB color read by the color scanner 1 is converted into L''a*b* space, which is a standard uniform color space, by the color conversion unit 2. Each 8-bit color Data L, a, and b are stored in the concentration-preserving N-value conversion unit 3.
, N such that the concentration is conserved in a certain region
Value conversion is performed, and each n-bit data Lx, at, b
Becomes @. Then, the data Lm is compressed and encoded in the compression unit 4 to improve communication efficiency, and sent to the communication control unit 5.
am, t)m is sent and transmitted. On the receiving side, the communication control unit 6 receives data from the communication control unit 5 on the transmitting side. Received data Ls, &s+bn
Since this is data encoded for communication, the decompression unit 7 restores it to Lx, at, and bt. Here, since compression and expansion are reversible, Lx, which is the output of the expansion section 7,
, ax. b2 is the output L2+a of the concentration-preserving N-value converter 3 on the transmitting side.
It is obtained as exactly the same data as 2+bz. This N
In order to perform color conversion on the value color data L21 & 21b2, the multivalue data L',
are converted into a' and b'. Here, this L', a
', b' are the standard uniform color space data, but they are the output of the color conversion unit 2, and a and b do not strictly match in the minute area, but the N-value processing Since it is a concentration conservation type, the concentration is conserved within a certain amount range, and L', a', and b' are the same, and a and b are the same. These color data L', a', b' are converted into color reproduction range color data L'', a'', b'' specific to the color printer 11 in the color changing section 9.
In order to be able to output the value to the color printer 11, the density-preserving N-value conversion unit 10 converts it into L”2.”z,b”2.
The data is converted into a value and sent to the color printer 11, which outputs an N-value color image. The overall flow of color image data has been explained above with reference to FIG. 1, but individual processing will be explained below. The color converter 2 converts R, G, and B data read by the color scanner 1.
The part that converts data into L 11 all b@data is performed in the following process. In advance, the color scanner 1 reads pixels for which colorimetric data of L@, am, and b@ has been obtained using a measuring device.
Its outputs are R, G, B and L", a". A 3×3 matrix representing the relationship with bo is derived. By increasing the number of sample points, the matrix becomes more accurate (. The matrix obtained in this way is , and converts R, G, and B data into a and b data.In addition, a so-called look-up table that simply inputs the read data as an address and outputs the corresponding data is used in the ROM. It goes without saying that the 5 matrix is not limited to 3 x 3. In the concentration-preserving N-ary conversion unit 3 and IO, binarization processing is performed using, for example, an error diffusion method.Error diffusion The method is a type of pseudo-gradation display method. Binarization is performed using a fixed amount of value, and the error that occurs at that time is weighted as shown in Figure 2 and diffused to surrounding unbinarized pixel positions to increase the density. By applying this error diffusion method to each data of a and b, Lx+aa,b, in which the density is preserved, is derived in a 2×2 pixel area. The method of N-value processing will be explained in detail using ternarization as an example using FIG. D. Addition with a o + b, 3 is performed, and L H, a
N, b 0 that becomes N, where L o r a o
The initial value of +b o is "0". Then, the selector 72 determines how many of the 8 bits each of LN, aN, and bs has. As shown in FIG. 8, the level from 0 to 255 is divided into three stages, and when it is in the range 0 to 63, it is converted to "00" by the N value converter 73, and 64
“01” when at level ~191, 192~25
When it is at level 5, it is converted into ternarized data L Rr a t + b t such as "lO". On the other hand, in the representative value setting section 74, L*-a*+
b *When each value is “00”, set it to “0”;
When it is "l", it is "128", when it is "lO", it is "255".
”, set the representative value L mr a l + b s, and subtract it from the original data or a and b using the subtractor 75,
Calculate each error L t+ a t, b y. The error diffusion unit 76 applies an error L to surrounding non-ternarized pixel positions by weighting as shown in FIG.
Diffuse o. Incidentally, the diffusion matrix is of course not limited to the 2×2 matrix shown in FIG. Moreover, it goes without saying that the representative value and the threshold value are not taken from this example. Further, the information is not limited to ternary conversion, but may be binary conversion, quaternary conversion, or N-value conversion. In the compression section 4 and expansion section 7, N-value image data is compressed, expanded, encoded, and decoded, and the procedure is based on the MH method. The MH method is one in which data of all scanning lines is one-dimensionally encoded. One-dimensional encoding is a method of encoding the continuous length of equivalent pixels, 1 pixel and 0 pixel, that alternately appear on one line. This will be explained in detail using FIG. Here, the discussion will proceed assuming that the "1" pixel is a white pixel and the "0" pixel is a black pixel. As shown in Fig. 3(b), a table is prepared in advance in which the continuous length and the code are in one-to-one correspondence, and the pixel and continuous length are determined according to the input data, and the code is determined. You can get a percentage. When image data as shown in FIG. 3(a) is input, first, pixels and continuous length are determined. Since there are 5 consecutive white pixels, the table shown in FIG. 2(b) is searched and the code data "1100" is obtained. It is encoded using this procedure. On the other hand, in decoding,
In contrast to this, the data is decoded using the opposite method. Therefore, the data is not degraded during this encoding and decoding. 2
+b2 is converted into multivalued (8-bit) data L', a', b'. Specifically, for example, processing is performed to set the average value in a 3×3 matrix as the value of the pixel of interest position (center pixel), but at this time, the desired weighting is done by a coefficient (the weighting of the pixel of interest position is ), the value of the pixel of interest position may be calculated. The color conversion unit 9 converts the gold space data L'', a', b' into color reproduction range data L'', a specific to the color printer 11.
", b"'. FIG. 4 shows a hardware embodiment for realizing this color space compression. Here, γ represents the distance from the L° axis, and can be calculated as γ=Ia''+b''. Further, θ is an angle from the a° axis, and is determined by b# or θ==5in−'□I a"+b" weight, etc. The output from the γ calculation section 41 is multiplied by the output d from the product coefficient table in a multiplier 44, and the data after color space compression is given to the synthesis section 45. Therefore, the multiplier 44 multiplies the compression ratio d and the golden space of 8 bits (0 to 255), and outputs data that sets the color reproduction range of the color printer 11 to 8 bits (0 to 255). The compression ratio d at this time is read from the product coefficient table. Then, from θ and γ3, a” and b” are found in the synthesis section 45,
The color calculation unit 46 calculates L'', a'', and b''. <Other Examples> In the example, a standard uniform color space is used, but L*
The full v11 space may be used, or a more uniform color space may be used. Density-preserving binarization methods are not limited to the error diffusion method, but also the average density-preserving method and the method that best preserves the density. You may use a binarization method such as compression,
Regarding decompression, it is not limited to the MH method;
The R method, the MMR method, or even a binarization method that is more efficient and suitable for this embodiment may be used. In addition, although the color conversion unit 9 uses a method of uniformly compressing the golden color space, there is no reason to limit it to this method. For example, there is a method of converting to a part of the golden space, a method of compressing a part of the golden space and leaving the central part as is, etc.
The conversion method is not limited. As explained above, according to this embodiment, deterioration in image quality due to N-value conversion is prevented by performing N-value conversion in which color density is preserved in a certain area during color image communication, and color A well-balanced image can be obtained, and by having a means for color conversion to a color reproduction range specific to a device such as a color printer, it is possible to faithfully reproduce the colors of the original. Furthermore, the uniform color space in this example is L* a
In addition to *b*, L@u11va or any other color space may be used as long as the spatial distance corresponds equally to the color difference to the human eye. [Effects of the Invention] As explained above, according to the present invention, communication is performed between communication devices using N-value data corresponding to a uniform color space, and the receiving device converts the data into its own unique color space. Since the processing is performed using the same method, communication efficiency is high and color reproduction can be performed with high fidelity. Furthermore, when the data between devices is encoded data, communication efficiency is further improved.

[Brief explanation of drawings]

Fig. 1 is a diagram showing the relationship between the communication side device and the receiving side device and their configuration in this embodiment, Fig. 2 is a diagram for explaining the concentration preserving binarization method, Fig. 3 (a) , (b) is a diagram for explaining the concept of compression and expansion, Figure 4 is a concrete block configuration diagram that performs color conversion, and Figures 5 and 6 are block configurations of devices that perform conventional image communication. FIG. 7 is a block diagram of the N-ary conversion section of the embodiment, and FIG. 8 is a diagram for explaining the principle of ternarization. In the figure, l...color scanner, 2...color conversion unit, 3
．． 10... Density-preserving binarization section, 4... Compression section, 5
．． 6... Communication control unit, 7... Decompression unit, 8... Multivalue conversion unit, 9... Color conversion unit, 11... Color printer. Patent applicant: Canon Co., Ltd. Sheep, convulsions 2 in 1 '+ 9
1111 Figure 1 Figure 2 Figure 1 7°'71100 10 1111 1
+ Figure 3 Jiang I 1ull
ξHi IB 1
Figure 5 ff1('m(i'1 Figure 6)

Claims (2)

[Claims] (1) In a color image communication system for communicating color images between different models, the sending device includes a first color conversion means for converting color image data unique to the sending side into data corresponding to a uniform color space. and N-ary conversion means for converting the converted data into N-values so as to preserve density within a predetermined area, and the receiving device converts the received data into multi-value color data corresponding to the uniform color space. a second unit that converts the multivalued color data obtained by the multivalued color data into data corresponding to a color space specific to the receiving device;
A color image communication system comprising: color conversion means. (2) The transmitting device further includes means for compressing and encoding the N-valued data by the N-ary encoding means, and the receiving device further comprises compressing and encoding the data encoded by the N-ary encoding means. 2. The color image communication system according to claim 1, further comprising means for decompressing and decoding the data.