JPS62225073A - Method and apparatus for compressing/restoring print picture data - Google Patents

Abstract

PURPOSE:To reduce the data quantity of a restored picture with high quality by converting an RGB data into an HSV data, quantizing and compressing it, converting it into a variable length code and sending so as to restore the send data in the reverse process to that at the compression. CONSTITUTION:The RGB data having a density at each color in the unit of picture elements extracted from a picture to be printed is converted into the HSV data having the content of hue,saturation, lightness by an HSV converter l. The HSV data obtained by the HSV converter l has a visual redudancy for each axis, then a quantizer 2 assigns a bit number corresponding thereto to apply HSV quantization by using a quantized representative number. The quantized data is converted into a variable length code data by an RICE converter 3 according to a prescribed conversion table and the result is sent. In restoring the data, the RICE inverter 5 inverts the variable length code and the inverse HSV converter 6 obtain data R',G',B' and a color inverse device 7 obtains a CMY data.

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial Field of Application) The present invention relates to a printing image data compression/restoration method and an S!
Regarding the location.

(Prior art) Conventionally, when transmitting image data for printing, CMYK
Each color (cyan, magenta, yellow, ink) is recorded in 8-bit format.

On the other hand, image data is widely used in the field of television and facsimile for purposes such as band compression, and two methods are used: complete restoration encoding and non-recovery encoding.

The former uses Huffman codes, linear codes, and arithmetic, and completely restores the information from the information source. The latter includes transform coding, vector quantization, predictive coding, etc., which reduce information from an information source to some extent and perform coding.

(Problems to be Solved by the Invention) As mentioned above, since the printing image data is not compressed, the data m is huge, for example 100 MB for A4 size.
data handling is complicated.

Furthermore, although image compression in television and facsimile machines can achieve island ratio compression, the image quality is poor and is not suitable for printing.

An object of the present invention is to provide a method for compressing/restoring image data that can ensure image quality suitable for printing while reducing data conversion.
3 and to provide equipment.

(Means for Solving the Problem) In order to achieve the above object, the present invention transforms RGB image data into H (hue), S (saturation), and ■ (lightness) data in line with human visual sense. The converted data is quantized and compressed to such an extent that no visual deterioration is felt, and then variable length encoding is performed to obtain compressed data. This data is later encoded into H3
It is converted into V data, roughly converted into RGB data, and a blackout version is generated based on the V data therein.

(Example) FIG. 1 is a block diagram showing the configuration of an apparatus according to the present invention, and the overall configuration is RGB data C'M'Y'
This shows a system for converting to K' data, and is configured to perform data compression and restoration midway through the system.

The RGB data to be given to this system is, for example, a signal obtained by scanning a photographic film and color-separating it with a filter. Furthermore, the data extracted from this system may be based on RGB instead of CMY. From input to output, (1) HSV conversion (2) quantization (3) RICE conversion (4) RICE inverse conversion (5) -18V conversion is performed, and color inversion is added as necessary. do.

Next, each of these conversion operations (1) to (5) will be explained in sequence: J2.

(1) HSV conversion Refers to the operation of converting RGB data into H (hue), S (saturation), and ■ (lightness) data, and refers to l-I S in Figure 1.
This is done by V converter 1. This l-ISV conversion is an algorithm devised to facilitate the handling of color information in the field of computer graphics, and it is simple and simple in many aspects, although it is not strictly discussed in terms of color. Its effectiveness is beginning to be recognized. The details of the algorithm are as follows.

When 0≦RXG, B≦1, MAX=maximum value of R, G, B MIN=minimum value of R, G, B, deviation range RNG is RNG=MAX−M I N Also, brightness ■ is V= It is determined as M A X . Then, when MAX=O, it is set as S-O, and H=undefined. On the other hand, when MAX≠O, saturation S
is 5=RNG/MAX, and the following operators r, g, and b are found.

That is, r= (MAX-R)/RNG Q= (MAX-G)/RNG b-(MAX-B)/RNG When R=MAX from this operator r, 9, b, h=b-g G= When MAX, h=r-b+2 When B=MAX, h=g-r+4 are obtained, and the hue H is: When h≦0, H=60 x h + 360h>
When O, it is obtained as H=60Xh.

Figure 2 shows an example of data analysis that was the subject of conversion from RGB data to l-ISV data. It is a film, generally R, G.

B Each data can be said to be a concentration distribution centered on the intermediate value.

Figure 3 shows the results of HSV conversion of the RGB data in Figure 2, where the hue H is most frequently red (R).
Others are generally average. Furthermore, the saturation SG is concentrated at about t0.03. The brightness V had substantially the same characteristics as the density distribution of RGB data.

(2) Dumpling This is the operation performed by the d-dumpling device 2 in Figure 1, and is the same as (1) above.
Since the data obtained by HSV conversion has visual redundancy in each axis, the first stage of compression can be achieved by allocating the corresponding number of bits and requantizing it. The number of m child bits is set to 5.2 for each of H, S, and V as a value that saturates the S/N ratio and does not cause visual deterioration on printed matter.
It is better to set it to 7 bits. This is determined so that the sum of squared errors is minimized by quantization of MAX.

When the original pixel value of the data obtained by the above 1-ISV conversion is replaced with its dumpling representative number, the result is as shown in FIG. 4. The quantization representative number is H (hue)
If you look at it, it is any one of 32 (5 bits) colors, and if you look at S (color 0), it is 4 (2 bits).
This indicates which of the 128 (7 bits) levels it is in for (brightness). In the example of Figure 4, for example, 1'' is R and Y.
The fact that there are many hues closer to R between , and the fact that S has low saturation indicates that V also has low brightness, and the target image part is that J: Una H , S, and V. As can be seen from this example, pixels in the vicinity have the same or similar numbers and form a group. Therefore, by utilizing this property, more efficient compression can be performed using methods such as run length. Here, in order to perform compression with emphasis on image quality, complete restoration encoding is used to return to this stage.

(3) RICE'a conversion (variable length encoding) In this embodiment, this is performed by the RICE converter 3 in FIG.
Variable length encoding is performed using the RICE method. RICE
The encoding is complete restoration encoding for compressing one-dimensional data (scan data, etc.). In this encoding, the minimum unit of encoding is called a block, and it is desirable that the data length J can be shortened. For this purpose, the characteristics of the original data are checked and encoding is performed accordingly, but here the data is set to J=21.

FIG. 5 shows an example of this characteristic, and shows that the closer the difference between adjacent pixels is to O, the higher the tAi degree is.

From this, the difference between pixels is determined, and the smaller the absolute value of each difference, the shorter the code is assigned. This is shown in the table below.

Table 1, Figure 6 shows 414 examples of the resulting data, and this format includes ID, RF, variable code,
The data consists of four parts: 10
is the code number of this block data, RF is the differential reference data, and E is the block terminator.

FIG. 7 shows an example of a device configuration for creating such variable length data. This device includes a difference calculator 31, F
It consists of an S generator 32, a code selector 33, and an 8-word data generator 34.

The difference calculator 31 is constituted by a subtraction unit, and calculates the difference between adjacent pixels for each pixel block made up of, for example, 21 pixels.

This calculation result is given to the FS generator 32.

The FS generator 32 is realized by one vinyl counter and a comparator, and assigns a variable length code as shown in Table 1 above according to the difference value from the differential compensator 31. This variable length code, that is, the code is given to the code selector 33.

The code selector 33 has 3J stages (3×21=6 in this example).
The FS code from the FS generator 32 is encoded from the 3-stage shift register. This encoding is performed by selecting one of three modes depending on the bit length F of the FS code. This is the second below
It is as shown in the table.

This code is then sent to the 8-word coder 34 or output as is.

Table 2 8 The word coder 34 consists of a ROM having an appropriate table, and converts the bit string code from the code selector 33 into 3 bits.
Convert into code in 1-pitch increments. If the bit string length is not divisible by 3, dummy bits are added. Table 3 below shows this conversion operation.

Table 3 What has been described so far is the process of data compression. The compressed data is transmitted, eg, stored, communicated, etc. by the transmission means 4 shown in FIG. 1, and then restored. Next, the restoration operation will be explained.

(4) RICE inverse transformation The decoding procedure using RICE inverse transformation is as follows. Basically, the procedure of RICE transformation is followed in reverse, and 1
) mode selection, 2) 3-bit data reconstruction, 3) differential data reproduction, and 4) reverse differential reproduction.

1) Mode Selection Corresponding to the mode selection according to Table 2 above, it is determined whether processing by the 8-word decoder is required or not.

2) 3-bit data reproduction This is carried out when the mode of bit data reproduction by the 8-word decoder is selected as a result of the above 1) mode selection.

3) Partial data reproduction The differential data is reproduced using a table having the opposite relationship to the above-mentioned Table 1.

4) Add each difference value to the reverse difference reference data as the value of the first pixel in the block to reproduce the H, S, and ■ values of each pixel.

In this way, the 5V of the reconstructed data is the maximum value of RGB. Also, if you look at CMY,
is the inversion of RGB (187 subtracted from 255), so it can be said that it is the minimum of CMY (111).

Since the smear data in printing is basically a non-linear function of the minimum value of CMY, it is possible to obtain the smear data '' by passing the V data through an appropriate look-up table (LUT) 8 as shown in FIG. Sumidata has the desired gradation such as skeleton black and full plaque.

(5) Inverse) ISV conversion This is performed by the inverse HSV converter 6 shown in FIG. 1, and is performed by the following method. In other words, 1-1 (hue) is 36
Those having an extent of 0°, S (saturation) J3 and ■ (brightness) are assumed to be positive numbers less than or equal to 1, and are calculated as follows.

0≦H≦360.0≦S≦1.0≦■≦1 However, all real numbers h' = H/60 h'' = integer part of h' = h' - h'' (decimal part of h') p = (
1-8) Let xV q = (1-3xf)xV t = (1-(Sx(1-f))xV. Next, when h″-Q, (R, G, B) = (V, t , (1>=
1 = (Q, V, p)-2
= (p, V, t) = 3 = (p, q, V) =
4 = (t, l), V
)=5=(V, p,
q) The RGB data obtained in this way is, for example, the first
The data may be converted into CMY data by the color inverter 7 shown in the figure, or it may be subjected to normal processing.

Figure 8 shows the overall system from capturing RGB data from the target image, compressing and restoring it, performing color correction, and final output. This is the device shown in Figure 1.

In this system, the image data importing device 1R10 extracts RGB data, sends it to the compression system 20, compresses the data, gives it as HSV data to the transmission system 3o, and sends the data from this transmission system 30 to the restoration system 40, which then sends the CMYK data. is obtained, and the color is corrected by the color corrector 50. Further gradation correction may be performed as necessary.

〔Effect of the invention〕

As described above, the present invention converts RGB data to 1-1 (hue).
, S (saturation), and V (lightness), each of which is converted into an octad and compressed, and then converted into a variable length code and transmitted. By restoring, the restored image can be of high quality and the data speed can be kept to a minimum.

I can do that. Further, the blackout data can be formed using the ■ data in the restored 1-ISV data.

[Brief explanation of drawings]

Fig. 1 is a block diagram showing the configuration of the \Ft position according to the present invention, Fig. 2 is an example of an image of an indoor dance picture, RGB Hist 1 - Gram, and Fig. 3 is H3V Hist 1 - of the same image. Figure 4 is an explanatory diagram of the t-+ S V distribution situation for a part of an image, Figure 5 is a pixel difference histogram of J3 to the H8v distribution in Figure 4, and Figure 6 is an illustration of the pixel difference histogram of the H8v distribution in Figure 1. 1 on the device
FIG. 7 is a block diagram showing the configuration of the RICE converter shown in FIG. 1. FIG. 8 is a printing image data creation device including the device shown in FIG. FIG. Applicant's agent Sato -Yu No. 1 煕 Keyaki Matters Meng Shu - Higashiji○kosu rJ1i117'))i+l'H1'J'l;117-
Sun Moon layer 1 nf / # I 3rd Yoshishi IL Shi J - Sugi degree) Figure procedure ne 111 Positive basket F (Method) June 1, 1985

Claims (1)

[Claims] 1. Convert RGB data, which has the density of each color in pixel units extracted from the image to be printed, into HSV data containing hue, saturation, and brightness, and convert this HSV data into quantized representative data. A method for compressing image data for printing, which performs HSV quantization using value numbers, converts the quantized data into variable-length code data according to a predetermined conversion table, and transmits the data. 2. Convert RGB data, which has the density of each color in pixel units extracted from the image to be printed, to HSV data containing hue, saturation, and brightness, and convert this HSV data into HSV quantum data using the quantization representative value number. This quantized data is converted into variable length code data according to a predetermined conversion table, and this variable length code data is converted to HSV using an inverse conversion table.
A method for compressing/restoring image data for printing, comprising: inversely converting the HSV quantized data into quantized data; converting the HSV quantized data into RGB data; and forming corner data using the V data in the HSV quantized data. 3. An HSV converter that converts the RGB data extracted from the image to be printed and having the density of each color in pixel units into HSV data containing hue, saturation, and brightness, and HSV from this HSV converter. a quantizer that converts data into HSV quantized data using quantization representative value numbers determined for each of hue, saturation, and brightness; and a quantizer that converts the HSV quantized data into variable length code data and transmits the data. A printing image data compression device equipped with a transmission system. 4. An HSV converter that converts RGB data extracted from an image to be printed and having density of each color in pixel units into HSV data containing hue, saturation, and brightness, and HSV from this HSV converter. a quantizer that converts data into HSV quantized data using quantization representative value numbers determined for each of hue, saturation, and brightness; and a quantizer that converts the HSV quantized data into variable length code data and transmits the data. a transmission system; a code inverse converter that converts variable length code data from this transmission system into difference data and further converts it into HSV quantized data; and a code inverse converter that converts the HSV quantized data from this code inverse converter into RGB data. an inverse HSV converter; a color inverter for creating CMY data according to the output of the inverse HSV converter; and means for creating CMY data in accordance with the V data in the HSV quantized data from the code inverse converter. A compression/decompression device for printing image data.