JP2658031B2 - Color image reading method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial application field) The present invention reads a color image using a color sensor in which pixels for reading three colors (for example, R, G, B) are sequentially arranged adjacently. About the method.

(Prior Art) There is a color sensor used in a color image input device in which pixels for reading three colors (for example, R, G, B) are sequentially arranged. For example, one-dimensional contact type CCD
In the color sensor, R, G, and B filters are sequentially deposited in one pixel, and data is basically output serially in the order of R → G → B. Then, color calculation processing for converting the set of data into three colors used for printing is performed.

(Problems to be Solved by the Invention) In this color operation processing method, the density of image reading is equal to the density of pixels of the sensor. Therefore, the reading density cannot be higher than the pixel density of the available sensor (for example, 16 lines / mm).

Conventionally, a method of electrically processing read image data has been used to increase the density of pixels.

An object of the present invention is to provide a new color image reading method for reading an image at a reading density higher than the density of an image of a sensor.

(Means for Solving the Problems) The present inventor has considered that it is possible to increase the density three times by performing this calculation even between adjacent pixels, and has arrived at the present invention.

In the color image reading method according to the present invention, a set of three types of elements for detecting three colors is arranged in a predetermined order, and a plurality of sets of elements are arranged one-dimensionally. And a color arithmetic processing unit that performs a primary conversion of a set of three-color image data detected by the one-dimensional color sensor into three-color print data of one pixel for printing. In the reading method, in a color operation processing device, a set S _i (i = 1, 3) of three color elements for detecting image data to be converted into print data of one pixel.
Multiple elements that constitute a one-dimensional color sensor as (2, ...)
A set P _j (j = 0, 0) of C _j (j = 0, 1,...) To (C _j , C _{j + 1} , C _{j + 2} )
1, ...) to calculate the print data required for printing for each of C _j , C _{j + 1} , C _{j + 2} at a speed higher than the transport speed of the input color data, and output them sequentially. Features.

(Operation) The pixel density of a color image is tripled by setting a set of three color elements of a one-dimensional color sensor so as to overlap each other.

Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings.

FIG. 2 shows a block diagram of the image reading circuit. The original is illuminated by a light source (fluorescent lamp) (not shown), and the image information of the reflected light is detected by a one-dimensional contact type color CCD sensor (1) and converted into an electric quantity. This sensor (1)
Is a CCD in which four CCD sensors are arranged in a staggered line and connected, and each sensor has R (red), G
(Green) and B (blue) filters are sequentially deposited in the scanning direction. (The shape of the pixel is not a square but a parallelogram in order to prevent the occurrence of a moiré pattern.) Therefore, each CCD sensor sequentially sends R, G, B signals to the analog multiplexer (2). . The analog multiplexer (2) selects signals output from the four sensors, converts the signals into one signal bus, and sends the signal bus to the sample and hold circuit (3). The sample and hold circuit (3)
Color separation is performed on each of the R, G, and B signals. The next signal amplifying circuit (4) performs differential amplification with respect to the black level while correcting the sensitivity of each of the signals R, G, and B. Each of the amplified signals is converted into an 8-bit digital value by the AD converter 5, then subjected to shading correction by the shading correction circuit (6), and sent to the color operation circuit (7).

The color operation circuit (7) converts each data of the read image signals R, G, and B that have been subjected to shading correction to C, which is print image data.
(Cyan), M (Magenta), and Y (Yellow).
In the color conversion processing, the following conversion is performed: print data = r · R + g · G + b · B. r, g, b are predetermined constants, and C, M, Y
Can be rewritten as follows:

However, r _1-3 , g _1-3 , and b _1-3 are constants. Therefore, in the color calculation process, after multiplying each of the constants r, g, b by the read image data R, G, B, data r · R, g · G, b · B are added.

In the color calculation processing according to the present invention, as described below,
By calculating the color image density between the adjacent pixels, 3
Up to double. As shown in FIG. 1, the i-th pixel of the sensor is arranged in the order of R _i , G _i , and B _i . Now, the 3i-th C value (C _3i ) is _calculated by the following equation using the i-th pixel data (R _i ,
G _i , B _i ).

C _3i = r ₁ · R _i + g ₁ · G _i + b ₁ · B _i ... At this time, the 3i + 1th C value (C _{3i + 1} ) is obtained by using the data of the adjacent pixel.

C _{3i + 1} = r ₁ · R _{i + 1} + g ₁ · G _i + b ₁ · B _i Further, the 3i + 2nd C value (C _{3i + 2} ) is obtained by using the data of the next pixel. Ask.

C _{3i + 2} = r ₁ · R _{i + 1} + g ₁ · G _{i + 1} + b ₁ · B _i ... Similar color calculation processing is performed for M and Y.

As described above, by performing the color calculation processing using the data of the adjacent pixels, it is possible to obtain a pixel density three times the reading density.

FIG. 3 is a block diagram of the color operation circuit (7). The multiplication block (11) is a block for performing multiplication such as rR, and its value is sequentially sent to three addition blocks (12, 13, 14) for C, M, and Y connected in parallel. , Each added value is
The data are sequentially sent to the data synthesis circuit (15). The operation of this circuit will be described later using a timing chart (FIG. 5).

FIG. 4 is a circuit diagram of a multiplication block (11) and one addition block (for example, 12). The other two addition blocks are not shown for simplicity.

As shown in the flowchart of FIG. 5, in the multiplication block (11), each data (8-bit digital data) of the input pixel data R, G, B read by the CCD sensor (1) is converted into parallel data. The data is input to the serial conversion circuit (21), and is serially transferred as data DA to the multiplier (22) in the order of R, G, and B in synchronization with the rise of the clock CK1 that is three times the dot clock. On the other hand, the selector circuit (23) selects one of the three constants r, g, and b (for example, r ₁ , g ₁ , and b _{1 in} the case of the calculation of C), and outputs the result to the multiplier (22). send. This selection is performed in synchronization with the rise of the constant select clocks CK2 and CK3 generated in the select clock generation circuit (24). Thus, r and R data, g and G data, and b and B data are input to the multiplier (22) in synchronization with each other. Therefore, the multiplier (22) sequentially performs multiplication of r · R data, g · G data, and b · B data, and multiplies the addition block by the latch circuit (2) in synchronization with the clock ▲. Send value.

In the addition block, the selector circuit (34) periodically adds the data DC added by the adder (31) to (3).
(Once each time) It clears to “φ”. Then, in order to synchronize with the output DB from the latch (25), the latch circuit (3
5), and the data DC 'is fed back to the adder (31). The adder (31) adds the input data DB from the multiplication block and the DC 'fed back, and outputs data DC through the latch circuits (32, 33).

The output DC of the latch circuit (32) is periodically cleared by the selector (34). The output DC is latched by the latch circuit (33) in synchronization with the timing of the clear, thereby realizing the operation of the expression. . That is, the addition block is r ·
While performing the addition of R + 0, r · R + g · G, r · R + g · G + b · B, only the data of r · R + g · G + b · B is output to the next processing block.

As shown in FIG. 3, the addition blocks shown on the right side of FIG. 4 are actually connected in parallel. However, the difference between the operation of each addition block is that the selection of “φ” by the selector 34 and the latch performed simultaneously therewith. Only 34 latch timings. That is, as shown in the timing chart of FIG. 5, the selector circuit 34 in the addition blocks I to III
The output data DC from 1 is periodically cleared to “φ”, but the timing of clearing each block of I to III to “φ” is different (see DC ′), and the latch circuit 33 of each block is Then, the DC is latched in synchronization with the timing of the selector 34 clearing to "φ" (see DD in FIG. 5). Therefore, if the image data r · R _n , g · G _n , b · B _n output from the multiplication block is represented by R _n ′, G _n ′, B _n ′ for simplicity, the addition block I has R _n ′ + G _n ′ + B _n ′, II
In Rn _{+ 1} '+ _Gn ' + _Bn ', in III, Rn _{+ 1} ' + _{Gn + 1} '+ B
_{This means} that _n 'is not output, and the pixel density can be tripled. Now, each block I, II, III
Are output as DE _n , DF _n , and DG _n , the output of the block is subjected to parallel-serial conversion in the order of DE _n , DF _n , and DG _n in the data synthesis circuit (15) in FIG. send. Thus, in the color operation processing circuit (7), the print data C,
When converting to M or Y, data is output three times the read pixel density.

In order to convert the print data into print data, the image input device that obtains the color information of the original repeats scanning for each of the print data C, M, and Y, and performs three scans.

Next, the output data (DH) of the color operation circuit (7) is converted to a pixel density corresponding to the print dot density on the output side, that is, the printer side, by the image density conversion circuit (8), and then sent to the printer. . FIG. 7 is a block diagram of the image density conversion circuit (8), and FIG. 8 is a timing chart of the operation. This circuit (8) is used for converting the pixel density in the main scanning, and can be dealt with by changing the scanning speed of the optical system in the sub-scanning direction (the density is improved if the scanning is slowed down). Now, assuming that the reading pixel density of the CCD (1) is 16 lines / mm, the output (DH) of the data synthesizing circuit has a density of 48 lines / mm. The latch clock in the latch circuit (42) is generated by the image density conversion clock generation circuit (41) from the dot clock CK4 synchronized with the change of DH. The image density conversion data signal is normally set according to the print dot density on the output side, and is 24 lines / mm. The image density conversion clock generation circuit (41) generates a clock (RCK) having a cycle designated by the image density conversion data signal from ▼ and sends it to the AND gate (42). The image density conversion instruction signal is a signal of negative logic, and when this signal is "L", the clock RCK output from the image density conversion clock generation circuit (41) is used as a latch clock to OR with the AND gate (42). The signal is sent to the latch circuit (44) via the gate (43). on the other hand,
When the image density conversion signal is "H", the signal is sent to the latch circuit via the AND gate (42) and the OR gate (43) using ▼ / 3 as a latch clock. Therefore, the latch circuit (44)
Outputs the pixel data of 48 lines / mm or less according to the number of print dots on the output side when the image density conversion instruction signal is “L” and 16 pixels / mm of the read pixel density when “H”. Output data.

As the color operation processing circuit, a conventionally used system shown in FIG. 9 may be used. That is, the color operation circuit (7) is composed of three multipliers (51, 52, 53) and one adder (57).

FIG. 10 shows a timing chart of this circuit. In this color operation processing circuit, a multiplier is provided for each of R, G, and B data, and the results are respectively latched (54, 55, 56)
Synchronize by. Then, 3 is added by the adder (57).
Data a, R, b, G, c and B are added in parallel, and the result of the addition is latched by a latch circuit (58) using a clock obtained by inverting the latch clock of the first latch circuit (54, 55, 56). To obtain print data C, M or Y. If the timing chart of the latch of the latch circuit (58) is thinned out, zooming (electrical reduction) becomes possible. C, M, and Y data are obtained by three scans. This circuit is more expensive than the scheme shown in FIG. 3 because it uses three rather expensive multipliers.

(Effect of the Invention) The reading density can be tripled without changing the element density of the one-dimensional color sensor.

[Brief description of the drawings]

FIG. 1 is a diagram showing the correspondence between elements of a one-dimensional color sensor and printing pixels. FIG. 2 is a block diagram of the color image reading apparatus. FIG. 3 is a block diagram of a color operation circuit. FIG. 4 is a circuit diagram of a part of the color operation circuit. FIG. 5 is a timing chart of the multiplication block. FIG. 6 is a timing chart of the addition block. FIG. 7 is a circuit diagram of the latch synthesis circuit. FIG. 8 is a timing chart of the latch synthesis circuit. FIG. 9 is a block diagram of a modified embodiment of the color operation circuit. FIG. 10 is a timing chart of the circuit of FIG. R _n , B _n , G _n ... Image data of one-dimensional color sensor, C _n ... Print data of cyan, 1... Color CCD sensor, 7.

Claims (2)

(57) [Claims] 1. A detecting means for detecting a color pixel using a one-dimensional color sensor in which a set of three kinds of elements for detecting three colors is arranged in a predetermined order and a plurality of sets of elements are arranged one-dimensionally. And a color operation processing device that performs a primary conversion of a set of three-color image data detected by the one-dimensional color sensor into three-color print data of one pixel for printing. In the arithmetic processing unit, a set S _i (i = 1,2,2,3) of three color elements for detecting image data to be converted into print data of one pixel.
..) As a plurality of elements C _j constituting a one-dimensional color sensor
(J = 0,1,...) To (C _j , C _{j + 1} , C _{j + 2} ) _Pj (j = 0,1,
…)), Print data required for printing for each of C _j , C _{j + 1} , C _{j + 2} is calculated at a speed higher than the transport speed of the input color data, and is sequentially output. Color image reading system. 2. A color image reading system according to claim 1, wherein said color arithmetic processing device converts a set of image data (R, G, B) detected by the element into a conversion coefficient. The image data (R, G, or B) serially sent from the detection means to convert the print data into one color D by using (r, g, b) according to the conversion formula of D = r, R, g, G, b, B ) Is multiplied by the corresponding transform coefficient (r, g or b); three adding means connected in parallel to the multiplying means for sequentially adding the multiplied values sent from the multiplying means; When the multiplication data is sent serially, the adding means is provided with different timing for each adding means,
Addition control means for adding 0 to each of the three multiplication data and adding the value to the multiplication value sent from the multiplication means instead of the previous addition result, and holding the print data calculated by the addition means And an output means for sending the image to an output device.