JP3147234B2 - Color image reader - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a color image reading apparatus, and more particularly, to a color image reading apparatus that performs dot-sequential color separation of picture elements and reads them.

[Prior Art] A conventional color image reading apparatus reads a picture element by, for example, color separation into an RGB signal. The color separation method mainly includes a three-color illumination sequential switching method, a three-line color sensor method, and a dot-sequential color sensor method.

[Problems to be Solved by the Invention] In the three-color illumination sequential switching method, a monochrome sensor of one line and illumination of three colors of RGB are provided, and an image is read by changing illumination three times for each line. However, this method does not increase the image reading speed.

The three-line color sensor system includes RGB filters and three line sensors provided for each of the RGB, and includes a contact type and a reduction type. In the close contact type, each line sensor can be arranged without a gap, but at present it is very expensive and has a great effect on the device cost. On the other hand, in the reduced type, since there is a gap, each of the RGB signals is captured with a time delay. Therefore, a buffer memory is required,
Also, if there is vibration in the scanning mechanism, the time delay will fluctuate,
Causes image degradation.

The dot-sequential color sensor system uses an RGB filter and a one-line sensor provided in a dot-sequential manner for a picture element. According to this system, RGB color information can be extracted in one line. However, the total number of pixels in one line is limited due to manufacturing restrictions, and there is a problem that a sufficient resolution cannot be obtained.
That is, since the RGB filters are adjacent to each other, a pixel pitch of about 20 microns is required, and since the length that can be taken out from the wafer is determined, the total number of pixels is only about 2500. Since the number of picture elements is about 840, the reading density of four picture elements per millimeter for the width of the A4 document is only small. So-called 4pel (100dpi).

As a method of solving this and increasing the reading density, there is a method of arranging a plurality of sensors in a staggered manner. However, this method not only raises costs, but also has many problems, such as correction of sensitivity unevenness at a joint portion and mounting accuracy of each sensor.

The present invention is to solve the above-mentioned problem,
An object of the present invention is to obtain a read signal more faithful to the color of an original image in a color image reading apparatus using a reading unit in which a plurality of image sensors corresponding to a plurality of color components are repeatedly arranged in a point-sequential manner.

[Means for Solving the Problems] The present invention has the following configuration as one means for achieving the above object.

That is, the color image reading apparatus according to the present invention has a reading unit in which a plurality of image sensors that respectively correspond to a plurality of color components and output color component signals of the corresponding color components are repeatedly arranged in a point-sequential manner, A color image reading device that outputs a plurality of color component signals for each pixel of the picture element, wherein each of the imaging elements is a pixel, and pixels of a plurality of color components continuous in the dot sequential direction in the reading means are one picture element. The plurality of picture elements indicated by the color component signal by referring to a color component signal output from the reading means for a plurality of picture elements of a target pixel and an adjacent pixel adjacent to the target pixel in the dot-sequential direction; Detecting an edge of an image present in a pixel, detecting means for detecting the position as a position between pixels, and, if an edge exists on the pixel of interest, based on the edge position, A plurality color component signals of the serial target picture element pixel, and an outputting means for outputting the color component signals for each pixel of the plurality pixels selectively.

Preferably, when an edge exists on the pixel of interest, the output unit outputs a plurality of color component signals of a pixel adjacent to the pixel as a plurality of color component signals of pixels located at both ends of the pixel of interest. And selecting one of the plurality of color component signals of the adjacent picture element based on the position of the edge as the multi-color component signal of the pixel at the center of the target picture element.

[Description of Embodiments] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

FIG. 6 is a sectional view showing a reading mechanism of the color image reading apparatus of the embodiment. In the figure, a fluorescent lamp 1 and a mirror 2
The first mirror unit 3 on which the mirror is mounted and the second mirror unit 6 on which the mirrors 4 and 5 are mounted move at a two-to-one speed by driving the sub-scanning motor unit. 10 is a platen glass,
Reference numeral 11 denotes a document pressing plate. As described above, the reflected image of the original 7 illuminated by the fluorescent lamp 1 is transmitted through the mirrors 2, 4, 5 and the lens 8,
Color sensor (CC
D) Image on 9.

FIG. 7 is a diagram showing details of the light receiving surface of the CCD 9. In the figure, on the light receiving surface of the CCD 9, each pixel of the light receiving surface (dot sequential)
, A red filter (R) 12, a green filter (G) 13, and a blue filter (B) 14 are repeatedly provided. One pixel pitch is 20 microns and the total number of effective pixels is 2550 pixels. The lens 8 projects an image of 0.083 mm on the original onto one pixel on the CCD 9. That is, the image of 4pel on the original is the pixel of 12pel on the CCD
The relationship is read by B, R, and G.

FIG. 14 is a diagram schematically showing the relationship between the original color projected on the CCD 9 and the raw output of the CCD 9. Below, the raw output (pixel data) of 5 bits (32 gradations) from CCD9 is on the dark side
A description will be given of a case where the bright side corresponds to level 0 in FIG.

When the original color is read by the CCD 9 as shown in FIG. 14, since the resolution and the like of the lens 8 are actually finite, the influence between picture elements appears in the raw output signal. Although there is no problem in the range of the same color, the influence between picture elements cannot be ignored at the boundary between white and black or black and white. For example, assuming that the effect of the color directly above the pixel is 80% and the effect of the adjacent color is 20%, the first raw output R at the boundary between white and black is white R = 0 and black R = 30, R = 0 × 0.8 + 30 × 0.2 = 6. The raw output G is given by G = 0 × 0.2 due to G = 0 for white and G = 30 for black.
+ 30 × 0.8 = 24. Further, at the next boundary between black and white, B = 30 × 0.8 + 0 × 0.2 = 24 and R = 30 × 0.2 + 0
× 0.8 = 6. Therefore, if the read data is directly output to an external printer or the like without performing any processing on such a raw output signal, a reproduced image faithful to the original color can no longer be obtained. The solution of this problem will be described in detail below according to this embodiment.

FIG. 1 is a block diagram of a signal processing section of a color image reading apparatus according to an embodiment. In the figure, a read signal from a CCD 9 is amplified by an amplifier 15 and then converted into 5-bit pixel data by an A / D converter 16. Reference numeral 17 denotes a data alignment unit mainly composed of a latch circuit, which sets pixel data serially sent from the A / D converter 16 into one set for each BRG, and sets a target picture element D1 (B
1, R1, G1) and the picture elements D0 (B0, R0, G0) and D2 (B2, R2,
G2). Reference numeral 20 denotes a select signal generation circuit, which performs detection processing of edge, on-pixel, thin line, boundary position, and shading, which will be described later, on each of the pixel data D0, D1, and D2 sent from the data alignment unit 17. And a select signal A reflecting the detection result. Reference numeral 60 denotes a data transmission circuit, which includes a memory for storing several sets of data strings obtained by rearranging the BRG data of the picture element data D0, D1, and D2. A set of data corresponding to A is transmitted. With this configuration, the original color arrangement pattern is sequentially detected from the raw output signal of the CCD 9, and the data of the set closest to the original color arrangement pattern is read from the memory according to the detection result, and the target pixel is reproduced. It is sent out sequentially at the timing.

FIG. 2 is a circuit diagram showing details of the select signal generation circuit 20. In the figure, reference numeral 30 denotes a basic signal generator, which is described later with comparators 41 to 46, an edge detector 31, an ON picture element detector 32, a boundary position detector 35, a thin line detector 33, and a density detector 34.
Consists of The basic signal generation section 30 detects the arrangement pattern of the original color by each of the above-described sections, and generates basic signals S1 to S8 indicating the detection result. Reference numeral 50 denotes a condition setting unit that generates appropriate select signals A1 to A6 for reading out a set of data closest to the original color array pattern from the memory in accordance with the state of generation of the basic signals S1 to S8.

<Basic Signal Generation Unit 30> The comparators 41 to 46 calculate a difference between predetermined pixel data, and output a logical 1 when the result is equal to or more than a set value (for example, 10).

The edge detecting section 31 is a circuit for detecting that there is an original color edge on the target picture element D1 or on the boundary of the target picture element D1, and outputs a logic 1 to the edge detection signal S1 when an edge exists.

The on-pixel detecting section 32 is a circuit for detecting that the document color has not changed at least on the target pixel D1, and outputs a logic 1 to the on-pixel detection signal S2 when detecting the state of the on-pixel. I do.

The thin line detection unit 33 is a circuit for detecting that all or a part of the thin line of the original color exists on the target picture element D1, and outputs a logical 1 to the thin line detection signal S3 when the thin line is detected.

The boundary position detection unit 35 is a circuit that detects which of the values of the pixels R0 and R2 in the picture elements on both sides thereof is closer to the value of the center pixel R1 of the target picture element D1, and the value of R1 is closer to R0. At the time, a logic 1 is output to the boundary position detection signal S4, and when the value of R1 is close to R2, a logic 1 is output to the boundary position detection signal S5.

The density detection circuit 34 is a circuit that, when a thin line due to a dark portion or a light portion is detected, detects a position of such a thin line as viewed from the target picture element D1. The details of the density detection circuit 34 will be described with reference to FIG.

FIG. 3 is a circuit diagram showing details of the density detection circuit 34.
In the figure, the density detection circuit 34 includes a density detection section 70, a light section detection section 72, and a priority setting section 75.

In the dark part detection unit 70, reference numerals 701 to 707 denote comparators, each of which outputs a logic 1 when the pixel signal R0, G0, B1, R1, G1, B2 or R2 is equal to or greater than a set value (for example, 22). 711-7
Reference numeral 15 denotes an AND circuit, each of which outputs a logical AND of three consecutive comparator outputs as shown. As a result, the position of the dark portion can be determined.

In the light part detector 72, reference numerals 721 to 727 denote comparators, each of which outputs a logic 1 when the pixel signal R0, G0, B1, R1, G1, B2 or R2 is equal to or less than a set value (for example, 10). 731-7
An AND circuit 35 outputs a logical AND of three consecutive comparator outputs as shown in the figure. As a result, the position of the light portion can be determined. Then, the dark part detecting unit 70 and the light part detecting unit 72
No conflict occurs between the corresponding AND circuits (for example, 711 and 731), and these are logically ORed by the OR circuits 741 to 745 and input to the priority setting unit 75.

The priority order setting unit 75 assigns priorities to the outputs of the OR circuits 741 to 745. For example, when the output of the OR circuit 741 (the target picture element D1) is logic 1 (when there is a dark portion or a light portion), the output of the OR circuit 742 is invalidated. When the output of the OR circuit 741 is logic 0, the output of the OR circuit 742 has the following priority.
When the output of 752 becomes logic 1, the output of OR circuit 743 is invalidated. The output of the OR circuit 741 becomes the gray level detection signal S6. The outputs of the AND circuits 752 and 755 are passed through the OR circuit 762 to detect the gray level detection signal S7.
become. The outputs of the AND circuits 754 and 751 become the gray level detection signal S8 via the OR circuit 761.

<Condition setting unit 50> In FIG. 2, the density detection signals S6 to S8 are thin line detection signals.
It becomes valid when S3 is logic 1 (AND circuits 80, 81, 82). The edge detection signal S1 becomes valid when both the ON picture element detection signal S2 and the output S61 of the AND circuit 80 are at logic 0 (AND circuit 84).
As for the density detection signals S7 and S8, the output S10 of the AND circuit 84 is also logic 1
Becomes effective at the time of (AND circuits 81 and 82). Boundary position detection signal
In S4 and S5, the signal S10 is logic 1 and the thin line detection signal S3 is logic 0.
Becomes valid at the time of (AND circuits 85 and 86). The select signal A1 is a signal S11 obtained by inverting the signal S10,
A2 is an output signal S41 of the AND circuit 85. Select signal A3
The output signal S52 of the AND circuit 87 indicates that the select signal A2 has priority over the select signal A3. The select signal A4 is the output signal S71 of the AND circuit 81 and the select signal A4
A5 is an output signal S81 of the AND circuit 82. The select signals A1 to A5 are all logic 0, and the edge detection signal S1
Is logic 1, the select signal A6 becomes logic 1 (NAND
Circuit 89).

The operation of the condition setting section 50 will be more clear from the description with reference to a specific example described later.

<Data Transmission Circuit 60> FIG. 4 is a block diagram of the data transmission circuit 60.
In the figure, reference numeral 67 denotes a selector, and pixel data D0, D1, D2
Pixel data (B0, R0, G0), (B1, R1, G1), (B2, G2,
R2) is allocated to each of the memories 61 to 66 in an array pattern as shown. That is, the memory 61 stores the set of picture element data (D1, D1, D1
Remember 1). The memory 62 stores a set of picture element data (D0, D0, D
2) Remember. The memory 63 stores a set of picture element data (D0, D2, D
2) Remember. Here, the pixel data D3 to D6 will be described.

FIGS. 5A and 5B are diagrams for explaining the pixel data D3 to D6. In the figure, a pair d3 of three consecutive pixels (G0, B1, R1) around the pixel B1 of the target picture element D1 is taken, and
Rearrange in BRG order to obtain picture elements D3 (B1, R1, G0). Less than,
Similarly, pairs d4 to d6 are taken as shown in the figure, and each is rearranged in BRG order to obtain picture elements D4 to D6. Thus, D0 → D5 → D3
A picture element (window) having a phase shift of one pixel at a time in the order of → D1 → D4 → D6 → D2 is obtained.

Referring back to FIG. 4, the memory 64 stores a set of picture element data (D3, D3, D
6) Remember. The memory 65 stores a set of picture element data (D5, D4, D
4) Remember. The memory 66 stores a set of picture element data (D3, D1, D
4) Remember. Then, the selector 68 selects any one of the sets of the picture element data in the memories 61 to 66 according to the select signals A1 to A6 (relationship in Table 69) at the timing of reproducing and outputting the target picture element D1. Output.

8 to 13 are diagrams schematically showing the relationship between the original color projected on the CCD 9 and the target picture element D1. Here, for convenience of explanation, 100 is a black area of the document, and 200 is a white area. The picture element of interest is D1 (B1, R1, G1), and the picture elements before and after it are D0 (B0, R0, G0) and D2 (B2, R2, G2). Hereinafter, the operation of each detecting means will be described.

<Edge Detection> In the example of FIG. 8, there is no edge on the target picture element D1 or on the boundary of the target picture element D1. Therefore | B1-B2 | ≧ 0 and | G1-G0 | ≧
Since 10 is not satisfied, the edge detection signal S1 becomes logic 0. In the example of FIG. 9 to FIG.
(Fig. 11) or on the boundary (Figs. 11 and 12). Therefore, | B1−B2 | ≧ 10 or | G1−G0 | ≧ 10 is satisfied, and the edge detection signal S1 becomes logic 1.

Here, consider a colored original image. For example, 10 in Fig. 10
When 0 is black and 200 is yellow, | B1−B2 | ≧ 10 is not satisfied because neither contains blue (B). But yellow 200
Contains green (G1), | G1−G0 | ≧ 10 is satisfied. Therefore, edges are detected at the black and yellow boundaries. On the other hand, when the case where 100 is black and 200 is red is considered, since green (G1) is not included in red 200, | G1−G0 | ≧ 10 is not satisfied. Therefore, no edge is detected at the boundary between black and red. As described above, in the boundary portion of a document color such as red and black, blue and black, and red and blue, which is generally dark and has low contrast, it is necessary to accurately detect and reproduce such a boundary. In this embodiment, such a boundary is not detected because it does not significantly affect the boundary.

<On-picture element detection> In the example of FIG. 11, an edge is detected, but the edge is not on the target picture element D1. When reproducing such a picture element, it is desired to output the contents of the target picture element D1 as it is. The ON picture element detection circuit 32 is a circuit that detects such a state. In the example of FIG. 11, since neither | G1−G0 | ≧ 10 and | R1−R0 | ≧ 10 are satisfied, the on-picture element detection signal S2 becomes logic 1. G0
The same applies to the case of white up to B1 and black after B1. In this case, since neither | B1−B2 | ≧ 10 and | R1−R2 | ≧ 10 are satisfied, the ON picture element detection signal S2 becomes logic 1.

<Thin Line Detection> In the example of FIGS. 8 to 10, the vicinity of the target pixel D1 is wide black.
Generally, | B1−B0 | ≧
10 and | G1−G2 | ≧ 10 are not satisfied. Accordingly, the thin line detection signal S3 becomes logic 0. However, in the examples shown in FIGS. 12 and 13, since the target picture element D1 is covered by all or part of the narrow black 100, | B1−B0 | ≧ 10 or | G1−G2 | ≧ 10 is satisfied. I do. Therefore, the thin line detection signal S3 becomes logic 1.

In addition, in the example of FIG.
However, since | G1−G2 | ≧ 10 is satisfied, the thin line detection signal S3 becomes logic 1. However, at the same time, since the on-picture element detection signal S2 becomes logic 1, the signal S10 in FIG. 2 becomes logic 0, and the thin line detection signal S3 in this case is effectively invalidated (AND circuits 81, 82).

<Boundary position detection> Edge is detected, not a thin line, and attention pixel D1
May change color (ie, it is not an ON picture element). In this case, it is desirable to reflect the color of the next picture element when reproducing the target picture element D1. For example, when there is a boundary between skin color and black on the target picture element D1, at least the signals at both ends of the set of three signals output when reproducing the target picture element D1 are set to the skin color and black, respectively. The center signal is determined by the boundary position detection circuit 35 to determine which color to use. That is, the boundary position detection circuit 35 is a circuit that detects which of the center pixels R0 and R2 of the picture elements D0 and D2 adjacent to the picture element D1 of interest is closer to the center pixel R1. In the example of FIG. 9, black 100 extends to B1. Therefore, | R1−R0 | ≧ 10 is satisfied, and the boundary position detection signal S4 becomes logic 0. On the other hand, | R1
Since −R2 | ≧ 10 is not satisfied, the boundary position detection signal S5 becomes logic 1. Therefore, in the example of FIG. 9, the color of R1 is close to the picture element D2 side. In the example of FIG. 10, the black 100 extends to R1. Therefore, since | R1−R0 | ≧ 10 is not satisfied, the boundary position detection signal S4
Becomes logical 1. Since | R1−R2 | ≧ 10 is satisfied, the boundary position detection signal S5 becomes logic 0. Therefore, in the example of FIG.
The color of R1 is close to the pixel D0 side. Note that the boundary position detection signals S4 and S5
Are both logic 1, the S4 side is prioritized in the AND circuit 87 of FIG.

<Shade Detection> In the example of FIG. 13, there is a thin black line 100 across the target pixel D1 and the preceding pixel D0. In such a case, if the complete black line 100 is reproduced by the picture element D0 or the picture element D1, faithful reproduction of the original image cannot be obtained. Thus, in the present embodiment, the picture element d3 (G
0, B1, R1), that is, D3 (B1, R1, G0). The density detection circuit 34 shown in FIG.
This circuit searches for a picture element (dark part) with a large signal for all three pixels or a picture element (light part) with a small signal for all three pixels from a set of pixels.

In the example of FIG. 13, since the set of d3 (G0, B1, R1) is large,
The output of the AND circuit 712 shown in FIG. Also in this example
Since the set d6 (G1, B2, R2) is small, the output of the AND circuit 735 in FIG. Accordingly, the outputs of the OR circuits 742 and 745 both become logic 1, and both pass through the priority setting section 75 while being valid, and the gray level detection signal S7 becomes logic 1. The density detection signal S7 satisfies the AND circuit 81 of FIG.
To logic 1. Thereby, the target picture element D1 in FIG. 13 is reproduced by the data set (D3, D3, D6) read from the memory 64. That is, as apparent from FIG. 5 (A), the first two pixels of the target picture element D1 are reproduced twice by the data D3 of the phase d3, and the remaining one pixel of the target picture element D1 is
It is reproduced once by d6 data D6. As a result, a fine line faithful to the original is reproduced.

FIG. 15 is a diagram for explaining the relationship between an example of a document color and a reproduction color. In the figure, reference numeral 300 denotes a color signal column representing the basic colors on the document and their original BRG components, reference numeral 301 denotes a document color column schematically representing the arrangement of the document colors, reference numeral 302 denotes a CCD9 read pixel array column, 303 is a picture element number column attached to a picture element sequentially, 304 is a column showing a raw output signal of CCD9, 305 is a column showing the presence or absence of edge detection, 306 is a column showing the presence or absence of on-picture element detection, and 307 is a thin line A column indicating the presence or absence of detection, a column 308 indicates a state of boundary position detection, a column 309 indicates a state of density detection, a column 310 indicates a select signal generated for each pixel of interest,
1 is a column showing a set of output data selected by the select signal, and 312 is a column showing an arrangement of reproduced colors. Raw output field 304
To explain, at the boundary between skin color and black in picture element No. 3, B = 16 × 0.
8 + 30 × 0.2 = 18.8 ≒ 19. Also, R = 10 × 0.2 + 30 × 0.8 = 26 due to the skin color R = 10 and the black R = 30.
Thus, a raw output signal of a component different from the original document color appears. Hereinafter, the processing of this embodiment will be described according to the picture element numbers.

<Picture element No. 2> The edge detection column 305 is "none" because | B1-B2 | = | 16-19 | = 3 <10 | G1-G0 | = | 18-18 | = 0 <10.

When no edge is detected, the other detections become substantially invalid (indicated by a symbol / in the figure), and the select signal A1 becomes logic 1 preferentially, so that the memory 61 is selected and the three sets of data (D1, D1 , D1) is sent out. Therefore, reproduction of this picture element is faithful to the original document color.

<Picture element No. 3> The edge detection column 305 is “Yes” because | B1−B2 | = | 19−30 | = 11 ≧ 10 | G1−G0 | = | 30−18 | = 12 ≧ 10.

The ON picture element detection column 306 is “none” because there is an edge on the target picture element.

The thin line detection field 307 is "none" because | B1-B0 | = | 19-16 | = 3 <10 | G1-G2 | = | 30-30 | = 0 <10. That is, a thin line (black or white) that can be called a thin line is not present in the target picture element.

 The boundary position detection column 308 is close to “R2” because | R1−R0 | = | 26−10 | = 16 ≧ 10 | R1−R2 | = | 26−30 | = 0 <10.

The density detection column 309 is substantially invalidated by “none” in the thin line detection column 307.

As a result, the select signal A3 becomes logic 1, and the memory 63
Is selected and three sets of data (D0, D2, D2) are transmitted.
Therefore, even if the raw output signal of the CCD 9 is affected by the flesh-black boundary, the reproduced output after processing is faithful to the original document color.

<Picture No. 4> The edge detection field 305 is "none" because | B1-B2 | = | 30-27 | = 3 <10 | G1-G0 | = | 30-30 | = 0 <10. Therefore, select signal A1 is logic 1
And the memory 61 is selected, and three sets of data (D1, D1, D
1) is sent.

<Picture element No. 5> The edge detection column 305 is “Yes” because | B1−B2 | = | 27−16 | = 11 ≧ 10 | G1−G0 | = | 18−30 | = 12 ≧ 10.

The ON picture element detection column 306 is “none” because there is an edge on the target picture element.

The thin line detection column 307 is “Yes” because | B1−B0 | = | 27−30 | = 3 <10 | G1−G2 | = | 18−28 | = 10 ≧ 10. That is, a part of the skin-colored thin line is over the target picture element.

The boundary position detection column 308 is substantially invalidated by “Yes” in the thin line detection column 307.

In FIG. 3, R0 ≧ 22, G0 ≧ 22, B1
Since both satisfies ≧ 22, this indicates that the dark portion was detected at d5 in FIG. 5A, and is “D5”. R1 ≦ 1
Since both 0, G1 ≦ 10 and B2 ≦ 10 are satisfied, this indicates that a light portion was detected at d4 in FIG. 5A, and is “D4”. As a result, a signal S8 is output from the priority setting unit 75. As a result, the select signal A5 becomes logic 1, and three sets of data (D5, D4, D4) of the memory 65 are transmitted.

Looking at the reproduction output, the BRG of D5 is (27, 30, 30), which is slightly different from the original original black (30, 30, 30). This is because the boundary of the original color is slightly affected by the optical performance, but there is no problem since the color is actually reproduced as almost black. Similarly, D4 is slightly different from the flesh color of the document. However, this is also almost skin tone. Although there are some differences as described above, for example, the size of one pixel on printing is (1/12) mm, and there is almost no influence.

<Picture No. 6> Similar to picture element No. 5, a thin line is detected and the dark part is “D6”
Is detected, the select signal becomes A4, and three sets of data (D3, D3, D6) are transmitted.

<Picture element No. 7> Although a fine line is detected, the dark portion is detected by "D1", so that the density detection signal S6 is satisfied and the AND circuit 80 of FIG. 2 is satisfied. This invalidates the AND circuit 84, and the select signal A1
Is output. Accordingly, the memory 61 is selected, and three sets of data (D1, D1, D1) are transmitted.

<Picture element No. 8> The ON picture element detection field 306 becomes “Yes” by | B1−B2 | = | 19−16 | = 3 <10 | R1−R2 | = | 10−10 | = 0 <10 Become. If ON picture element is detected, AN in Fig. 2 will be displayed.
The D circuit 84 is blocked, and the select signal A1 becomes logic 1.
Therefore, three sets of data (D1, D1, D1) are transmitted.

As described above, the original document color 301 was reproduced very faithfully as shown by the reproduction color 312.

FIG. 16 is a diagram showing reproduced colors in the case where no processing is performed on the original color of FIG. In the case of no processing, reproduction is performed using the select signal A1, that is, three sets of data (D1, D1, D1).
In the picture element No. 3 in FIG. 16, as a result of the CCD 9 reading the boundary image, the raw output BRG is (19, 26, 30), but since this is output as it is, the dark blue color is actually reproduced. Similarly, picture element No. 5 becomes an eyebrow color, picture element No. 6 becomes shrimp tea, and is reproduced in a completely different color. Also, since the presence or absence of an edge, the location of the edge, and the like are not detected, the unit of reproduction is naturally the unit of picture element, and the texture of the reproduced image is coarse. On the other hand, when the processing is performed as shown in FIG. 15, the output data can be changed in units of one-third the size of one picture element, that is, in units of pixels, and the original can be accurately reproduced.

FIG. 17 is a view for explaining the relationship between another example of the document color and the reproduction color. In this example, the processing of a document having fine lines finer than FIG. 15, that is, black and white thin lines having a width of 3 pixels (about 0.25 mm) and a width of 2 pixels (about 0.17 mm) is shown. In the figure, edges are detected in all picture elements No. 2 to No. 8, and fine lines are detected. In the picture elements NO.2 to NO.6, a dark part or a light part is detected, and predetermined processing is performed on each of them. However, if the line width of the document is less than three pixels, the document is not reproduced in the color of the document. For example, in picture element No. 3, a light portion (white) is detected at the phase of d5, but the subsequent data d4, that is, D4, becomes (B, R, G) = (6,24,24). Not black. This is blue.

Further, in the picture elements No. 7 and No. 8, the original color is a series of thin lines having a width of 2 pixels, and there is no part where all the three pixels are large or small, and therefore, no shaded part is detected in the shade detection. At this time, all the signals S6, S7, S8 in FIG. 2 become logic 0, and the output A6 of the NAND circuit 89 eventually becomes logic 1. As a result, three sets of data (D3, D1, D4) are transmitted. This data array pattern is a pattern for outputting three consecutive pixels centered on each pixel of the target picture element.
This is a suitable pattern to reflect the state of NO.7 and NO.8. As a result, even if black and white cannot be clearly reproduced, they can be reproduced with a considerable density difference, and the outline is correctly reproduced.

In this way, black and white may be reproduced in blue and green, and white in yellow and light blue, especially in a document with continuous thin fine lines.However, the boundaries between shades are reproduced correctly, and the outline is composed of pixels. To obtain a detailed image.

FIG. 18 is a diagram showing reproduced colors in the case where no processing is performed on the document color of FIG. Looking at the reproduction state shown in FIG. 18, not only the line width and position are incorrect, but also the white line on the manuscript over picture elements Nos. 7 to 8 has disappeared from the reproduction. Therefore, according to the present embodiment, a very fine thin line (about 0.17 mm)
This has the effect that the reproducibility is extremely good even for a document having a series of characters. Next, as another example, a case where the contrast is low will be described.

FIG. 19 is a view for explaining the relationship between the original color consisting of white and light blue and the reproduction color. In picture element No. 3, ON picture element detection is "Yes" because | B1-B2 | = | 1-7 | = 6 <10 | R1-R2 | = | 24-30 | = 6 <10. By the way, looking at the original color, there is a border between white and light blue on the picture element No. 3,
Normally, this attention pixel is not in the on-pixel state, but B2
Is light blue, B2 = 7, which is small, and as a result, the ON picture element detection is “Yes”. To cope with such a case, it is possible to cope with it by increasing the judging conditions. Conversely, the fact that the two data are close means that the colors are also close. However, there is no problem in color. Therefore, no special countermeasure is taken in this embodiment. For this reason, in the picture element No. 3, the ON picture element is detected, the select signal becomes A1, and the data of the target picture element D1 is transmitted as it is. However, light blue is originally a color having a small B value, and there is almost no chromatic effect even if the B portion of the pixel of interest has a low value once. However, the position of the light blue reproduction pixel is shifted by one pixel. The same applies to picture element No. 5. However, in the case of the picture elements No. 6 and No. 8, the on-picture element is not detected, and the boundary position of the place where the normal processing is performed is detected.
The boundary is correctly reproduced.

As described above, when the contrast of the document is low, an error of one pixel may occur in some cases. However, an error of one pixel in a low contrast portion, that is, an error of 0.083 mm, is hardly present in an actual image. Has no effect.

In the above description, three sets of pixel data are output when one picture element is reproduced. However, the present invention is not limited to this. For example, it is easy to use two sets, and if four or five sets are used, there is an effect that the boundaries can be reproduced more finely.

FIG. 20 is a diagram for explaining a case where five sets of pixel data are output for one picture element. In the picture element No. 1, the color is black up to the boundary of the pixel B, but in the picture elements Nos. 2 and 4, the color is black up to the middle of the pixel G or the pixel R. For this reason, CC
The raw output of D9 is smaller than that of pixel NO.
In the case of small. For example, G of picture element No. 2 is 16. Therefore, in the example shown in the figure, when G = 16, only one set of black signals is transmitted, and when B = 24, two sets of black signals are transmitted. As described above, the difference in the raw output of the CCD 9 also occurs due to the subtle difference in the position of the boundary, and it is possible to consider five sets of output using this.

As described above, according to the present embodiment, a boundary is formed in a high contrast part at a pixel unit, that is, at a pitch of 1/12 mm. Further, when the contrast is low, for example, even in the boundary between white and light blue, an error of one pixel occurs in some cases, and the original document color is reproduced with high fidelity. Also, thin lines (0.17 mm or less) less than one picture element can be reproduced with distinct shading. That is, an image is reproduced at a pixel pitch which is 1 / of one picture element pitch, and a reproduced image that is fine and sharp is obtained. For example, in the dot-sequential sensor used in the present embodiment, the number of picture elements usually translates an A4 original into 4 pel (100d
pi), but according to the invention 12pel (300
dpi) and can be played back.

Although the above-described embodiment has been described with respect to a scanner for reading a document image, the present invention can also be applied to an imaging apparatus such as a video camera.

As described above, according to the present embodiment, since each pixel of the dot sequential sensor is effectively used, the original image can be read and reproduced at a pitch finer than the pixel pitch, and a fine reproduced image faithful to the original can be obtained. Can be provided. Also,
Since it can be implemented with a simple circuit configuration and does not involve the complexity of optical adjustment, the reading resolution can be increased without a significant increase in cost.

[Effects of the Invention] As described above, according to the present invention, a plurality of image pickup elements (pixels) corresponding to a plurality of color components, respectively, are arranged in a dot-sequential direction in a dot-sequentially-reading unit. When a pixel of a color component is set as one picture element and a plurality of color component signals are output for each pixel of the picture element, if there is an edge on the target picture element, the edge position is expressed as a position between pixels. The color component signals of the pixels of the pixel of interest and the pixels of the plurality of pixels of the pixel adjacent to the pixel of interest in the dot-sequential direction are selectively output as the multiple color component signals of the pixels of the pixel of interest. Therefore, in the color image reading apparatus using the above-mentioned reading means, a reading signal more faithful to the color of the document image to be read can be obtained.

[Brief description of the drawings]

FIG. 1 is a block diagram of a signal processing section of a color image reading apparatus according to an embodiment, FIG. 2 is a circuit diagram showing details of a select signal generation circuit 20, and FIG. 3 is a circuit diagram showing details of a density detection circuit 34. FIG. 4 is a block diagram of the data transmission circuit 60, FIGS. 5A and 5B are diagrams for explaining the pixel data D3 to D6, and FIG. 6 is a reading mechanism of the color image reading apparatus of the embodiment. FIG. 7 is a view showing details of the light receiving surface of the CCD 9, FIG. 8 to FIG. 13 are original colors projected on the CCD 9 and target picture elements.
FIG. 14 is a diagram schematically showing the relationship between D1 and D1, FIG. 14 is a diagram schematically showing the relationship between the original color projected on the CCD 9 and the raw output of the CCD 9, and FIG. FIG. 16 is a diagram illustrating the relationship, FIG. 16 is a diagram showing a reproduction color in the case of no processing with respect to the document color of FIG. 15, FIG. 17 is a diagram illustrating a relationship between another example of the document color and the reproduction color, FIG. 18 is a view showing a reproduction color in the case of no processing with respect to the original color of FIG. 17, FIG. 19 is a view for explaining a relation between the original color composed of white and light blue and the reproduced color, FIG. FIG. 6 is a diagram for explaining a case where five sets of pixel data are output for one picture element. In the figure, 9: CCD, 12: Red filter, 13: Green filter, 14: Blue filter, 17: Data alignment unit, 20
…… Select signal generator, 30 …… Basic signal generator, 31
... Edge detection section, 32... On-picture element detection section, 33... Thin line detection section, 34.
... Condition setting unit, 60 data transmission circuit, 61 to 66 data storage memory, 67, 68 selector, 70 dark part detecting unit, 72 light part detecting unit.

Claims (2)

(57) [Claims] A plurality of image pickup devices corresponding to a plurality of color components, each of which outputs a color component signal of the corresponding color component. A color image reading device that outputs a plurality of color component signals for each pixel of the picture element, using pixels of a plurality of color components continuous in the dot sequential direction in the reading means as one picture element; The color component signal is referred to as a pixel of interest and a plurality of pixels adjacent to the pixel of interest in the dot-sequential direction, and the image of the image present in the plurality of pixels indicated by the color component signal is referred to. Detecting means for detecting an edge and detecting the position as a position between pixels; and, when an edge exists on the target pixel, a plurality of color component signals of the pixel of the target pixel based on the edge position. As a color image reading apparatus characterized by an output means for outputting the color component signals of each pixel of the plurality pixels selectively. 2. The image processing apparatus according to claim 1, wherein, when an edge exists on the pixel of interest, a plurality of color component signals of pixels located at both ends of the pixel of interest are output as a plurality of color component signals of a pixel adjacent to the pixel. A component signal is selected, and a plurality of color component signals of a pixel at a central portion of the pixel of interest selects one of the plurality of color component signals of the adjacent pixel based on the position of the edge. 2. The color image reading device according to 1.