JPS62186653A - Picture reader - Google Patents

Abstract

PURPOSE:To ensure the reading of a prescribed color component by correcting a picture of a discriminated color into a white or black picture at the time of the part of the discriminated color is positioned singly at the part of black/ white picture. CONSTITUTION:A picture is spectrally diffracted into a discriminated color and its complementary color respectively, the result is photoelectrically transduced by line image sensors 8, 9, and whether or not the discriminated color is decided by comparing the output signal of the line image sensors by picture element. As to the picture element discriminated, a color picture signal is formed by applying a prescribed calculation 20 to a discriminated color picture signal and a complementary color picture signal and the complementary color picture signal is outputted as a black/white picture signal. If the discriminated color picture element is detected in a pseudo way in the part of the black/white picture, exclusive OR circuits 45, 46 convert a pseudo red picture element into a white or black picture element for the output.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field] The present invention relates to an image reading device that reads color information of images other than black and white.

[Prior Art] When inputting an image to a facsimile device, an image processing device, etc., it is often convenient if the color information of the image can also be determined.

For example, if a transmission document to be transmitted by a facsimile machine has red writing, a stamp, or a red underline, if this red information can be output, more effective image information transmission will be possible.

By the way, many methods have been proposed for distinguishing colors in this way, but most of them involve color separation of one color using one filter or the like.

However, in such a conventional method, if we consider the case where a red color is separated by a red filter and an image is formed on a line image sensor in order to distinguish red, for example, an image of a color containing more than a certain amount of red components, such as magenta, is read. When the output level of the line image sensor exceeds a predetermined threshold level, it may be determined as red, which is an inconvenience in that colors other than colors that are visually recognized as red are also recognized as red. was there.

Further, the background color of a typical document is white, so the output level of the line image sensor is the highest when reading the background. For this reason, when a red image signal is formed from only the red component light reception signal as described above, the output level of the line image sensor when reading the red part is close to the level of the white part, so the red color of the image is There has also been an inconvenience in that areas where the color is unclear or faint are not determined to be red but are determined to be white.

Furthermore, the color area that is visually recognized as red is wide;
This visual characteristic is different from the spectral characteristic when spectrally analyzed using a single filter, and as a result, a color that is visually recognized as red may not be recognized as red by some devices, resulting in the inconvenience that an appropriate red image signal cannot be output. There was also.

[Objective] The present invention has been made in order to solve the above-mentioned disadvantages of the conventional technology, and it uses a predetermined color component other than black and white components.
An object of the present invention is to provide an image reading device that can read images reliably.

[Structure] In order to achieve this object, the present invention separates an image into a discrimination color and its complementary color, photoelectrically converts them using a line image sensor, and compares the output signals of these line image sensors pixel by pixel. It is determined whether a pixel is of a discrimination color or not, and for pixels for which the discrimination has been made, a color image corresponding to the discrimination color is generated by performing a predetermined calculation on the image signal of the discrimination color and the image signal of the complementary color. form a signal.

Other pixels are determined to be black and white pixels, and complementary color image signals are output as black and white image signals. Furthermore, the image quality is improved by correcting the pixels of the discrimination color that appear pseudo-in the monochrome image part due to the reading error of the optical system into white pixels or black pixels.

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

FIG. 1 shows an optical system of an image reading device according to an embodiment of the present invention, and in this embodiment, red is used as the discrimination color.

In the figure, an original to be read 1 is illuminated with an image on its reading line by a light source 2, and the image light reflected from the reading line is reflected by a mirror 3 and guided to a lens 4, and then a half mirror or prism. A luminous flux splitting device 5 such as
It is divided into two parts.

The image light thus branched is separated through a red filter 6 that transmits a red component and a cyan filter 7 that transmits a cyan component, and then images are formed on line image sensors 8 and 9.

Note that the red filter 6 has a transmission limit wavelength of, for example, 580 nm+, and has a characteristic of transmitting light of wavelengths higher than that, and the cyan filter 7 has a transmission limit wavelength of, for example, 620 nm, and has a characteristic of transmitting light of wavelengths lower than that. It has the property of being transparent.

Now, when an image is divided into a red component and its complementary cyan component in this way, all colors are divided into these two components and appear as a predetermined level in the output signal of the line image sensor 8゜9. .

Therefore, when we measured the levels of the red component and cyan component of commercially available color samples such as red writing utensils, we found that these color samples were located in the shaded area in FIG. In addition, in FIG. 2, the output level of the line image sensors 8 and 9 when reading the white part of the background of the document where the light receiving level is the highest is 1, and the output level of the line image sensor 8 when reading the actual image is assumed to be 1. The output level of .9 is normalized and displayed.

That is, the red color appearing in the original image satisfies the relationship expressed by the following equation.

Kc/Kr<a, Kr>β where Kc is the normalized level of the cyan component (i.e., the output level of the line image sensor 9), and Kr is the level of the normalized red component (i.e., the level of the line image sensor 8). ) Hail. Also, according to actual measurements, α=
0.4. β=0.4. Note that this value is based on the case where the red filter 6 and cyan filter 7 having the characteristics as described above are used, and this value may change when filters with different characteristics are used.

From the above, the line image sensor 8 is used for each pixel.
.

Further, as the black and white image signal for determining black and white pixels other than red pixels, the output signal from the line image sensor 9 may be used as is.

By the way, when an image is read by the two line image sensors 8 and 9 in this way, the reading positions of the respective line image sensors 8 and 9 may shift. For example, if the reading position shifts in the sub-scanning direction, the same image cannot be read, and if the reduction ratio and reading start position of each of the line image sensors 8 and 9 shifts, even if the same image is read. The reading position is shifted by several bits.

Normally, the positional deviation of the line image sensors 8 and 9 is corrected by adjusting the mounting position and reduction ratio of the line image sensors 8 and 9, but it is impossible to reliably align a small deviation of about 1 bit. This is extremely difficult and may leave a difference of about 1 bit.

Therefore, for example, if the reading positions of line image sensors 8 and 9 in the sub-scanning direction are the same, and the reading start position of line image sensor 8 is one bit ahead of the reading start position of line image sensor 9, then 3. Output levels of the line image sensor 8 and line image sensor 9 when reading a black and white image as shown in Figure 3(a) (
(normalized) changes as shown in (b) and (c) of the figure.

At this time, the red image signal R calculated according to the above-mentioned criteria
D and the monochrome image signal BW become as shown in (d) and (e) in the same figure, and therefore, a pseudo red pixel is inserted at the point where the image changes from a black pixel to a white pixel9.As a result, the read image When recorded, a problem arises in that red pixels are scattered in a black and white image.

Here, Table 1 shows the combinations of the logical states of the red pixel signal RD and the black and white pixel signal BW, and the discrimination results of red pixels, white pixels, and black pixels.

” In order to solve this problem, when a pixel that is independently identified as a red pixel appears in a monochrome image, that red pixel is determined to be a pseudo red pixel, and this pseudo red pixel is replaced with a white pixel or a black pixel. It is sufficient to perform correction processing to correct the pixels.

That is, for example, when the red image signal RD and the black and white image signal 8 are distributed in the main scanning direction and the sub-scanning direction in the pattern shown in Table 2, the center pixel is determined to be a pseudo red pixel, and that pixel is determined as a white pixel. All you have to do is correct it pixel wise. In addition,
In this example, an area of three pixels in each of the main scanning direction and the sub-scanning direction is used as one judgment unit, but this area may be cut out in other ways.

Beast ■ ■ ■ ■ However, the main scanning direction is the direction of A-+84C, and the sub-scanning direction is the direction of A-+D)G. Therefore, pixel A,
B, C (A', B', C'') are central pixels E (E')
It is a pixel l lines before the pixel G, H, T (G'
, H', I') are the pixels one line after the center pixel E (E'), and the pixel D (D') is the pixel after the center pixel E (E').
'), and pixel F (F') is one pixel after the center pixel E (E'). Furthermore, the pixels indicated by "/" in Table 2 indicate that the pixels are not directly related to the determination of pseudo-red pixels.

FIG. 4 shows an image reading device u according to an embodiment of the present invention that realizes the principle of the present invention described above. In addition,
In this figure, drive control systems for the line image sensors 8 and 9 and control systems such as the sub-scanning mechanism, which are not directly related to the present invention, are omitted.

In the figure, image signals PL and P2 output from line image sensors 8 and 9 are applied to either a white waveform memory 13.14 or a normalization circuit 15.16 by switches U and tZ.

The switching devices 11 and 12 are controlled by control means (not shown).

When reading a reference white image (for example, a white original), the white waveform memories 13 and 14 are selected. As a result, at this time, the line image sensor 8
, 9 output the image signals Pi and P2 from the white waveform memory 1.
3.14.

Further, the switching devices 11 and 12 are controlled by the above-mentioned control means.
When reading the image of the original document 1, the normalization circuit 15.
6 normalization circuit 15 switched to select 16
．． 16 converts the level of the image signals Pi and P2 obtained when reading the original 1 using the signals added from the white waveform memory 13 and 14 in synchronization with the image signals PL and P2 as a reference, and The normalization circuits 15 and 16 absorb the difference in transmittance between the red filter 6 and the cyan filter.

That is, the shading correction circuit SDI that performs shading correction on the image signal P1 output from the line image sensor 8 by the switch 11, the white waveform memory 13, and the normalization circuit 15 is also connected to the switch 12. The white waveform memory 14 and the normalization circuit 16 each form a shading correction circuit SD2 that performs shading correction on the image signal P2 output from the line image sensor 9.

The output signals R and C of the normalization circuits 15 and 16 are applied to the color calculation circuit 20.

The color calculation circuit 20 executes the processing shown in FIG. 5 to determine the color of each pixel, and forms a red image signal R1 and a black and white image signal Bl.

That is, first, the ratio of the input signals R and C is calculated (step 101), and it is determined whether the ratio K is larger than a predetermined value α (determination 102).

If the result of this judgment 102 is YES, then the signal R
is larger than a predetermined value β (determination 103
). Note that the predetermined values α and β in this case are both 0.4 as described above.

When the result of this judgment 103 is '/ES, the pixel in question is a red pixel, so the color calculation circuit 20 sets the logic level rlJ to the red pixel signal R1 (104), and outputs the black and white pixel signal corresponding to this pixel. B1 has logic level “0”.
” (processing 105), and outputs the red image signal R1 and the monochrome image signal B1.6 When the result of judgment 102 is NO, and 1 judgment 10
If the result of step 3 is NO, the pixel in question is other than a red pixel,
That is, since it is either a white pixel or a black pixel, the red pixel signal R1 has a logic level "O" indicating that it is not red.
106), and the signal C is set for the monochrome image signal B1.
is binarized at a predetermined threshold level, the logic level of the result is set (binarization processing 107), and each is output.

In this way, the red component and black and white component of the image on the read original 1 are converted into a red image signal R1 and a black and white image signal B1, respectively, and are output.

The red image signal R1 output from the color calculation circuit 20 is applied to a shift register 31 and a line buffer 32 having a capacity capable of storing one line worth of red image signal R1, and the output of the line buffer 32 is applied to a shift register 33 and a line buffer 34. The output of the line buffer 34 is added to the shift register 35.

Further, the black and white image signal nt output from the color calculation circuit 20 is applied to a shift register 36 and a line buffer 37, and the output of the line buffer 37 is applied to a shift register 38 and a line buffer 39.
The output of 9 is applied to a shift register 40.

Addresses to these line buffers 32, 34, 37, and 39 are specified by address data DA output from an address counter 41, and writing/reading is controlled by a control signal RW output from a write/read controller 42. Ru. In addition, the shift register 31,
The shift clock CP output from the address counter 41 is added to 33, 35, 36, 38, and 40.

Address counter 41 and write/read controller 4
2 is a clock signal C output from a control section (not shown)
Its operation is controlled by KI, CK2 and control signal CL.

In this way, the red image signal R1 of one line before the line being added to the shift register 31 is transferred from the line buffer 32 to the shift register 33, and the red image signal R1 of two lines before is transferred from the line buffer 34 to the shift register 33.
5, and the black and white image signal B1 one line before the line added to the shift register 36.
is applied from the line buffer 37 to the shift register 38, and the monochrome image signal B1 of two lines before is applied from the line buffer 39 to the shift register 40.

Therefore, pixels G and H are output from the shift register 31.

A red image signal R1 corresponding to
From the shift register 35, pixels A, 13. The red pixel signal R1 corresponding to pixel C is sent from the shift register 36 to pixel G'
, t+'.

1' is sent to the shift register 38.
From the shift register 40, black and white image signal generators corresponding to pixels D', E', and F'' are output, and from the shift register 40, pixels A', B', and
A black and white image signal corresponding to C' is output.

Then, the red image signal devices corresponding to pixels A, B, C, D, E, F, G, II, and I and the black and white image signal B1 corresponding to pixel E' are stored in the ROM (read-only memory) 44. has been added.

Further, the black and white image signals B1 corresponding to pixels D' and E' are applied to an exclusive OR circuit 45, and the black and white image signals B1 corresponding to pixels 11' and ++' are respectively applied to an exclusive OR circuit 46. , the outputs of these exclusive OR circuits 46 and 47 are added to the ROM 44. Note that the pixels A-I and A' to I' herein correspond to each pixel in Table 1 and Table 2.

The ROM 44 stores the center pixel E based on the relationship shown in Table 1.
Based on the combination of the logical state of the red pixel signal R1 corresponding to the center pixel E' and the logical state of the black and white pixel signal B1 corresponding to the center pixel E', it is determined whether the pixel is a red pixel, a white pixel, or a black pixel. When the relationship shown in Table 1 and Table 2 is detected, a logical operation is performed to determine the pixel as a white pixel, and the results are output as a red image signal RD and a black and white image signal BW.

In this way, shift registers 31, 33, 35, 3
6°38.40, line buffer 32, 34, 37, 3
9. Address counter 42. Write/read control unit 4
3. ROM44° and exclusive OR circuit 45.4
6 forms a color determination circuit 30 that determines the color of each pixel from the output of the color calculation circuit 20 and corrects pseudo red pixels.

By the way, the exclusive OR circuit 45 detects ■■ in the case of Table 2, and the exclusive OR circuit 46 detects ■■ in the case of Table 2.

That is, in this embodiment, since pixel color discrimination is performed according to the relationship shown in Table 1, there is no case where the red image signal R1 and the black and white image signal B1 become "1" at the same time. Therefore, when the logic states of pixels D' and F' are different, if pixel R is "1", pixel D' is always "0", and pixel R is rO
J, the pixel D' will always be "1". Therefore, even if the logical states of pixels D' and F' are not directly known, pixel D'
As long as it is known that the logical states of and F' are different, it is possible to determine the case ■■ in Table 2.

Therefore, the exclusive OR circuit 45 selects pixels O' and F'.
The exclusive OR of the black and white image signals B1 is formed and the result is input to the ROM 44. Furthermore, pixel H'
, B' for the same reason, as long as it is known that their logical states are different, it is possible to determine ■■ in Table 2. Form an exclusive OR and RO the result
I am inputting it into M44. Also, this allows ROM/1
The number of inputs required for the ROM 44 is reduced, and the storage capacity required for the ROM 44 can be suppressed.

In the above-described embodiment, the area is cut out by three pixels each in the main scanning direction and the sub-scanning direction.

Although the pseudo red pixel is determined based on the pattern of the area adjacent to the center pixel of the area, the combination of the area to be cut out and the pattern determined as the pseudo red pixel is not limited to the above.

Further, in the above embodiment, when a pseudo red pixel is detected, it is corrected to a white pixel, but the pseudo red pixel may be corrected to a black pixel. Furthermore, the pseudo red pixel may be corrected to either a white pixel or a black pixel depending on the state of the black and white image surrounding the pseudo red pixel.

Furthermore, in the embodiment described above, the color calculation circuit performs the binarization process, and the occurrence of pseudo red pixels is detected using the binarized red image signal and black and white image signal. This can be done even with the signal intact.

Note that although the above-described embodiment discriminates the red color of an image, the present invention can also be applied to a device that discriminates other colors. Further, the means for dividing an image into color components is not limited to the above-mentioned method.

[Effects] As explained above, according to the present invention, an image is separated into a discrimination color and its complementary color, photoelectrically converted by a line image sensor, and the output signals of these line image sensors are compared pixel by pixel. It is determined whether a pixel is of a discrimination color or not, and for pixels for which the discrimination has been made, a color image signal corresponding to the discrimination color is generated by performing a predetermined calculation on the image signal of the discrimination color and the image signal of the complementary color. , and other pixels are distinguished as black and white pixels, and complementary color image signals are output as black and white image signals.Furthermore, when a portion of the discrimination color is located independently in a portion of a black and white image, this pixel is Since the image quality is improved by correcting white pixels or black pixels, an advantage is obtained that predetermined color components other than black and white components can be read reliably and with high quality.

[Brief explanation of drawings]

FIG. 1 is a schematic configuration diagram showing an example of an optical system of an image reading device according to an embodiment of the present invention, and FIG. 2 is a graph diagram for explaining the principle of determining red color. FIG. 3 is a waveform diagram for explaining the generation of pseudo red pixels, FIG. 4 is a block diagram showing an example of a device according to an embodiment of the present invention, and FIG. 5 is an example of processing of a color calculation circuit. This is a brooch yard showing 6...Red filter, 7...Cyan filter. 8.9...Line image sensor, 11.12...
・Switcher. 13.14... White waveform memory, 15, 16... Normalization circuit, 20... Color calculation circuit, 30... Color determination circuit. 31.33,35,36,38,40...shift register, 32.34,37,39...line buffer,
42... Address counter, 43... Write/read control section, 44... ROM (read only memory), 45.46... Exclusive OR circuit. Agent Patent Attorney Crest 1) Makoto Figure 1 Figure 2 Figure 1J, 4 Yatsume No. 1 Figure 3 ＾See White White White↓ Gumatagu 7 Figure 5

Claims (1)

[Claims] a spectroscopy means for separating the image light on the reading line into a predetermined first color and a second color complementary to the first color; and receiving the image light of the first color and the second color. two line image sensors that output image signals corresponding to each pixel; a color discrimination means that discriminates the first color based on a comparison of the image signals output from these line image sensors; and this color discrimination means. For pixels for which there is a discrimination output from the color discrimination means, a color image signal corresponding to the first color is calculated from the image signal, and pixels for which there is no discrimination output from the color discrimination means are distinguished as black and white pixels, and the pixels for which there is a discrimination output from the color discrimination means are discriminated as black and white pixels, color separation means for outputting an image signal corresponding to the color image signal as a black and white image signal; and color correction means for correcting a pixel of the color image signal to a white pixel or a black pixel when a predetermined pattern is located around the pixel. An image reading device comprising: