JP5590911B2 - Image reading apparatus and method - Google Patents

Description

  BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image reading apparatus mounted on a scanner, a FAX, a copying machine, and the like, and in particular, interpolates data between sensor chips when a document is read by a line sensor in which a plurality of sensor chips are linearly arranged. The present invention relates to an image reading apparatus and a method for outputting the same.

  In a contact image sensor in which a plurality of sensor chips are linearly arranged, the boundary between adjacent image pickup elements (pixels) across the boundary between adjacent chips is set to be twice the pixel pitch in the sensor chip, and the boundary There is known a technique that interpolates image data between pixels adjacent to each other with respect to (see, for example, Patent Document 1).

JP 2003-101724 A (pages 5-6, FIGS. 3-5)

  However, in the above-described prior art, a curve of a quartic expression using the average value of the pixel values of two adjacent pixels across the boundary or using the pixel values of four pixels adjacent across the boundary. Therefore, there is a problem that accurate interpolation cannot be performed when a high-frequency component is included in the document image.

The present invention has been made to solve the above-described problems, and the image reading apparatus of the present invention includes:
An image reading apparatus having a line sensor for reading a document,
A plurality of color sensor chips each having a plurality of RGB image sensors, and a color image sensor that outputs a first electric signal photoelectrically converted by the RGB image sensors;
A first AD converter for converting the first electrical signal into first digital data;
A first shading correction unit that corrects characteristic variations for each element of the image sensor included in the first digital data and outputs the corrected data as first corrected data;
Has a monochrome sensor chip having a plurality of monochrome image pickup device, and a monochrome image pickup unit for outputting a second electrical signal that the monochrome image pickup device is photoelectrically converted,
A second AD converter for converting the second electrical signal into second digital data;
A second shading correction unit that corrects characteristic variations for each element of the monochrome imaging element included in the second digital data and outputs the corrected data as second corrected data;
A pixel interpolation unit that performs interpolation on the first corrected data using the second corrected data,
In the plurality of color sensor chips, an interval between adjacent image sensor elements of adjacent color sensor chips among the plurality of color sensor chips is between the image sensor elements in the color sensor chip. It is arranged to be twice the interval,
The monochrome sensor chip of the monochrome image pickup unit, the next Ri suit the color sensor chip, in time for the end adjacent to one another, in the color sensor chip in missing position can not read the document, to read the document Are arranged so that
The pixel interpolation unit
Generating brightness data from the first corrected data;
The included in the second corrected data, and wherein the I row interpolation for the previous SL luminance data by using the data generated by reading the document in the monochromatic sensor chip in the missing position .

  According to the present invention, accurate interpolation can be performed even when a high-frequency component is included in an image of a document.

1 is a block diagram illustrating a configuration of an image reading apparatus according to a first embodiment of the present invention. It is a figure for demonstrating the positional relationship of the color sensor chip which the color imaging part 1C has, and the monochrome sensor chip which the monochrome imaging part 1M has. It is an enlarged view of the boundary part of mutually adjacent color sensor chips 6C1 and 6C2. 3 is a block diagram illustrating a configuration example of a pixel interpolation unit 4. FIG. (A)-(c) is a figure which shows an example of the image of the original read, and the data obtained by imaging. It is a block diagram which shows the structure in the case of implement | achieving the process of the shading correction | amendment part and pixel interpolation part of FIG. 1 by software. (A) And (b) is a flowchart which shows the procedure of the process implemented by CPU of FIG.

Embodiment 1 FIG.
FIG. 1 is a diagram showing a configuration of an image reading apparatus according to Embodiment 1 of the present invention.

  The image reading apparatus according to Embodiment 1 includes a color imaging unit 1C, a first AD conversion unit 2C, a first shading correction unit 3C, a monochrome imaging unit 1M, a second AD conversion unit 2M, 2 shading correction unit 3M and pixel interpolation unit 4.

The color imaging unit 1C includes a plurality of color sensor chips each having a plurality of RGB imaging elements, photoelectrically converts reflected light from the document, and outputs electrical signals S1R, S1G, and S1B.
A color filter that transmits only the wavelength of red light is placed on the surface of the image sensor on the R image sensor, and the R image sensor receives only red light from the reflected light from the document. An electric signal S1R corresponding to the amount of light is output.
Similarly, a color filter that transmits only the wavelength of green light is also placed on the surface of the image sensor on the G image sensor, and the G image sensor receives only the green light of the reflected light from the document. An electric signal S1G corresponding to the amount of light is output.
Similarly, a color filter that transmits only the wavelength of the blue light is also placed on the surface of the image pickup device for the B image pickup device, and the image pickup device for B receives only the blue light of the reflected light from the document. An electric signal S1B corresponding to the amount of light is output.
The electrical signals S1R, S1G, and S1B are input to the first AD converter 2C.

The monochrome imaging unit 1M has a plurality of monochrome sensor chips each having a plurality of monochrome imaging elements so that a document portion located at the boundary of the plurality of color sensor chips of the color imaging unit 1C can be read. Has been placed.
The monochrome imaging unit 1M photoelectrically converts reflected light from the document and outputs an electrical signal S1M corresponding to the amount of reflected light. The electric signal S1M is input to the second AD converter 2M.

FIG. 2 shows the positional relationship between the color sensor chip included in the color imaging unit 1C and the monochrome sensor chip included in the monochrome imaging unit 1M.

In each of the color sensor chips 6C1 to 6C6, a plurality of sets of RGB image sensors are linearly arranged, and the color sensor chips 6C1 to 6C6 are linear in the same direction as the arrangement direction of the image sensors in each sensor chip. Is arranged.
In each of the monochrome sensor chips 5M1 to 5M5, a plurality of monochrome imaging elements are linearly arranged, and the monochrome sensor chips 5M1 to 5M5 are linearly arranged in the same direction as the arrangement direction of the imaging elements in each sensor chip. It is arranged.
The arrangement direction of the monochrome sensor chips 5M1 to 5M5 is the same as that of the color sensor chips 6C1 to 6C6 (parallel to each other), and the monochrome sensor chips 5M1 to 5M5 are arranged in parallel to the color sensor chips 6C1 to 6C6. .

  Although FIG. 2 shows an example of an image reading apparatus having a configuration of six color sensor chips and five monochrome sensor chips, the present invention is not limited to this.

  Further, although FIG. 2 shows eight sets of RGB image sensors and eight image sensors of monochrome sensors in the color sensor chip, the present invention is not limited to this.

  An interval between adjacent image sensors (pixels) sandwiching a boundary between adjacent color sensor chips is larger than a pixel pitch in the chip.

Each of the monochrome sensor chips 5M1 to 5M5 is arranged such that one of the image sensors is aligned at a position where the RGB image sensors do not exist at the boundary between the adjacent color sensor chips.
For example, the monochrome sensor chip 5M1 is arranged so that one of the imaging elements is aligned at a position where the RGB imaging elements do not exist at the boundary between the color sensor chips 6C1 and 6C2 adjacent to each other.
Similarly, the monochrome sensor 5M2 is arranged so that one of the image sensors is aligned at a position where the RGB image sensors do not exist at the boundary between the color sensor chips 6C2 and 6C3 adjacent to each other.
Generally speaking, the i-th (i = 1 to 5) monochrome sensor chip 5Mi has one of its image sensors, the boundary between the i-th color sensor chip 6Ci and the (i + 1) -th color sensor chip 6C (i + 1). And are arranged so as to be aligned at positions where RGB image sensors do not exist.
The total number (I) of monochrome sensor chips 5Mi is smaller by 1 than the total number (I + 1) of color sensor chips 6Ci.

  The color sensor chips 6C1 to 6C6 and the monochrome sensor chips 5M1 to 5M5 are scanned (moved) in the direction of the arrow 7 or in the opposite direction (direction perpendicular to the sensor chip arrangement direction) to read the original. Conversely, the document may be read by conveying the document in the direction of arrow 7 or in the opposite direction. The relative movement direction or conveyance direction of the sensor chip and the document is referred to as the sub-scanning direction, and the arrangement direction of the imaging elements in the sensor chip (the arrangement direction of the imaging elements of the same color in the color sensor chip, the imaging element in the monochrome sensor chip) ) Is sometimes referred to as the main scanning direction.

FIG. 3 is an enlarged view of the boundary between the color sensor chips 6C1 and 6C2. The detailed positional relationship between the color sensor chip and the monochrome sensor chip will be described with reference to FIG. The horizontal axis is the horizontal position (main scanning direction), and the vertical axis is the vertical position (sub-scanning direction).
The position in the horizontal axis direction is represented by k-2, k-1, k, k + 1, k + 2, etc. as indices, and the position in the vertical axis direction is j-2, j-1, j, j + 1, j + 2, etc. as indices. It is represented by The image sensors of the same color in the color sensor chip are arranged at equal intervals in the horizontal direction, and this interval is referred to as one pixel interval and is represented by “1”. The image sensors of different colors in the color sensor chip are arranged at equal intervals in the vertical direction, and this interval is referred to as one pixel interval and is represented by “1”. However, the dimension is not necessarily the same as “1” in the horizontal direction and “1” in the vertical direction.

On the color sensor chips 6C1 and 6C2, an R image sensor is arranged at the vertical position j, a G image sensor at the vertical position j-1, and a B image sensor at the vertical position j-2. In the monochrome sensor chip 5M1, an image sensor is arranged at a vertical position j + 2.
Since the frame region 8 exists at the periphery of the image sensors of the color sensor chips 6C1 and 6C2, the image sensors between the sensor chips cannot be arranged at the same pitch as in the sensor chip. Therefore, the image sensor at the right end of the color sensor chip 6C1 is arranged at the horizontal position k-1, and the image sensor at the left end of the color sensor chip 6C2 is arranged at the horizontal position k + 1. Then, the color sensor chips 6C1 and 6C2 cannot read the document portion at the horizontal position k. This horizontal position is called a missing position.

However, the boundary between the color sensor chips adjacent to each other, and the interval between the image sensors adjacent to each other across the boundary is set to twice the interval between the image sensors in the color sensor chip, and the corresponding positions between the color sensor chips. By interpolating data at (missing position), pixel data at equally spaced positions can be obtained. That is, if the interpolation data is handled in the same way as the sampling data, the sampling interval becomes substantially uniform.

  The monochrome sensor chip 5M1 is arranged at a horizontal position where a document at a horizontal position k which is a missing position can be read. Since each of the color sensor chip and the monochrome sensor chip has a frame region at the peripheral portion, the color sensor chip and the monochrome sensor chip are arranged in a row of R imaging elements of the color sensor chip (a row of imaging elements in the color sensor chip). Among them, the row of image pickup elements located on the side closest to the monochrome sensor chip) and the row of image pickup elements of the monochrome sensor chip are separated at an interval twice the vertical pixel interval in the color sensor chip. Be placed. Therefore, the image sensor of the monochrome sensor chip 5M1 is disposed at the vertical position j + 2.

  By setting the position of the image sensor of the color sensor chip and the image sensor of the monochrome sensor chip to an integer multiple of the pixel interval in the vertical direction in the color sensor chip, the sampling interval including the interpolation data becomes constant. It is possible to easily align the data acquired by the image sensor and the data acquired by the image sensor of the monochrome sensor chip.

  The first AD converter 2C outputs digital data D2R obtained by analog-digital conversion of the electric signal S1R, digital data D2G obtained by analog-digital conversion of the electric signal S1G, and digital data D2B obtained by analog-digital conversion of the electric signal S1B. Digital data D2R, D2G, and D2B are input to the first shading correction unit 3C.

  The second AD converter 2M outputs digital data D2M obtained by analog-digital conversion of the electrical signal S1M. The digital data D2M is input to the second shading correction unit 3M.

The first shading correction unit 3C corrects the characteristic variation for each element of the image sensor included in the color imaging unit 1C for each RGB color.
The first shading correction unit 3C
For the digital data D2R, output first corrected data D3R in which the characteristic variation of each image sensor of R is corrected;
For the digital data D2G, output first corrected data D3G in which the characteristic variation of each G image sensor is corrected;
For the digital data D3B, first corrected data D3B obtained by correcting the characteristic variation of each B image sensor is output.
The first corrected data D3R, D3G, and D3B are input to the pixel interpolation unit 4.

  The second shading correction unit 3M corrects variation in characteristics of each image sensor of the monochrome imaging unit 1M. The second shading correction unit 3M outputs second corrected data D3M obtained by correcting the characteristic variation for each element of the image sensor with respect to the digital data D2M. The second corrected data D3M is input to the pixel interpolation unit 4.

  By the shading correction by the first shading correction unit 3C and the second shading correction unit 3M, the level difference between the digital data D2R, D2G and D2B and the digital data D2M is also adjusted.

  Since the detailed operation of the shading correction units 3C and 3M is a known technique, detailed description thereof is omitted.

The pixel interpolation unit 4 interpolates the sensor chip boundary data (data at the horizontal position k) included in the first corrected data D3R, D3G, and D3B using the second corrected data D3M, and outputs image data D4R. , D4B and D4B are output.
Image data D4R, D4G, and D4B are output as output data of the image reading apparatus.

  Next, the configuration of the pixel interpolation unit 4 will be described in detail. FIG. 4 is a diagram illustrating a configuration example of the pixel interpolation unit 4.

  The pixel interpolation unit 4 includes a vertical alignment unit 8, a luminance / color difference conversion unit 9, a color difference interpolation unit 10, a luminance interpolation unit 11, and an RGB conversion unit 12.

The first corrected data D3R, D3G, and D3B and the second corrected data D3M are input to the vertical alignment unit 8. As shown in FIG. 3, the vertical positions of the RGB image sensor of the color sensor chip and the image sensor of the monochrome sensor chip are shifted. Therefore, since the document positions to be read simultaneously are different, it is necessary to align the positions in the vertical direction. When the image reading apparatus is scanned in the direction of the arrow 7, when the G image sensor reads the document portion at the vertical position j-2, the B image sensor reads the document portion at the vertical position j-3. It is necessary to store the data of the image pickup element B read at -2. Similarly, when the R image sensor reads the original portion at the vertical position j-2, the B and G image sensors read the original portions at the vertical positions j-4 and j-3, respectively. It is necessary to store the data of the B and G image sensors. Further, when the image sensor of the monochrome sensor chip reads the document portion at the vertical position j-2, the image sensors at B, G, and R respectively read the document portions at the vertical positions j-6, j-5, and j-4. Therefore, it is necessary to store the data of the B, G, and R image sensors read at the vertical position j-2. The vertical alignment unit 8 has a storage element such as a line memory inside, and uses the data stored for 4, 3, and 2 lines of R, G, and B data read by the color sensor chip, respectively. Adjust. With this storage, the vertical alignment unit 8 delays the first corrected data D3R, D3G, and D3B data by a time corresponding to 2, 3, and 4 lines, respectively, and thereby the first corrected vertical position is aligned. The post-alignment data D8R, D8G and D8B and the second post-alignment data D8M are output. The first aligned data D8R, D8G, and D8B are input to the luminance / color difference conversion unit 9, and the second aligned data D8M are input to the luminance interpolation unit 11.

  The luminance / color difference conversion unit 9 converts the first aligned data D8R, D8G, and D8B into luminance data D9Y and color difference data D9R and D9B. Conversion from RGB to luminance color difference is performed using equation (1). However, Formula (1) is an example of a conversion formula from RGB to luminance color difference, and the conversion formula is not limited to Formula (1). The luminance data D9Y is input to the luminance interpolation unit 11, and the color difference data D9R and D9B are input to the color difference interpolation unit 10.

D9Y = 0.3 × D8R + 0.59 × D8G + 0.11 × D8B
D9R = 0.7 × (D8R−D8G) −0.11 × (D8B−D8G)
D9B = 0.89 × (D8B-D8G) −0.3 × (D8R-D8G)
... (1)
Note that the calculation of Expression (1) is performed using RGB data after alignment of pixels at the same horizontal position, and the calculation results are used as luminance data and color difference data for pixels at the same horizontal position. It is done.

The color difference interpolation unit 10
Interpolate the data corresponding to the boundary of the color sensor chip of the color difference data D9R (data at the horizontal position k) and output the color difference interpolation data D10R,
Data corresponding to the boundary of the color sensor chip of the color difference data D9B (data at the horizontal position k) is interpolated to output color difference interpolation data D10B.
The color difference interpolation data D10R and D10B are input to the RGB conversion unit 12.
The color difference data is interpolated using the average value of the data corresponding to the left and right positions of the boundary. For example, referring to FIG. 3, the color difference interpolation data at the horizontal position k is
D10R (k) = (D9R (k-1) + D9R (k + 1)) / 2
D10B (k) = (D9B (k-1) + D9B (k + 1)) / 2
... (2)
Given in.

The luminance interpolation unit 11 interpolates data corresponding to the boundary of the color sensor chip of the luminance data D9Y (data at the horizontal position k) using the second aligned data D8M, and outputs luminance interpolation data D11Y. The luminance interpolation data D11Y is input to the RGB conversion unit.
Interpolation of the luminance data is performed by inserting a position corresponding to the boundary of the color sensor chip of the data the luminance data D9Y position corresponding to the boundary of the color sensor chip of the second alignment post data D8m. That is, interpolation is performed by a method in which the value of the second aligned data D8M is used as it is as the value of the luminance data D9Y. For example, referring to FIG. 3, the luminance interpolation data D11Y at the horizontal position k is
D11Y (k) = D8M (k) (3)
Given in.

To perform the interpolation as described above, it is necessary to keep performing level adjustment for aligning the signal level of the second alignment after data D8m the signal level of the bright Dode over data D8m. Such level adjustment is performed by shading correction by the shading correction units 3C and 3M.
Since the output of the image sensor other than the image sensor at the position aligned with the position (missing position) of the color sensor chip among the image sensors of the monochrome sensor chip is used for the above level adjustment, it is shown in FIG. Thus, even if the image sensor is not arranged over the same length as the color sensor chip, it is sufficient to use an image sensor having an image sensor only in a position aligned with the missing position and a region in the vicinity thereof. Also, when using a monochrome sensor chip in which the image sensor is arranged over the same length as the color sensor chip, only output signals from the image sensor at the position aligned with the missing position and the image sensor in the area in the vicinity thereof are used. May be used (supplied to the AD conversion unit 2M).

The RGB converter 12 converts the luminance interpolation data D11Y and the color difference interpolation data D10R and D10B into RGB data D12R, D12G, and D12B. The luminance color difference is converted to RGB using Equation (4). However, Formula (4) is an example of a conversion formula from luminance color difference to RGB, and the conversion formula is not limited to Formula (4). The RGB data D12R, D12G, and D12B are output from the pixel interpolation unit 4 as image data D4R, D4G, and D4B, respectively.
D12R = D10R + D11Y
D12G = (0.59 × D11Y−0.3 × D10R−0.11 × D10B) /0.59
D12B = D10B + D11Y
... (4)

  Next, the effects of the present invention will be described using specific examples. FIGS. 5A to 5C are diagrams for explaining a specific example. FIG. 5A shows a white and black vertical stripe read original (“black” is indicated by hatching), and FIG. 5B shows first corrected data D3R, D3G corresponding to FIG. D3B, FIG. 5C shows second corrected data D3M corresponding to FIG. 5A. The horizontal axis of FIG. 5A is the sampling position of the read document, the horizontal axis of FIGS. 5B and 5C is the horizontal position, the vertical axis of FIG. 5B is the first corrected data, The vertical axis of 5 (c) shows the second corrected data. 5B and 5C, the data values on the vertical axis are normalized with the maximum value being “1”. K indicating the horizontal position in FIGS. 5A to 5C corresponds to k in FIG. That is, the color sensor chip cannot read the document at the horizontal position k.

  When the color sensor chip reads the original in FIG. 5A, the original is black and white, so the first corrected data is “0” and “0” for each pixel, as shown in FIG. 5B, for D3R, D3G, and D3B. The data has a periodicity of repeating “1”. However, it is not possible to acquire data at the horizontal position k, which is the boundary between the color sensor chips. On the other hand, when the original of FIG. 5A is read by the monochrome sensor chip, the data has a periodicity of repeating “0” and “1” for each pixel as in FIG. 5B. Further, as shown in FIG. 3, since the monochrome sensor chip is arranged so that the pixel (imaging device) is positioned at the horizontal position k, the monochrome sensor chip can read the document at the horizontal position k.

  The vertical alignment unit 8 aligns the vertical positions of the first corrected data D3R, D3G, and D3B and the second corrected data D3M.

The luminance / color difference conversion unit converts the first post-alignment data D8R, D8G, and D8B into luminance / chrominance data D9R, D9B, and D9Y using Equation (1). When the first post-alignment data D8R, D8G, and D8B are “0”,
D9Y = 0.3 × 0 + 0.59 × 0 + 0.11 × 0 = 0
D9R = 0.7 × (0-0) −0.11 × (0-0) = 0
D9B = 0.89 × (0-0) −0.3 × (0-0) = 0
It becomes. When the first post-alignment data is “1”, from the equation (1),
D9Y = 0.3 × 1 + 0.59 × 1 + 0.11 × 1 = 1
D9R = 0.7 × (1-1) −0.11 × (1-1) = 0
D9B = 0.89 × (1-1) −0.3 × (1-1) = 0
It becomes.

The color difference interpolation unit 10 performs interpolation using the average value of the color difference data corresponding to the left and right pixels of the boundary of the color sensor chip.
D10R (k) = (D9R (k-1) + D9R (k + 1)) / 2 = (0-0) / 2
D10B (k) = (D9B (k-1) + D9B (k + 1)) / 2 = (0-0) / 2

The luminance interpolation unit 11 interpolates the data of the monochrome sensor chip corresponding to the position of the boundary of the color sensor chip into the luminance data corresponding to the position of the boundary of the color sensor chip.
D11Y (k) = D8M (k) = 1

The RGB converter 12 converts the luminance interpolation data D11Y and the color difference interpolation data D10R and D10B into RGB data. When the luminance interpolation data is “0” and the color difference interpolation data is “0”, from Equation (4),
D12R = 0 + 0 = 0
D12G = (0.59 × 0−0.3 × 0−0.11 × 0) /0.59
D12B = 0 + 0 = 0
It becomes. When the luminance interpolation data is “1” and the color difference interpolation data is “0”, from the equation (4),
D12R = 0 + 1 = 1
D12G = (0.59 × 1-0.3 × 0−0.11 × 0) /0.59=1
D12B = 0 + 1 = 1
It becomes.

  As described above, even if the read document is a document including high-frequency information as shown in FIG.

  On the other hand, even if a colored document is a natural image, high-frequency information is aggregated into luminance data. Therefore, even if the color difference data is interpolated with the average value on the left and right of the boundary of the color sensor chip, the image of the present invention is used. The reading apparatus can output image data without degrading the image quality.

  That is, according to the image reading apparatus according to the first embodiment of the present invention, even a line sensor in which a plurality of color sensor chips are linearly arranged can output a read image of an original without damaging information. it can.

Embodiment 2. FIG.
6 shows a configuration in which a part of the image reading apparatus according to the first embodiment is realized as software, that is, by a programmed computer. FIGS. 7A and 7B are executed by the CPU shown in FIG. The procedure of the processing to be performed is shown.
As shown in FIG. 6, the image reading apparatus according to the second embodiment includes a color imaging unit 1C, a first AD conversion unit 2C, a monochrome imaging unit 1M, a second AD conversion unit 2M, a CPU 21, A program memory 22, a data memory 23, a first interface 24, a second interface 25, and a bus 26 for connecting them are provided.

  The CPU 21 operates according to a program stored in the program memory 22. The first interface 24 receives the digital data D2R, D2G and D2B output from the first A / D converter 2C and the digital data D2M output from the second A / D converter 2M, and receives a bus. 26 to the CPU 21. Data subjected to processing by the CPU 21 is output from the second interface 25 via the bus 26. The data memory 23 stores data input from the first interface 24, data to be output from the second interface 25, and data generated during the processing by the CPU 21.

  The processing of the CPU 21 performed using the digital data D2R, D2G and D2B from the first A / D converter 2C and the digital data D2M from the second A / D converter 2M is shown in FIG. As shown, the first shading correction step ST3C, the second shading correction step ST3M, and the pixel interpolation step ST4 are included.

  As shown in FIG. 7B, the pixel interpolation step ST4 includes a vertical alignment step ST8, a luminance color difference conversion step ST9, a color difference interpolation step ST10, a luminance interpolation step ST11, and an RGB conversion step ST12.

The first shading correction step ST3C generates first corrected data D3R, D3G, and D3B obtained by shading correction of the first digital data D2R, D2G, and D2B. This operation is equivalent to the operation of the first shading correction unit 3M.
The second shading correction step 3M generates second corrected data D3M obtained by shading correction of the second digital data D2M. This operation is equivalent to the operation of the second shading correction unit 3C.

  The pixel interpolation step ST4 generates post-interpolation data D4R, D4G, and D4B by interpolating the first post-correction data D3R, D3G, and D3B using the second post-correction data D3M. This operation is equivalent to the operation of the pixel interpolation unit 4.

In the vertical alignment step ST8, the first post-alignment data D8R, D8G, and D8B and the second post-alignment data obtained by aligning the vertical positions of the first post-correction data D3R, D3G, and D3B and the second post-correction data D3M D8M is generated. This operation is equivalent to the operation of the vertical alignment unit 8.
The luminance color difference conversion step ST9 converts the first post-alignment data D8R, D8G, and D8B from RGB to luminance color difference, and generates luminance data D9Y and color difference data D9R, D9B. This operation is equivalent to the operation of the luminance / color difference conversion unit 9.

The color difference interpolation step ST10 generates color difference interpolation data D10R and D10B obtained by interpolating data corresponding to the position of the color sensor chip boundary of the color difference interpolation data D9R and D9B as an average value of the left and right data of the boundary. This is equivalent to the operation of the interpolation unit 10.
In the luminance interpolation step ST11, luminance interpolation data D11Y is generated by interpolating data corresponding to the position of the boundary of the color sensor chip of the luminance data using the second aligned data D8M. This operation is equivalent to the operation of the luminance interpolation unit 11.

The RGB conversion step ST12 generates RGB data D12R, D12G, and D12B obtained by converting the luminance interpolation data D11Y and the color difference interpolation data D10R and D10B from luminance color difference to RGB. This operation is equivalent to the operation of the RGB converter 12.
The above is the operation of the image reading apparatus according to the second embodiment.

  The image reading apparatus according to the second embodiment is obtained by realizing a part of the image reading apparatus according to the first embodiment as software, and the effect is the same.

  1C color imaging unit, 2C first AD conversion unit, 3C first shading correction unit, 1M monochrome imaging unit, 2M second AD conversion unit, 3M second shading correction unit, 4 pixel interpolation unit, 5M1 to 5M5 Monochrome sensor chip, 6C1 to 6C6 color sensor chip, 8 vertical alignment unit, 9 luminance color difference conversion unit, 10 color difference interpolation unit, 11 luminance interpolation unit, 12 RGB conversion unit.

Claims (10)

An image reading apparatus having a line sensor for reading a document,
A plurality of color sensor chips each having a plurality of RGB image sensors, and a color image sensor that outputs a first electric signal photoelectrically converted by the RGB image sensors;
A first AD converter for converting the first electrical signal into first digital data;
A first shading correction unit that corrects characteristic variations for each element of the image sensor included in the first digital data and outputs the corrected data as first corrected data;
Has a monochrome sensor chip having a plurality of monochrome image pickup device, and a monochrome image pickup unit for outputting a second electrical signal that the monochrome image pickup device is photoelectrically converted,
A second AD converter for converting the second electrical signal into second digital data;
A second shading correction unit that corrects characteristic variations for each element of the monochrome imaging element included in the second digital data and outputs the corrected data as second corrected data;
A pixel interpolation unit that performs interpolation on the first corrected data using the second corrected data,
In the plurality of color sensor chips, an interval between adjacent image sensor elements of adjacent color sensor chips among the plurality of color sensor chips is between the image sensor elements in the color sensor chip. It is arranged to be twice the interval,
The monochrome sensor chip of the monochrome image pickup unit, the next Ri suit the color sensor chip, in time for the end adjacent to one another, in the color sensor chip in missing position can not read the document, to read the document Are arranged so that
The pixel interpolation unit
Generating brightness data from the first corrected data;
The included in the second corrected data, and wherein the I row interpolation for the previous SL luminance data by using the data generated by reading the document in the monochromatic sensor chip in the missing position Image reading device.   The monochrome sensor chip of the monochrome imaging unit is arranged so that the document can be read at an intermediate position between imaging elements at adjacent ends of the adjacent color sensor chips.
  The image reading apparatus according to claim 1. The monochrome imaging unit
Among the image sensor rows of the color sensor chip, the distance between the image sensor row located closest to the monochrome sensor chip of the monochrome image pickup unit and the image sensor row of the monochrome sensor chip is a color sensor chip. 3. The image reading apparatus according to claim 1, wherein the image reading device is arranged so as to be an integral multiple of an interval in a direction perpendicular to the direction of the row of the image pickup elements. The pixel interpolation unit
The first aligned data obtained by aligning the first corrected data in the vertical direction and the second aligned data obtained by aligning the second corrected data with the vertical position of the first aligned data. A vertical alignment unit that outputs
A luminance and color difference converting section for converting the first alignment post data to the luminance data and chrominance data,
Using the data obtained by reading the document with the image sensors at the adjacent edges of the adjacent color sensor chips included in the color difference data , the color difference data of the missing position between the edges is generated. A color difference interpolation unit that outputs color difference interpolation data obtained by interpolating the color difference data at the missing position ;
Using the data obtained by reading the original with the monochrome sensor chip at the missing position, which is included in the second post-alignment data , as the missing position luminance data, the luminance data at the missing position is interpolated. A luminance interpolation unit that outputs the luminance interpolation data,
The image reading apparatus according to any one of claims 1 to 3 the color difference interpolation data and the luminance interpolation data; and a RGB conversion section for converting the R GB data. The color difference interpolation unit includes an average value of data obtained by reading a document with image sensors at adjacent ends of the adjacent color sensor chips included in the color difference data, between the ends. The image reading apparatus according to claim 4, wherein the image reading apparatus outputs the color difference data at a missing position . An image reading method using an image reading apparatus having a line sensor for reading an original,
The image reading device is
A plurality of color sensor chips each having a plurality of RGB image sensors, and a color image sensor that outputs a first electric signal photoelectrically converted by the RGB image sensors;
A first AD converter for converting the first electrical signal into first digital data;
Has a monochrome sensor chip having a plurality of monochrome image pickup device, and a monochrome image pickup unit for outputting a second electrical signal that the monochrome image pickup device is photoelectrically converted,
A second AD converter that converts the second electrical signal into second digital data;
A first shading correction step of correcting characteristic variations for each element of the image sensor included in the first digital data and outputting as first corrected data;
A second shading correction step of correcting characteristic variations for each element of the monochrome imaging element included in the second digital data and outputting as second corrected data;
A pixel interpolation step for performing interpolation on the first corrected data using the second corrected data,
In the plurality of color sensor chips, an interval between adjacent image sensor elements of adjacent color sensor chips among the plurality of color sensor chips is between the image sensor elements in the color sensor chip. It is arranged to be twice the interval,
The monochrome sensor chip of the monochrome image pickup unit, the next Ri suit the color sensor chip, in time for the end adjacent to one another, in the color sensor chip in missing position can not read the document, to read the document Are arranged so that
The pixel interpolation step includes:
Generating brightness data from the first corrected data;
The included in the second corrected data, and wherein using the data generated by reading the document in the monochromatic sensor chip at missing position will row interpolation for the previous SL luminance data Image reading method.   The monochrome sensor chip of the monochrome imaging unit is arranged so that the document can be read at an intermediate position between imaging elements at adjacent ends of the adjacent color sensor chips.
  The image reading method according to claim 6. The monochrome imaging unit
Among the image sensor rows of the color sensor chip, the distance between the image sensor row located closest to the monochrome sensor chip of the monochrome image pickup unit and the image sensor row of the monochrome sensor chip is a color sensor chip. 8. The image reading method according to claim 6, wherein the image reading method is arranged so as to be an integral multiple of an interval in a direction perpendicular to the direction of the row of the image pickup devices. The pixel interpolation step includes:
The first aligned data obtained by aligning the first corrected data in the vertical direction and the second aligned data obtained by aligning the second corrected data with the vertical position of the first aligned data. A vertical alignment step that outputs
A luminance and color difference converting step of converting the first alignment post data to the luminance data and chrominance data,
Using the data obtained by reading the document with the image sensors at the adjacent edges of the adjacent color sensor chips included in the color difference data , the color difference data of the missing position between the edges is generated. A color difference interpolation step for outputting color difference interpolation data obtained by interpolating the color difference data at the missing position ;
Using the data obtained by reading the original with the monochrome sensor chip at the missing position, which is included in the second post-alignment data , as the missing position luminance data, the luminance data at the missing position is interpolated. Luminance interpolation step for outputting the luminance interpolation data,
The image reading method according to any one of claims 6 to 8, characterized in that it comprises a RGB conversion step of converting the color difference interpolation data and the luminance interpolation data R GB data. In the color difference interpolation step, an average value of data obtained by reading a document with an image sensor at an end portion adjacent to each other of the adjacent color sensor chips included in the color difference data is calculated between the end portions. The image reading method according to claim 9 , wherein the image difference data is output as the color difference data at a missing position .

Images