JPH07273941A - Image reader - Google Patents

Abstract

PURPOSE:To make the reader low in cost and to eliminate moire in reflected distortion by using a solid-state image pickup device on which two line sensors with a different phase are integrally arranged at the same sampling pitch. CONSTITUTION:Suppose that picture elements of a CCD 101 are arranged on an original at a pitch of 63.5mu, then each signal in R1, G1, B1 lines in a digital signal being an output of an A/D converter 102 is delayed in a delay memory 103 by a distance of subscanning lines of a couple of read lines, the same line read signal as that of R2, G2, B2 lines is obtained, When the CCD 101 with the picture element arrangement as above reads a line chart of 12 lines/mm, an area of an overlapped part between an aperture of the CCD 101 and the chart is obtained as a signal output to cause moire in which the phase is inverted at an interval of nearly 4 lines/mm. When data of a 1st line and a 2nd line are interpolated alternately by an interpolation circuit 106, the sampling frequency of the spectrum is virtually doubled and overlap by reflected distortion is eliminated to avoid the production of moire.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention is an image in which an original is scanned while being irradiated with light, the reflected light is imaged on a solid-state image pickup device, and image information of the original is converted into a one-dimensional electric signal to be read. The present invention relates to a reading device, and more particularly, to an image reading device for a printer that reads a document image discretely.

[0002]

2. Description of the Related Art In an image reading apparatus which samples an original using a one-dimensional line sensor such as a CCD, the spatial frequency band in which the information of the original can be transmitted without distortion is 1/2 (the sampling frequency of the CCD). It is up to the Nyquist frequency). When the information in the frequency band higher than the Nyquist frequency enters the CCD, the information of the original cannot be accurately reproduced due to the aliasing distortion, and the moire,
Image quality deterioration such as color bleeding of thin line portions occurs.

For example, in the case of a 400 dpi scanner, since the sampling pitch is 63.5 μ, the sampling frequency is about 15.7 (lines / mm) and the Nyquist frequency is about 7.9 (lines / mm). This 7.9
If the original contains frequency components of (lines / mm) or more, moiré is generated and image quality is deteriorated. In particular, in the case of a color scanner, moiré due to color, color tint of fine lines, etc. occur, resulting in significant deterioration of image quality.

Conventionally, the devices disclosed in the following publications have been proposed to solve these problems.

In Japanese Patent Laid-Open No. 5-196894, two types of optical low-pass filters are provided in the optical path, and C
By preventing high frequency components from entering the CD,
A method for removing moire has been proposed.

In Japanese Patent Application Laid-Open No. 59-123367, 2
It has been proposed to reduce the moire by arranging two line sensors with a slight shift in the main scanning direction and adopting a high level signal.

In Japanese Utility Model Application Laid-Open No. 52-114556 and Japanese Utility Model Application Laid-Open No. 52-114557, in a color sensor in which RGB pixels are arranged in order, two lines of two color lines and one color line are provided, and two color lines are provided. It has been proposed to arrange the lines of.

In Japanese Patent Laid-Open No. 3-129964, CC
It has been proposed to make the pixel pitch of D non-uniform to suppress moire.

Further, Japanese Patent Laid-Open No. 4-111676 proposes to perform spatial filtering on a signal decomposed into RGB.

[0010]

However, in Japanese Unexamined Patent Publication No. 5-196894, since a low-pass filter is arranged in the optical path, it is necessary to manufacture and adjust with high accuracy, and the device becomes expensive. was there.

Further, in Japanese Patent Application Laid-Open No. 59-123367, two line sensors are arranged with a slight shift in the main scanning direction, but the two sensors are independent and are arranged independently. Therefore, the optical paths are different,
Due to the error in the placement of mirrors in the optical path, the error in the magnification of the lens, etc., it is necessary to make adjustments with a very high degree of precision in order to place them with a very small amount of adjustment, and it is unique to each CCD itself. Since there are twists, warps, bends, pitch irregularities, etc., and it is difficult to accurately arrange the two sensors with each other, there is a problem that accurate sampling may not be performed.

Further, in Japanese Patent Application Laid-Open No. 59-123367, since the signal with the higher level is adopted, the density of the image changes, and the original becomes different from the original. There were also points.

In Japanese Utility Model Application Laid-Open No. 52-114556 and Japanese Utility Model Application Laid-Open No. 52-114557, one set of pixels corresponding to each sensor is not sampled at the same position on the document, and therefore one set is set. However, when the color information is extracted from the RGB pixels, there is a problem that position shift, color moire, bleeding, color shift and the like occur in an image having a high frequency or an edge portion.

Further, in Japanese Patent Laid-Open No. 3-129964, the pixel pitch of the CCD is set to be unequal.
Although it has an effect of reducing moire to a regular pattern, the line width is changed and the diagram becomes jagged when a thin line without a regular pattern or an edge part of an oblique line is read. There was a problem.

Further, in Japanese Unexamined Patent Publication No. 4-111676, there is a problem in that the input signal itself from the CCD is spatially filtered, so that moire due to aliasing distortion cannot be removed.

Further, a method of removing high frequency by using a spatial filter after sampling is generally used. In this method, when the CCD contains a frequency component higher than the Nyquist frequency, High-frequency components cannot be completely removed because it is not possible to distinguish between the spectrum of the image and the spectrum due to aliasing. Further, if the filter is applied too much, image quality deterioration such as resolution deterioration occurs.

Further, according to the above-mentioned conventional technique, even if the moire due to the aliasing distortion does not occur at the same magnification, the data is thinned out by the interpolation at the time of reduction, so that the sampling frequency is substantially reduced. However, there is a problem in that there is a case where the moiré is generated, which has not been generated at the same magnification.

Further, in order to reduce the moire, it is necessary to read the original at a higher density, but if the number of pixels in the main scanning direction of the CCD is simply increased, the area of the pixel itself becomes smaller and the light receiving area becomes smaller. Decrease, the amount of received light decreases and S
There was a problem that the / N ratio deteriorates.

The present invention has been made in view of the above, and it is an object of the present invention to easily remove moire due to aliasing distortion at low cost.

The present invention has been made in view of the above, and it is an object of the present invention to easily perform accurate sampling and reduce moire without requiring highly accurate adjustment.

The present invention has been made in view of the above, and the same position of RGB pixels is sampled to obtain a high frequency image or a position shift, color moire, bleeding, and color in an edge portion. The purpose is to reduce the occurrence of misalignment.

Further, the present invention has been made in view of the above, and it is an object of the present invention to obtain a good image free from moire even during zooming such as reduction / enlargement.

Further, the present invention has been made in view of the above, and it is an object of the present invention to perform high-density reading at a low cost and without deteriorating the S / N ratio.

[0024]

In order to achieve the above object, the present invention is directed to an image reading apparatus according to claim 1 in which an original is scanned while irradiating it with light, and the reflected light is transmitted to a solid-state imaging device. An image reading apparatus which forms an image and converts image information of a document into a one-dimensional electric signal and reads the solid-state imaging apparatus in which at least two line sensors having the same sampling pitch and different phases are integrally arranged. Be prepared.

The image reading apparatus according to the second aspect scans the original while irradiating the original with light, forms an image of the reflected light on the solid-state imaging device, and converts the image information of the original into a one-dimensional electric signal. In the image reading apparatus which is read in this way, a solid-state image pickup apparatus in which a pair of line sensors having the same sampling pitch and different phases are integrally arranged is provided.

Further, in the image reading apparatus according to the third aspect, the phase of the pair of line sensors is about 180 degrees,
The size and pitch of each pixel are substantially the same.

The image reading apparatus according to the fourth aspect scans the original while irradiating the original with light, forms an image of the reflected light on the solid-state imaging device, and converts the image information of the original into a one-dimensional electric signal. In an image reading apparatus that reads the same, a solid-state image pickup device integrally arranged with a pair of line sensors having the same sampling pitch and different phases of approximately 180 degrees, and a distance in the sub-scanning direction between the pair of line sensors. Only, a delay means for delaying one of the signals of the pair of line sensors, an interpolating means for interpolating the delayed outputs with each other, and a filtering means for filtering the image data interpolated by the interpolating means, cut-off frequency of the filtering means are variable by the data output ratio, if the sampling pitch of the magnification m, 1 line was P _0, the frequency f _{1 is, f 1 = m / (2} * P 0)
Is what is.

The image reading apparatus according to the fifth aspect scans the original while irradiating it with light, forms an image of the reflected light on the solid-state imaging device, and converts the image information of the original into a one-dimensional electric signal. In an image reading apparatus that reads the same, a solid-state image pickup device integrally arranged with a pair of line sensors having the same sampling pitch and different phases of approximately 180 degrees, and a distance in the sub-scanning direction between the pair of line sensors. Only, a delay means for delaying one signal of the pair of line sensors, an interpolating means for interpolating the delayed outputs with each other, a filtering means for filtering the image data interpolated by the interpolating means, and the filtering The image data thus obtained is interpolated according to the scaling factor of the output image, and resampling means for resampling is provided.

The image reading apparatus according to the sixth aspect scans the original while irradiating it with light, forms an image of the reflected light on the solid-state imaging device, and converts the image information of the original into a one-dimensional electric signal. In an image reading apparatus that reads the same, a solid-state image pickup device integrally arranged with a pair of line sensors having the same sampling pitch and different phases of approximately 180 degrees, and a distance in the sub-scanning direction between the pair of line sensors. A delay means for delaying one of the signals of the pair of line sensors, an interpolation means for interpolating the delayed outputs with each other, a filtering means for filtering the image data interpolated by the interpolation means, and An area determination unit that inputs the inserted image data and determines an area of a character portion, a halftone dot portion, a pattern portion, and the like is provided, and the filtering unit includes a determination result of the region determination unit. Based on, it is to change the cutoff frequency or frequency characteristics.

The image reading apparatus according to the seventh aspect scans the original while irradiating it with light, forms an image of the reflected light on the solid-state imaging device, and converts the image information of the original into a one-dimensional electric signal. In an image reading apparatus that reads the same, a solid-state image pickup apparatus in which a pair of line sensors having the same sampling pitch and different phases of approximately 180 degrees are integrally arranged, and a distance in the sub-scanning direction between the pair of line sensors. And a delay means for delaying an interval obtained by subtracting 1/2 of the reading pitch in the sub-scanning direction.

[0031]

The image reading apparatus of the present invention (Claim 1) uses the solid-state image pickup apparatus in which at least two line sensors having the same sampling pitch and different phases are integrally arranged, thereby reducing the cost. Moire due to folding and distortion can be easily removed. Furthermore, highly accurate adjustment is no longer necessary.

An image reading device of the present invention (claim 2)
Reduces the cost by using a solid-state imaging device in which a pair of line sensors with the same sampling pitch and different phases are integrally arranged, and easily removes moire due to aliasing distortion. Furthermore, highly accurate adjustment is no longer necessary.

An image reading device of the present invention (claim 3)
, The phase of the pair of line sensors is approximately 180 degrees, and the size and pitch of each pixel are substantially the same, so moire can be reduced without making the diagram jagged.

An image reading device of the present invention (claim 4)
Is a solid-state imaging device in which a pair of line sensors having the same sampling pitch and different phases of approximately 180 degrees are integrally arranged to read image information of an original document, and read the pair of line sensors in the sub-scanning direction. When one of the signals of the pair of line sensors is delayed by the interval, the delayed outputs are interpolated with each other, and variable depending on the data output magnification. When the magnification is m and the sampling pitch of one line is P ₀ , The frequency f ₁ is f ₁ = m / (2 * P ₀ ).
By filtering the interpolated image data at a cut-off frequency that gives a signal with no aliasing distortion, an image without moire even after thinning by interpolation is obtained.

An image reading device of the present invention (claim 5)
Is a solid-state imaging device in which a pair of line sensors having the same sampling pitch and different phases of approximately 180 degrees are integrally arranged to read image information of an original document, and read the pair of line sensors in the sub-scanning direction. One signal of the pair of line sensors is delayed by the interval, the delayed outputs are interpolated with each other, the interpolated image data is filtered, and the filtered image data is interpolated according to the scaling factor of the output image. Then, by re-sampling, a signal without aliasing distortion can be obtained, and an image without moire can be obtained even during zooming.

An image reading apparatus of the present invention (claim 6)
Is a solid-state imaging device in which a pair of line sensors having the same sampling pitch and different phases of approximately 180 degrees are integrally arranged to read image information of an original document, and read the pair of line sensors in the sub-scanning direction. Delay one signal of the pair of line sensors by the interval, interpolate the delayed outputs with each other, input the interpolated image data,
Area determination of a character portion, a halftone dot portion, a pattern portion, etc. is performed, and the cutoff frequency or frequency characteristic at the time of filtering is changed based on the determination result to filter the interpolated image data. As a result, the optimum image quality is obtained for each area such as the character portion, the halftone dot portion, and the pattern portion.

An image reading device of the present invention (claim 7)
Is a solid-state imaging device in which a pair of line sensors having the same sampling pitch and different phases of approximately 180 degrees are integrally arranged to read image information of an original document, and read the pair of line sensors in the sub-scanning direction. By delaying the interval obtained by subtracting 1/2 of the reading pitch in the sub-scanning direction from the interval, high-density reading can be performed with a low-cost configuration.

[0038]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a case where the image reading apparatus of the present invention is applied to a color scanner will be described in detail in the order of [First Embodiment] and [Second Embodiment] with reference to the drawings.

[Embodiment 1] FIG. 1 is a schematic block diagram of a color scanner of Embodiment 1, in which an analog output from each reading line of a CCD 101 which is a solid-state image pickup device is a digital signal by an A / D converter 102. Is converted to.

Here, the CCD 101, which is an essential part of the present invention,
As shown in FIG. 2, a pair of reading lines (R1 line and R2 line, G1 line and G2 line, B1 line and B2 line, respectively) form a pair L2 in the sub-scanning direction.
At the same pitch in the main scanning direction with a distance (interval) of
The pixel size is the same, and the pixel pitches are shifted by ½ of the pixel pitch P (180 degrees in phase), and are arranged integrally.

Further, each pair of reading lines is arranged at three distances L1 in the sub-scanning direction, and three R, G, and B color separation filters are provided to read colors. is there.

A delay memory for delaying each of the R1 line, G1 line, and B1 line signals of the digital signals output from the A / D converter 102 by a distance (L2) corresponding to the sub-scanning of a pair of reading lines. The data of the first line (R
The 1st line, G1 line, B1 line) and the data of the 2nd line (R2 line, G2 line, B2 line) form a read signal for the same line of the document.

Next, the G and B data are sent to the delay memory 104 for delaying the distance (L1 or 2 × L1) between the color pixels, and R, G, and
The image signal of B is obtained.

Subsequently, the respective data are stored in the correction circuit 1
By 05, shading correction, γ correction, etc. are performed. It should be noted that this correction circuit 105 is used in the delay memory 10
You may arrange | position before 3,104.

Next, the interpolation circuit 106 alternately interpolates the data of the first line and the data of the second line. At this point, the image data is virtually sampled at the double sampling frequency. In the first embodiment, a clock twice the CCD clock is used,
Interpolated data is obtained by alternately selecting the data.

The interpolation data output from the interpolation circuit 106 is sent to the area determination circuit 107 and the filter circuit 108. The area determination circuit 107 determines areas such as a character portion, a halftone dot portion, and a pattern portion, and the result is determined by the filter circuit 10.
8 and a color conversion / color correction / gradation correction circuit 110 described later.
Notify and change the parameters of each circuit.

In the filter circuit 108, the frequency characteristic is determined by the parameter for correcting the magnification and the MTF of RGB by the lens, and the interpolation data is filtered.

The resampling / interpolation circuit 109 interpolates according to the scaling factor and resamples it to output the RGB signal converted into a sampling frequency that can be finally output.

Based on this signal, the next color conversion / color correction / gradation correction circuit 110 sequentially obtains K, C, M, and Y signals, which are output to the printer side.

On the printer side, this signal is subjected to video processing / modulation processing by the video processing / modulation circuit 111, and the LD is driven via the LD driver 112.
Alternatively, if the filing process is performed without directly outputting to the printer, the RGB after resampling is used.
The signal may be temporarily stored in an image memory (not shown), directly or after being subjected to a compression process or a conversion process into a Lab space, and then filed in a predetermined storage device.

The operation of the above arrangement will be described by taking a specific manuscript as an example. First, the CCD pixel density is 400 dpi (original surface sampling pitch is 6
3.5μ, sampling frequency is about 15.7 lines / mm)
Then, the case where the frequency of the read document is 12 lines / mm (about 1.5 times the Nyquist frequency) and the scaling factor is 100% will be described.

As shown in FIG. 3A, it is assumed that the pixels of the CCD 101 are arranged at a pitch of 63.5μ on the original surface. In FIG. 6A, the pixel lines on the first line and the pixel line on the second line are written separately, but since the signals are delayed by the delay memory 103 as described above, the same pixel in the sub-scanning direction is used here. The signal of the line is obtained.

With this arrangement of CCD 101, 12 lines / mm
When reading the original document (line chart), the CCD 10
The area of the portion where the opening 1 and the line chart overlap (the portion where the diagonal lines overlap in the figure) is obtained as the signal output. FIG. 3B shows the output of each reading line, which is a moire having a phase inversion of about 4 lines / mm.

Explaining this in the frequency domain, FIG.
As shown in (a), a spectrum due to aliasing appears at about 4 lines / mm. In the figure, f
_n is a Nyquist frequency of about 8 lines / mm, and f _s is a sampling frequency of about 16 lines / mm.

Next, when the data of the first line and the data of the second line are alternately interpolated by the interpolation circuit 106, as shown in FIG.
It becomes as shown by the solid line in (c). Since the sampling frequency is virtually doubled at this time, the overlapping due to the aliasing distortion is eliminated and the moire does not occur, as shown in FIG. 4B. In the figure, f _n 'is a virtual Nyquist frequency after interpolation of about 16 lines / mm, and f _s is a virtual sampling frequency after interpolation of about 32 lines / mm.

Next, if the scaling factor is 100%, then FIG.
When filtered with a filtering characteristic having a cutoff frequency in the vicinity of the Nyquist frequency (about 8 lines / mm) of one row of read lines as shown in (b), data without moiré as shown in FIG. 3 (c) Is obtained. In the frequency domain, as shown in FIG. 4C, only the DC component remains and the average density of the original is obtained. This filtering is performed by a digital filter, and the convolution method or the like is applied.

Next, in the resampling / interpolation circuit 109, if the scaling factor is 100%, resampling is performed at the original sampling pitch of 63.5 μ. In reality, it will be thinned to 1/2. Data after resampling is shown in FIG. 3C, and the spectrum state at that time is shown in FIG. 4D. Nyquist frequency f _n = 8 /
Within mm, the aliasing spectrum is removed, image data without moire can be obtained, and a signal with less moire can be obtained even for a document having a Nyquist frequency or higher.

Next, the frequency of the read document is 6 lines /
mm (0.75 times the Nyquist frequency) will be described in consideration of the influence of the frequency. As shown in FIG. 5A, data is sampled when the sampling is performed and the number of sampling pixels is 100.
As shown in FIG. 6A, the spectrum state at that time is as shown in FIGS. 6B and 7B.

In FIGS. 6 (b) and 7 (b), the high frequency component of the folded wave enters into the Nyquist frequency (f _n = 8 lines / mm) at a position of about 4 Hz, and the frequency of the original document is 6 Hz. Interference causes 6-4 = about 4 Hz moire.

The high frequency spectrum shows the MT of the CCD aperture.
Although it is actually small due to F (see FIG. 11) and Lence's MTF, etc., when it folds back within the Nyquist frequency range and interferes with the frequency of the original, moire is caused, and especially in a color machine, it becomes a color moire. Deteriorates.

In the first embodiment, since there is overlap with the adjacent pixels, the pixel pitch P is set to P / 2 and the number of pixels is simply doubled as shown in FIGS. 12 (a) and 12 (b). MTF in the high frequency region as shown in FIG.
Since the TF is deteriorated, the influence of high frequency aliasing distortion is reduced.

Further, if the area determination processing (separation processing of the character portion, the halftone dot portion, the pattern portion, etc.) is performed in the state where the above-mentioned moire occurs, there is a problem that a separation defect is likely to occur. In the first embodiment, since the area determination processing is performed with the data interpolated, the separation accuracy can be improved.

FIG. 8A shows the data after the interpolation of the first line and the second line, and FIG. 8B shows the spectrum state at that time. As shown in the figure, the high frequency components are also separated from each other, and there is no moire. Next, in the case of the same magnification, when the filtering process as shown in FIG. 8B is performed, the spectrum does not include the high frequency component as shown in FIG. 9A. Then, when thinning processing is performed to restore the original sampling frequency, there is no high-frequency aliasing distortion within the Nyquist frequency (8 lines / mm) as shown in FIG. can get.

Here, as another example of the magnification, in the case of 50%, the filter characteristic (cutoff frequency of 0.5 /
It is filtered by (2 × 0.0635) ≈4 lines / mm), and the spectrum of only the DC component is obtained as shown in FIG. Next, the data is decimated to 1/4,
The spectrum is as shown in FIG. 10C, and moire does not occur even during zooming.

FIG. 10D shows an example of the spectrum when the scaling process is performed without performing the filtering process. Plural folding distortions occur within the Nyquist frequency (f _nR ) at the time of zooming.

The filter circuit 108 is independently arranged for each RCB as shown in FIG.
It is also used to correct variations in GB MTF.
The MTF of the lens is RGB as shown in FIG.
There are variations, and differences also occur depending on the image height.

Therefore, in the first embodiment, the MTF is measured for each image height at the time of assembling and adjusting the lens, or a plurality of MTF measurement patterns (see FIG. 14) are provided in the area of the original surface before the original reading or the start. It is assumed to have been placed.

As shown in FIG. 14, the MTF measurement pattern is a white and black edge pattern having a high contrast and is regularly read before reading the original. Next, the RGB edge signal is differentiated for each image height by a differentiating circuit, and Fourier-transformed by the FFT circuit to obtain the MTF of each color. Next, the filter characteristics of RGB are obtained by taking the ratio for each frequency with the MTFs of the other two colors with the MTF of an arbitrary color as a reference.

For example, based on the MTF of G, the MTF characteristic of R = (MTF characteristic of G) / (MTF characteristic of R) MTF characteristic of B = (MTF characteristic of G) / (MTF characteristic of B) Becomes

Next, inverse Fourier transform is performed by the inverse FFT circuit to obtain the filter coefficient for each color. By reflecting this coefficient in the filter circuit 108, FIG.
The filter characteristics shown in FIG. 13B are obtained, and the MTF variation of each color of the lens after the filtering is shown in FIG.
The correction is performed as in (c), color moire and bleeding are reduced, and the accuracy of color conversion in the color conversion / color correction / gradation correction circuit 110 is improved. FIG. 15 shows the RGB MT
The schematic block diagram of the variation correction of F is shown.

Further, the filter circuit 108 may perform a two-dimensional filtering process using a delay memory (not shown). Based on the determination result of the area determination circuit 107, the filter circuit 108 described above
Optimum images can be obtained in each area by making it possible to change the parameters appropriately.

In the first embodiment, the CCD shown in FIG.
Although the configuration of the (solid-state image pickup device) is adopted, as shown in FIG. 16, a block having a phase shift of 180 ° may be arranged as another block. In the case of a monochrome CCD, one set of reading line may have only one reading line.

[Embodiment 2] FIG. 17 is a schematic block diagram of a color scanner according to a second embodiment. The interpolation circuit 170 in the schematic block diagram of the first embodiment shown in FIG. 1 is replaced with an interpolation filter 1701. It is a thing. The other configurations are common to those of the first embodiment, and illustration and description thereof are omitted.

The analog output from each reading line of the CCD 101 is converted into a digital signal by the A / D converter 102. Next, A / D converter 10
Of the digital signals output from 2, R1 line, G
The signals of line 1 and line B1 are sent to the delay memory 10
3 by the distance (L2:
That is, by being delayed by 1/2) of the reading pitch in the sub-scanning direction, a pair of reading lines overlap each other's intermediate points in both the main and sub-scanning directions, as shown in FIG. However, the original will be sampled. The sampling points at this time are shown in FIG.

The two-dimensional Nyquist frequency range at this time is shown in FIG. In the figure, the alternate long and short dash line is the Nyquist frequency range of the conventional CCD (CCD of FIG. 12A), and the broken line is the Nyquist frequency range of the second embodiment, which means that a frequency band twice the area can be obtained. Becomes
The generation of moire can be reduced and high-density sampling can be performed. 20A and 20B are explanatory diagrams showing sampling states and sampling points of a conventional CCD.

Next, the respective data are stored in the correction circuit 10
Shading correction, γ correction, etc. are performed according to 5.
The data obtained by the delay memory for several lines and the two-dimensional interpolation filter 1701 interpolate the intermediate points as shown in FIG. 21, and the data of double the sampling frequency is obtained for both the main and the sub. One of the data is sent to the area / area determination circuit 107, and the area determination of the character portion, the halftone dot portion, the pattern portion, etc. is performed. With this result,
The filter circuit 108 and color conversion / color correction
It is possible to change the parameters of the gradation correction circuit 110.

In the filter circuit 108, the frequency characteristic is determined by the parameter for correcting the scaling factor and the MTF of RGB by the lens, and the data of the double sampling frequency is filtered.

The resampling / interpolation circuit 109 interpolates according to the scaling factor and resamples it to output the RGB signal converted into a sampling frequency that can be finally output.

Based on this signal, the next color conversion / color correction / gradation correction circuit 110 sequentially obtains K, C, M, and Y signals and outputs them to the printer.

On the printer side, this signal is subjected to video processing / modulation processing by the video processing / modulation circuit 111, and the LD is driven via the LD driver 112.

[0081]

As described above, the image reading apparatus of the present invention (Claim 1) scans a document while irradiating it with light, and forms an image of the reflected light on the solid-state image pickup device. An image reading apparatus for reading information by converting information into a one-dimensional electric signal is provided with a solid-state imaging apparatus in which at least two line sensors having the same sampling pitch and different phases are integrally arranged. Can be easily removed at low cost. Also,
Moire can be reduced by performing accurate sampling easily without requiring highly accurate adjustment.

An image reading device of the present invention (claim 2)
Scans the original while irradiating it with light, forms an image of the reflected light on the solid-state imaging device, converts the image information of the original into a one-dimensional electrical signal, and reads the image at the same sampling pitch. Moreover, since the solid-state imaging device in which the pair of line sensors having different phases are integrally arranged is provided, it is possible to easily remove the moire due to the folding distortion at low cost. Further, it is possible to easily perform accurate sampling and reduce moire without requiring highly accurate adjustment.

An image reading apparatus of the present invention (claim 3)
Since the pair of line sensors have a phase of about 180 degrees and the pixels have substantially the same size and pitch, moire due to aliasing distortion can be easily removed at low cost. Further, it is possible to easily perform accurate sampling and reduce moire without requiring highly accurate adjustment. Furthermore, moire is reduced without making the diagram jagged.

An image reading device of the present invention (claim 4)
Is a solid-state imaging device in which a pair of line sensors having the same sampling pitch and different phases of approximately 180 degrees are integrally arranged to read image information of an original document, and read the pair of line sensors in the sub-scanning direction. When one of the signals of the pair of line sensors is delayed by the interval, the delayed outputs are interpolated with each other, and variable depending on the data output magnification. When the magnification is m and the sampling pitch of one line is P ₀ , The frequency f ₁ is f ₁ = m / (2 * P ₀ ).
Since the interpolated image data is filtered at a cutoff frequency that becomes, moire due to aliasing distortion can be easily removed at low cost. Further, it is possible to easily perform accurate sampling and reduce moire without requiring highly accurate adjustment. In addition, a signal with no aliasing distortion can be obtained, and an image without moire can be obtained even after thinning by interpolation.

An image reading apparatus of the present invention (claim 5)
Is a solid-state imaging device in which a pair of line sensors having the same sampling pitch and different phases of approximately 180 degrees are integrally arranged to read image information of an original document, and read the pair of line sensors in the sub-scanning direction. One signal of the pair of line sensors is delayed by the interval, the delayed outputs are interpolated with each other, the interpolated image data is filtered, and the filtered image data is interpolated according to the scaling factor of the output image. However, since resampling is performed, moire due to aliasing distortion can be easily removed at low cost. Also, without the need for highly precise adjustment,
Accurate sampling can be easily performed to reduce moire. In addition, a signal with no aliasing distortion can be obtained, and an image without moire can be obtained even during zooming.

An image reading apparatus of the present invention (claim 6)
Is a solid-state imaging device in which a pair of line sensors having the same sampling pitch and different phases of approximately 180 degrees are integrally arranged to read image information of an original document, and read the pair of line sensors in the sub-scanning direction. Delay one signal of the pair of line sensors by the interval, interpolate the delayed outputs with each other, input the interpolated image data,
Areas such as character areas, halftone areas, and pattern areas are determined, and the cutoff frequency or frequency characteristic at the time of filtering is changed based on the result of the determination to filter the interpolated image data. Moire caused by can be easily removed at low cost. Further, it is possible to easily perform accurate sampling and reduce moire without requiring highly accurate adjustment. Also,
Optimal image quality is obtained for each area such as the character portion, the halftone dot portion, and the pattern portion.

An image reading device of the present invention (claim 7)
Is a solid-state imaging device in which a pair of line sensors having the same sampling pitch and different phases of approximately 180 degrees are integrally arranged to read image information of an original document, and read the pair of line sensors in the sub-scanning direction. Since the interval obtained by subtracting 1/2 of the reading pitch in the sub-scanning direction from the interval is delayed, moire due to aliasing distortion can be easily removed at low cost. Further, it is possible to easily perform accurate sampling and reduce moire without requiring highly accurate adjustment. In addition, the low-cost configuration enables high-density reading.

[Brief description of drawings]

FIG. 1 is a schematic block diagram of a color scanner according to a first embodiment.

FIG. 2 is a configuration diagram of a CCD (solid-state imaging device) according to the first embodiment.

FIG. 3 is an explanatory diagram showing a specific operation example of the first embodiment.

FIG. 4 is an explanatory diagram showing a specific operation example of the first embodiment.

FIG. 5 is an explanatory diagram showing a specific operation example of the first embodiment.

FIG. 6 is an explanatory diagram showing a specific operation example of the first embodiment.

FIG. 7 is an explanatory diagram showing a specific operation example of the first embodiment.

FIG. 8 is an explanatory diagram showing a specific operation example of the first embodiment.

FIG. 9 is an explanatory diagram showing a specific operation example of the first embodiment.

FIG. 10 is an explanatory diagram showing a specific operation example of the first embodiment.

FIG. 11 is an explanatory diagram showing a conventional CCD array and an example in which the number of pixels is simply doubled.

FIG. 12 is an explanatory diagram showing MTF characteristics according to the present invention.

FIG. 13 is an explanatory diagram showing variations in RGB MTF.

FIG. 14 is an explanatory diagram showing an MTF measurement pattern.

FIG. 15 is a schematic block diagram of RGB MTF variation correction.

FIG. 16 is a configuration diagram of another CCD (solid-state imaging device).

FIG. 17 is a schematic block diagram of a color scanner according to a second embodiment.

FIG. 18 is an explanatory diagram showing sampling states and sampling points of the CCD according to the second embodiment.

FIG. 19 is an explanatory diagram showing a two-dimensional Nyquist frequency range in the second embodiment.

FIG. 20 is an explanatory diagram showing sampling states and sampling points of a conventional CCD.

FIG. 21 is an explanatory diagram showing data of twice the sampling frequency for both the main and sub sides obtained according to the second embodiment.

[Explanation of symbols]

 101 CCD (solid-state image pickup device) 102 A / D converter 103, 104 delay memory 105 correction circuit 106 interpolation circuit 107 area determination circuit 107 108 filter circuit 109 resampling / interpolation circuit 110 color conversion / color correction / gradation correction circuit 1701 Interpolation filter

Claims (7)

[Claims] 1. An image reading apparatus that scans an original while irradiating it with light, forms an image of the reflected light on a solid-state imaging device, converts image information of the original into a one-dimensional electric signal, and reads the same. An image reading device comprising a solid-state imaging device in which at least two line sensors having a sampling pitch and different phases are integrally arranged. 2. An image reading apparatus which scans an original while irradiating it with light, forms an image of the reflected light on a solid-state imaging device, converts image information of the original into a one-dimensional electric signal and reads the same. An image reading device comprising a solid-state imaging device in which a pair of line sensors having a sampling pitch and different phases are integrally arranged. 3. The phase of the pair of line sensors is approximately 180 degrees, and the size and pitch of each pixel are
The image reading apparatus according to claim 2, wherein the image reading apparatuses are substantially the same. 4. An image reading apparatus that scans an original while irradiating it with light, forms an image of the reflected light on a solid-state imaging device, converts image information of the original into a one-dimensional electric signal, and reads the same. A solid-state imaging device having a pair of line sensors integrally arranged at a sampling pitch and having a phase difference of about 180 degrees, and one signal of the pair of line sensors by the distance in the sub-scanning direction between the pair of line sensors. Delaying means for delaying, the interpolating means for interpolating the output after the delay, and the filtering means for filtering the image data interpolated by the interpolating means, and the cutoff frequency of the filtering means is It is variable depending on the output magnification, and when the magnification is m and the sampling pitch of one line is P ₀ , the frequency f ₁ is f ₁ = m / (2 * P ₀ ). An image reading device characterized by the above. 5. An image reading apparatus which scans an original while irradiating it with light, forms an image of the reflected light on a solid-state image pickup device, converts image information of the original into a one-dimensional electric signal and reads the same. A solid-state imaging device in which a pair of line sensors having a sampling pitch and a phase difference of approximately 180 degrees are integrally arranged, and one signal of one of the pair of line sensors by the distance in the sub-scanning direction between the pair of line sensors. Delaying means for delaying, the interpolating means for interpolating the delayed output with each other, the filtering means for filtering the image data interpolated by the interpolating means, and the filtered image data for transforming the output image. An image reading apparatus comprising: a resampling means for interpolating and resampling according to a magnification. 6. An image reading apparatus that scans an original while irradiating it with light, forms an image of the reflected light on a solid-state imaging device, converts image information of the original into a one-dimensional electric signal, and reads the same. A solid-state imaging device in which a pair of line sensors having a sampling pitch and a phase difference of approximately 180 degrees are integrally arranged, and one signal of one of the pair of line sensors by the distance in the sub-scanning direction between the pair of line sensors. Delaying means for delaying, interpolating means for interpolating the output after the delay, filtering means for filtering the image data interpolated by the interpolating means, and inputting the interpolated image data, Area filtering means for judging the area of the character portion, the halftone dot portion, the pattern portion, etc. is provided, and the filtering means determines the cutoff frequency to the cutoff frequency based on the determination result of the area determining means. An image reading device characterized by changing frequency characteristics. 7. An image reading apparatus that scans an original while irradiating it with light, forms an image of the reflected light on a solid-state imaging device, converts image information of the original into a one-dimensional electric signal, and reads the same. A solid-state imaging device in which a pair of line sensors having a sampling pitch and a phase difference of approximately 180 degrees are integrally arranged, and a distance between the pair of line sensors in the sub-scanning direction to a half of the reading pitch in the sub-scanning direction. An image reading apparatus, comprising: a delay unit that delays an interval obtained by subtracting.