JPH05122542A - Image processor - Google Patents

Abstract

PURPOSE:To make correction with fidelity possible by composing the processor of a source image illuminating means, photoelectric converting means to form image at a prescribed position with reflected light from this illuminating means and to output signals in plural colors and means to correct the position deviation of the plural chrominance signals based on a correcting coefficient calculated beforehand for each position. CONSTITUTION:This device is composed of an inline system image reading sensor 1 to read an original and to decompose it into R, G and B color signals, analog amplifier 2, A/D conversion circuit 3, shading correction circuit 4 to correct the characteristics of an original illumination lamp and the sensor, deviation correction part 5 to correct an influence caused by the R, G and B arrangement of the sensor 1, RAM 6 to store the correcting coefficient, and masking processing part 7 to correct the characteristic of the reading system and to standardize the respective R, G and B digital signals. Thus, a source image scanned by the sensor is replaced with R, G and B image signals, shading correction is executed after converting those signals to the digital signals at the converter and next, the masking processing is executed.

Description

Detailed Description of the Invention

[0001]

The present invention relates to a copying machine, a facsimile,
The present invention relates to an image processing apparatus suitable for optically reading an image from an image scanner or the like and digitally processing the read image information.

[0002]

2. Description of the Related Art Conventionally, in this type of apparatus, a color image is read using a sensor (CCD or the like) in which R, G and B optical filters are arranged in a predetermined order as shown in FIG. ..

Further, as an image reading characteristic using a sensor, a level difference (so-called color difference) between each RGB output of the sensor is caused by a physical deviation between an arrangement position of each RGB filter and a pixel position on the original image. It is generally well known that a shift phenomenon) occurs.

Therefore, in the conventional device of this type, it is an essential technique to correct the level difference between the RGB outputs of the sensor, that is, the color misregistration correction. As a means for performing this color misregistration correction, as shown in FIGS. 11 and 12, a method of using read image data of two adjacent pixels (see FIG. 2), and a pixel of interest and an image of two pixels adjacent thereto The method using data is general.

[0005]

However, in the above-mentioned conventional example, the color misregistration caused by the positional misalignment based on the arrangement of the R, G, and B filters of the sensor is merely corrected.

That is, when a sensor having this type of filter is used in an actual device, there are the following two other factors that cause color misregistration, and after considering these influences as well. It is necessary to correct the color shift.

A level difference between RGB outputs of the sensor due to chromatic aberration of an optical member (lens, reflection mirror, etc.) installed between the original image and the sensor.

The level difference between the RGB outputs of the sensor due to the change in the optical path length between the original image and the sensor due to the deflection of the platen glass or the inclination of the optical member.

Therefore, it is an object of the present invention to provide an image processing apparatus capable of correcting color misregistration in a form including the above-mentioned factors which cannot be corrected by the prior art.

[0010]

In order to solve the above-mentioned problems, an image processing apparatus according to the present invention comprises an illuminating means for illuminating an original image and a plurality of light beams reflected from the original image formed at a predetermined position. It is provided with a photoelectric conversion means for outputting color signals and a correction means for correcting the positional deviation of the signals of the plurality of colors based on a correction coefficient calculated in advance for each pixel position.

Here, the correction coefficient is determined from one of the three output data of RGB based on the output data of two or more adjacent pixels of the sensor, or is individually determined for each output data of RGB. Is preferred. Further, as the output data for determining the correction coefficient, it is preferable to use data obtained by reading a black fine line pattern formed on the reference white surface.

[0012]

By adopting the configuration of the present invention described above, the positional deviation of the RGB output of the photoelectric conversion means (sensor) can be faithfully corrected for each pixel position.

[0013]

EXAMPLES Examples of the present invention will be described in detail below.

Embodiment 1 FIG. 1 is a block diagram showing the overall configuration of an embodiment of the present invention. In FIG. 1, reference numeral 1 denotes an original document, R, G,
It is an in-line type image reading sensor that is separated into B color signals and is composed of a CCD.

Reference numeral 2 is an analog amplifier, 3 is an A / D conversion circuit for converting an analog signal into a digital signal, 4 is a shading correction circuit for correcting characteristics of a reading system, for example, characteristics of a document illumination lamp or a sensor. Is a deviation correction unit for correcting the influence of the arrangement of R, G, B of the sensor 1.

Reference numeral 6 is a random access memory for storing a correction coefficient used in the shift correction calculation.
Reference numeral 7 denotes a masking processing unit for correcting the characteristics of the reading system and standardizing each RGB digital signal.

In the above structure, the sensor 1 scans the document to convert the document information into R, G, and B image signals. The image signal converted into R, G, B is amplified to a target level by the amplifier 2 and converted from an analog signal to a digital signal (in this embodiment, 8-bit multivalued data) by the A / D converter 3. To be converted. The image signal converted into the digital signal is processed by the shading correction circuit 4
Shading correction is performed on the screen, and corrections related to characteristics such as reading and conversion are performed.
H, 0 OH for the darkest image.

The image signal subjected to the shading correction is
The deviation correction unit 5 corrects the deviation, and RGBR of the sensor 1 is corrected.
The corrections described in detail below are performed so as to eliminate the color shift caused by the physical shift due to the arrangement of GB ... RGB.

The image signal subjected to the above-mentioned deviation correction is subjected to masking processing in the masking processing unit 7, and RGB (each 8
(Bit) standardized signal.

Next, one of the characteristic portions of the embodiment of the present invention will be described in detail with reference to the drawings.

In the present embodiment, the deviation correction unit 5 and the memory 6 have a circuit configuration as shown in FIGS.

FIG. 3 shows a circuit for storing the correction coefficient used in the correction calculation of the deviation correction unit 5 in the memory 6 and a circuit for reading the correction coefficient. In the figure, 30 is image data (R or B) output from the shading circuit 4.
Is a D flip-flop for delaying for 1 pixel period, and the data of the Nth pixel and the image data of the (N + 1) th pixel are simultaneously input to the address of the ROM 31.

An appropriate correction coefficient value obtained from the two image data is stored in the ROM 31, and the correction coefficient corresponding to the address input is output. An address counter 37 outputs the pixel address in the sensor 1, and the output pixel address becomes an address input of the memory 6. Reference numeral 39 denotes the sum (a +) of the first correction coefficient value (b) output from the memory 6 and the preset correction coefficient value.
It is a subtracter that obtains a difference from b) and outputs a second correction coefficient value (a).

FIG. 5 shows a circuit for correcting the deviation of each image data of RGB based on the correction coefficient output from the memory 6 and the subtractor 39. In the figure, 10,1
Reference numerals 1 and 12 denote D-type flip-flops (F / F) for delaying each image data of RGB for one pixel period. This D
The R data of the Nth pixel output from the mold F / F10 is
In the multiplier 13, the second correction coefficient (a) output from the subtractor 39 is multiplied. On the other hand, the multiplier 14 multiplies the R data of the (N + 1) th pixel by the first correction coefficient (b) output from the memory 6, and after the respective multiplication results are added by the adder 18. , Divider 2
1 is divided by the sum of two correction coefficients and output as 8-bit image data.

Similarly, the B data of the Nth pixel output from the D-type F / F 12 is multiplied by the first correction coefficient (b) output from the memory 6 in the multiplier 15.
On the other hand, the multiplier 16 multiplies the R data of the (N + 1) th pixel by the second correction coefficient (a) output from the subtractor 39, and the respective multiplication results are added by the adder 20. After that, it is divided by the sum of the two correction coefficients in the divider 23 and output as 8-bit image data.

The Nth output from the D-type F / F11
The G data of the pixel and the G data of the (N + 1) th pixel are
After being added by the adder 19, the divider 22 halves the result and outputs it as 8-bit image data.

Next, the control of the memory 6 will be specifically described with reference to FIG.

(Difference Correction / Memory Read) In the case of performing the deviation correction on the image data read by the sensor 1, the RW By setting the SEL signal to 1 (Hi level), the OR gate 36 is closed, the memory 6 is fixed in the read mode, and the tri-state buffer 32 is disabled.

At this time, the scan start signal (S
When TART *) becomes 0 (Low), the address counter 37 is reset at the same time, and the address data synchronized with the pixel position of the image signal output from the sensor 1 is stored in the memory 6.
The deviation correction is performed by the second correction coefficient (b) input to the memory 6 and output from the memory 6 and the first correction coefficient (a) output from the subtractor 39.

The image data input from the shading correction circuit 4 to the shift correction circuit 5 and the address counter 37 are designed to operate in synchronization with the pixel clock (CLK). The data is read and the deviation is corrected.

(When storing correction coefficient / memory write) When storing correction data in the memory 6, the RW SEL
By setting the signal to 0 (Low level), the OR gate 36 is opened, and data is written to the memory 6 in synchronization with the write signal WEN. Further, the tri-state buffer 32 is enabled, and the correction coefficient data output from the ROM 31 is stored in the memory 6. At this time, the scan start signal (START *) of the sensor 1 is 0 (Lo
At the same time as w), the address counter 37 is reset and the address data synchronized with the pixel position of the image signal output from the sensor 1 is input to the memory 6.

On the other hand, the sensor 1 reads the black thin line pattern shown in FIG. 2 under the control of a control circuit (not shown). The relative position of the sensor 1 and the black thin line is controlled so as to have the positional relationship shown in FIG. 2 with respect to the pixel address output from the address counter 37.

In the present embodiment, the deviation correction processing using the data of the two adjacent pixels of the sensor has been described, but it goes without saying that the deviation correction using the pixel data of three adjacent pixels or more can also be executed. is there.

In this embodiment, the correction coefficient stored in the memory may be any correction coefficient, or both correction data may be stored in the memory.

Further, the output of the sensor is not limited to RGB, but any output in which two or more kinds of filters are periodically arranged can be easily applied.

As described above, the correction of the RGB output of the sensor can be corrected by the correction coefficient calculated from the data obtained by reading the black thin line pattern, which cannot be realized by the conventional apparatus. The optical path length between the original image and the sensor may change due to the deviation caused by the chromatic aberration of the optical member (lens, reflection mirror, etc.) installed between the original image and the sensor, the deflection of the platen glass, and the inclination of the optical member. R of the sensor considering the deviation caused by
It is possible to faithfully correct the positional deviation of the GB output for each pixel position, and it is possible to obtain clear and high-quality read image data without deterioration of resolution due to color misregistration or partial color fringing. An image processing device can be provided.

Second Embodiment Next, a second embodiment of the present invention is shown in FIGS.

The difference between this embodiment and the first embodiment is that the correction coefficients in this embodiment are stored in the memories 6-1 and 6-2 independently of the R signal and the B signal. That is, as shown in FIG. 8, deviation correction can be independently performed on each of the RGB signals input.

In the case of this embodiment, as in the first embodiment,
The original image and the sensor, which could not be realized by the conventional device, due to the chromatic aberration of the optical member (lens, reflection mirror, etc.) installed between the original image and the sensor, the bending of the platen glass, and the inclination of the optical member. R, G, and B of the sensor in consideration of the shift caused by the change in the optical path length between
The positional deviation of the output can be faithfully corrected for each pixel position, and the color deviation correction including the transmission frequency characteristic of the RGB filter and the sensitivity characteristic of the sensor which are slightly different for each pixel can be performed. it can.

Therefore, it is possible to provide an image processing apparatus with higher accuracy than that of the first embodiment.

Third Embodiment Next, a third embodiment of the present invention will be described in detail with reference to FIG.

In FIG. 9, reference numeral 42 is a latch circuit for taking in arbitrary pixel data in the sensor 1 by a data latch signal output from the CPU 40, and the latched pixel data is input to the CPU 40. CPU40
Then, the deviation correction amount for the sequentially input pixel data is calculated.

(When storing correction coefficient / memory write) In this embodiment, the correction data is stored in the memory 6 via the data bus of the CPU. The CPU 40 outputs a chip select signal (CS *) to the memory 6. As a result, the OR gate 36 is opened and at the same time, the address bus of the CPU 40 is selected by the selector 41. Also, the tri-state buffer 32 is enabled and the CPU 40
The correction coefficient data output by the CPU 40 is sequentially stored in a predetermined address of the memory 6 in synchronization with the write signal WEN output by the CPU 40.

(Difference Correction / Memory Read) When performing the deviation correction on the image data read by the sensor 1, the CPU 40 does not output the chip select signal (CS *) to the memory 6, and the OR The gate 36 is closed, the memory 6 is fixed in the read mode, and the tri-state buffer 32 is disabled. Further, at this time, the output of the address counter 37 is selected by the selector 41, the correction coefficients are sequentially read from the memory 6 as in the first embodiment, and the deviation correction is performed.

In this embodiment, the correction coefficient calculation process is performed by CP.
Since it can be performed by U, the hardware circuit can be easily configured. Further, by performing a highly accurate calculation by the CPU, it is possible to easily realize the deviation correction using a more faithful correction coefficient.

[0046]

As described above, according to the present invention, a shift due to chromatic aberration of an optical member (lens, reflecting mirror, etc.) installed between an original image and a sensor, which cannot be realized by a conventional apparatus, or an original document Correctly corrects the positional deviation of the RGB output of the sensor for each pixel position in consideration of the deviation caused by the change in the optical path length between the original image and the sensor due to the deflection of the base glass and the inclination of the optical member. can do.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of an image processing apparatus embodying the present invention.

FIG. 2 is a diagram showing a relationship between a black thin line pattern and output data.

FIG. 3 is a diagram showing a correction coefficient memory circuit.

FIG. 4 is a diagram showing read and write timing of a correction coefficient memory.

FIG. 5 is a diagram showing a color misregistration correction circuit.

FIG. 6 is a diagram showing another embodiment of a correction coefficient memory circuit.

FIG. 7 is a diagram showing another embodiment of a correction coefficient memory circuit.

FIG. 8 is a diagram showing another embodiment of the color misregistration correction circuit.

FIG. 9 is a diagram showing another embodiment of the correction coefficient memory circuit.

FIG. 10 is a diagram showing a conventional method of correcting color misregistration.

FIG. 11 is a diagram illustrating a color misregistration correction method of a conventional example.

FIG. 12 is a diagram showing a conventional method for correcting color misregistration.

[Explanation of symbols]

 1 optical sensor 3 A / D conversion circuit 4 shading correction circuit 5 misalignment correction unit 6 memory 7 masking processing unit

Claims (5)

[Claims] 1. An illumination unit for illuminating an original image, a photoelectric conversion unit for forming reflected light from the original image at a predetermined position and outputting signals of a plurality of colors, and a pre-calculated pixel position. An image processing apparatus comprising: a correction unit that corrects the positional deviation of the signals of the plurality of colors based on a correction coefficient. 2. The image processing apparatus according to claim 1, wherein the correction coefficient is determined based on output data of two or more adjacent pixels in the photoelectric conversion unit. 3. The correction coefficient according to claim 1, wherein the correction coefficient is RG.
An image processing device characterized in that it is determined from one of the three output signals of B. 4. The correction coefficient according to claim 1, wherein the correction coefficient is RG.
An image processing device, wherein each output signal of B is individually determined. 5. The image processing apparatus according to claim 1, wherein the output data for determining the correction coefficient is data obtained by reading a black fine line pattern formed on a reference white surface.