JPH02260886A - Color picture reader - Google Patents

Abstract

PURPOSE:To simplify an entire constitution and to obtain high picture quality by decomposing a read signal into a luminance signal and a color difference signal and applying optimum processing to them respectively. CONSTITUTION:Read signals VR, VG, VB from an image sensor circuit 8 are inputted to a matrix circuit 9 and converted into a luminance signal Y and color difference signals BY, GY. The luminance signal Y and color difference signals BY, GY are subject to gain adjustment processing by a processing circuit 10 respectively, A/D-converted by an A/D converter 11 into a digital picture signal. In this case, the frequency characteristic is important for the luminance signal Y, then the processing system for the luminance signal Y requires circuit components whose frequency characteristic covers a broad band, but since the luminance signal Y is a mean value of the signals VR, VG, VB, the S/N is considerably improved more than the processing for each color. Since the color difference signals GY, BY do not much require the frequency characteristic in comparison with that of the luminance signal Y, inexpensive circuit components of the processing system are employed. Thus, the entire constitution is simplified and high picture quality is attained.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a color image reading device that reads a color image using an image sensor for each color.

[Conventional technology]

Among the above-mentioned color image reading apparatuses, one having independent and equivalent processing circuits for three colors is conventionally known.

FIG. 5 shows such a conventional color image reading device, in which signals R, G, and B from a reading image sensor (for example, CCD) are processed by an R processing section 1. G processing section 2. The B processing section 3 performs the same processing. This will be explained using R as an example. The CCDCD output signal is subjected to preprocessing such as DC reproduction and sample hold in the preprocessing section 4, and then input to the variable gain amplifier 5. In the variable gain amplifier 5, in order to correct the sensitivity of each sensor, the gain is automatically corrected and set according to the background reflectance of the document or the reference reflectance. The output of variable gain amplifier 5 is given to input terminal IN of A/D converter 6. The reference value RF of the A/D converter 6 is
Usually, it is obtained by sampling the output of the variable gain amplifier 5 at a predetermined timing. The A/D output is then output as a digital image signal to the next processing stage via the interface 7. In addition, the G processing section 2゜B
The processing section 3 functions similarly.

By the way, in a nuclide color image reading processing system, when processing image signals, limiting the frequency band causes a decrease in resolution, so a wideband analog processing system is required.

[Problem to be solved by the invention]

In the above-mentioned conventional technology, three colors are processed simultaneously, so relatively high-speed reading is possible, but it is extremely expensive as a whole because it requires three broadband analog processing systems using expensive elements. , and the configuration is complicated.

Furthermore, in order to improve the S/N ratio, it is desirable to use a low-pass filter to remove noise from the digital signal, but as mentioned above, there is a problem in that a sufficient effect cannot be achieved in order to secure a wide band.

Note that one processing circuit processes each color sequentially, specifically R, G, and B for each scanning line.

There are known a line sequential method in which R, G, . Both of these have a cylindrical processing system, but have the disadvantage that the reading time is three times longer. Currently, both methods can only be practically adopted in low-speed reading devices because of the drawback that sufficient overlay accuracy of the reading positions cannot be ensured.

SUMMARY OF THE INVENTION An object of the present invention is to provide a color image reading device that secures a high S/N ratio and does not have a very complicated configuration for color image scanners such as Suyakina for color copying machines that require high speed and high image quality. It's about doing.

[Means to solve the problem]

The above purpose is to synthesize a brightness signal representing brightness information from the image sensor output for each color in a color image reading device that separates a color image and reads it with an image sensor for each color, and to represent the color information. This is achieved by including an arithmetic circuit for producing a plurality of color difference signals and making the frequency band for processing the color difference signals narrower than the frequency band for processing the luminance signals.

[Effect]

When reading a color original and making a copy, in order to obtain high image quality, the resolution must be high and color reproduction must also be good. The former is caused by the frequency characteristics (characteristics) of the signal processing system, and the latter is caused by the gradation (including S/N).

On the other hand, the characteristics of the human eye are that for color images,
It is known that the frequency characteristics are good for the luminance information, and the frequency characteristics are relatively low for the color information itself.

The present invention takes advantage of this point and decomposes the read signal into a luminance signal and a color difference signal and performs optimal processing on each, thereby simplifying the overall configuration and achieving high image quality.

Specifically, a matrix circuit composed of, for example, operational amplifiers receives the read signals Vm and V from the image sensor.
a, Vs are combined to obtain the average value of the three colors, and this is used as the luminance signal Y, and the obtained luminance signal Y and a predetermined VR,
Vc, Vm signals are compared to create color difference signals GY, BY, and these converted signals (luminance signal Y, color difference signals GY, B
Y) to the subsequent processing system. In this case, the luminance signal Y
Since the frequency characteristics are important, the processing system for the luminance signal Y requires circuit elements that cover a wide band, but since the luminance signal Y is the average value of VR, Va, and Vs, it is difficult to compare the processing for each color. and the S/N ratio can be greatly improved. In addition, since the color difference signals GY and BY do not require much frequency characteristics compared to the luminance signal Y, the processing circuit elements for this can be made inexpensive, and in some cases, it is also possible to use a low-pass filter. It becomes possible.

〔Example〕

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a functional block diagram of a color reading device according to the present invention, and 8 is a COD that reads R, G, and B for each color.
This is an image sensor circuit with an image sensor such as. Note that the read signal V* from the image sensor circuit 8,
It is assumed that Vc and Vwl are analog signals that have been preprocessed as described above. Read signal VII, Va, Vs
is input to a matrix circuit (MATL) 9 and converted into a luminance signal Y and color difference signals BY and GY. This conversion is performed as follows.

B Y = B −Y −・・・−
(2) GY=G-Y ・-・-=+
3) The luminance signal Y and the color difference signals BY, GY are each subjected to processing such as gain adjustment in the processing circuit 10, and then A/D converted by the A/D converter 11, and are converted into digital image signals. Next, matrix circuit (MAT2) 1 again
In step 2, perform the following calculation to obtain the original VG , VG , V
A digital image signal corresponding to the m signal is obtained.

R-3Y-(G+B) G=GY+Y B=BY+Y These R, G, and B digital image signals are outputted to the next processing stage via the interface 13. Such a digital data arithmetic circuit (MAT2) can be easily realized.

FIG. 2 is a detailed circuit diagram of the analog processing system, showing the matrix circuit 9. An analog processing system including a processing circuit 10 is shown.

The read signals Vl, VG, and Vl are respectively preprocessed analog image signals as described above. Pretreatment means CO
This is processing for obtaining an analog image signal from the output of an image sensor such as D, and includes sample hold, DC reproduction (zero clamp), white balance processing, etc. Note that the process of adjusting so that R, G, and B outputs are equal when a white original is read is known as white balance. These processing methods can be realized within the scope of conventional technology.

In FIG. 2, operational amplifiers 14 to 16 constitute an adder circuit together with a resistor group, and are
) performs the calculation of the expression. Inverting amplifiers 17 to 19 output a luminance signal Y and color difference signals GY and BY. Each of these is an A/D converter (A/D
i to A/D3) 20 to 22 and converted into digital signals. Note that each of the inverting amplifiers 17 to 19 can have its gain adjustable.

Here, since Y is a luminance signal, the f characteristic is important, and it is necessary to use wideband circuit elements for the operational amplifier 1, 2, and the inverting amplifier 17. However, the luminance signal Y is the read signal V, , V of each color. Since the average value of +Vl is calculated, the S/N ratio can be significantly improved compared to the conventional independent processing of each color signal.

For example, for the output of v, =v, =V, the luminance signal Y can have an S/N ratio three times that of each color signal.

On the other hand, since the color difference signals GY and BY are signals expressing color components, the f characteristic may be considerably lower than that of the luminance signal Y. i.e. operational amplifier 15.16 and inverting amplifier 18.1
The band 9 may be low, and a low pass filter can also be used. Also, low-cost circuit elements can be used. In this way, the S/N ratio can be improved for the color difference signals GY and BY by using the effect of time integration.

As mentioned above, by changing the signal processing system from independent processing of R, G, and H to synthetic conversion processing of the luminance signal Y and color difference signals GY and BY, it is possible to significantly improve the S/N while maintaining the resolution. Moreover, it is possible to reduce costs appropriately without using expensive cords for everything.

Now, regarding the A/D conversion mentioned above, the color difference signals BY, GY
Since the frequency band can be limited, the A/D conversion speed can also be reduced.

FIG. 3 is a modified circuit diagram of A/D conversion processing, in which the A/D conversion processing for the color difference signal BY and the A/D conversion processing for the color difference signal GY are shared by the A/D converter 21, A by time division
This shows an example in which the configuration is greatly simplified by performing /D conversion.

That is, the color difference signal BY is input by the analog switch 23 as an analog input to the A/D converter 21.

Select GY alternately. And analog switch 23
If the A/D output latches 24 and 25 are controlled in synchronization with the drive of the color difference signal BY.

GY can be independently A/D converted. Here A/D converter 2
When the conversion clock of 0.21 is the same, the color difference signals BY and GY are A/D at twice the period of the luminance signal Y.
It has been converted. By doing so, there is no problem with the frequency characteristics and the configuration can be simplified.

With the above configuration, it becomes easy to adjust the image quality (according to the user's preference or the type of document).

In other words, when the gain of the color difference signal GY or BY is adjusted by the gain of the inverting amplifier 18 or 19 in Fig. 2, the color density is adjusted independently of the brightness of the image (without changing the brightness). be able to.

Furthermore, if only the luminance signal Y is used, a high quality black and white image signal can be obtained.

By the way, in general, a color image processing system uses a color scanner to process R, G, B (or complementary color Y).

It is designed to receive signals from three systems: M and C).

Therefore, in the present invention, the output is returned to R, G, and B, but if the image processing system can process Y, GY, and BY signals as they are, recombination on the scanner side is unnecessary. Therefore, the memory capacity for image processing can be significantly reduced.

FIG. 4 is a transmission timing chart of the luminance signal Y and the color difference signals BY, GY, and as mentioned above, the image processing system
If GY and BY signals could be processed as they are, recombination would be unnecessary, signal transmission as shown in this figure would be possible, and communication cables, memory, etc. could be saved.

C. Effects of the Invention As explained above, according to the present invention, the overall configuration can be simplified and high image quality can be achieved by computationally converting the read signal into a luminance signal and a color difference signal and performing optimal processing on each. It is possible to provide a color image reading device that can obtain the following.

[Brief explanation of drawings]

Figures 1 to 4 relate to embodiments of the present invention; Figure 1 is a functional block diagram of a color image reading device, Figure 2 is a detailed circuit diagram of an analog processing system, and Figure 3 is a detailed circuit diagram of an analog processing system. FIG. 4 is a transmission timing chart of a luminance signal and a color difference signal, and FIG. 5 is a block diagram of a conventional color image reading device. 8... Image sensor circuit, 9... Matrix circuit, 10... Processing circuit, 11... A/D converter, 1
2... Matrix circuit, 13... Interface, 14-16... Operational amplifier, 17-19... Inverting amplifier. Figure 1 Figure 2 I7 Figure 3 Figure 4 Figure 5

Claims (2)

[Claims] (1) In a color image reading device that separates a color image and reads it with an image sensor for each color,
It is equipped with an arithmetic circuit for synthesizing a luminance signal representing luminance information from the image sensor output for each color and creating a plurality of color difference signals representing color information, and a frequency band for processing the color difference signal is set to the luminance signal. A color image reading device characterized in that the frequency band is narrower than the frequency band for processing. (2) The color image reading device according to claim 1, further comprising an arithmetic circuit for calculating and synthesizing an output signal corresponding to an original color from the luminance signal and the color difference signal.