JPH01246964A - Color picture reading system - Google Patents

Abstract

PURPOSE:To form a color picture reader which is simple in circuit configuration and high in performance by providing a storing means which stores the electric charges accumulated in each photodetector section until a prescribed output time. CONSTITUTION:Electric charges accumulated in photodetector sections 131-134 provided on a CCD image sensor are stored in a line shift gate 16 working as a storing means until a prescribed output time. Therefore, the electric charges accumulated in each photodetector section 131-134 can be outputted at the same time. As a result, the buffer memory circuit which is conventionally used for reducing the noises at the color boundary sections of originals becomes unnecessary. Accordingly, a color picture reader which is made simpler in circuit configuration and higher in performance can be formed.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a color image reading system for converting a color image into an electrical signal, and particularly to a method for driving a light receiving section thereof.

[Prior Art] FIG. 4 is a block diagram showing a general color image reading device. In FIG. 4, ▪ is a sensor substrate, 2 is a rod lens array as an imaging means, 3 is a fluorescent lamp as an illumination means for illuminating the original 4, and 5 is an original table.

Next, the operation of a general color image reading device configured as described above will be explained. A document 4 placed on a document table 5 is illuminated by a fluorescent lamp 3, and an image on the document 4 is formed by a rod lens array 2 onto a sensor substrate 1 as an erect, same-magnification real image.

Sensor board 1. The rod lens array 2 and the fluorescent lamp 3 work together to move the original 4 and the original platen 5 in the direction of the arrow 6.
The image information on the document 4 is sequentially converted into electrical signals for each scanning line.

5 and 6 are, for example, Japanese Patent Application No. 62-15484.
8 is a plan view of the sensor substrate 1 shown in No. 8, and a plan view schematically depicting the vicinity of the light receiving section on the sensor substrate 1. FIG. Fifth
In the figure, 11 is an insulating substrate, and 12 is a CCD image sensor arranged in a line on the insulating substrate 11. 6th
In the figure, 13 is one pixel, 131 to 134 are CC
This is a light receiving section disposed on the D image sensor 12. The light receiving section 131 is a light receiving section without a color filter, and the light receiving section 13
Reference numerals 2, 133 and 134 are light receiving sections having yellow, green and cyan color filters formed on their surfaces, respectively, and the four light receiving sections 131 to 134 correspond to one pixel 13.

Further, 141 and 142 are transfer gates for transferring charges generated by incident light in the light receiving sections 131 and 132 and the light receiving sections 133 and 134 to the CCD channels 151 and 152.

Further, FIG. 7 shows such a CCD image sensor 12.
3 is a time chart showing the general operation of FIG.

Next, the operation of the CCD image sensor 12 will be explained. When the transfer gate 141 is in the OFF state, light incident on the light receiving sections 131 and 132 is converted into electric charge and accumulated in the light receiving sections 131 and 132. Next, when the transfer gate 141 is turned on, the accumulated charge is transferred to the CCD channel 151. In the CCD channel 151, there is a potential well (not shown) corresponding to each of the light receiving sections 131 and 132, and a two-phase clock φ1. By applying φ2, the charges transferred from the light receiving section are sequentially transferred to the next potential well, and a floating diffusion amplifier (not shown) in the final stage is used to transfer the charges to the CCD image sensor 12.
is extracted externally as an analog signal. in fact,
As shown in the time chart of Fig. 7, the two-phase clock φ
1, φ2 (Fig. 7(b), (C1) are applied continuously, and the transfer gate pulse φT is input at a period of time TS (Fig. 7(a)). In this case, the second transfer gate pulse CCD channel 1 by gate pulse φT
The charges transferred to the light receiving section 131 during time 730
．． equal to the charge stored in 132. Therefore, C.C.
From the floating diffusion amplifier at the final stage of the D channel, signals proportional to the amount of light incident on the light receiving section 131 without a color filter and the light receiving section 132 on which a yellow color filter is formed can be alternately extracted as time-series signals. can. The other light receiving section 133°134
, transfer game) 142. The operation of CCD channel 152 is also similar to that described above.

Next, a method of converting the output values output by the above-described operation into RGB values, which are general color signals, will be described. The output values obtained through the color filters of each light receiving part of the i-th pixel 13 are Aiw, Aiy, Aig, A
ic, it is converted into RGB values using the following equation (1).

(1) The 3-by-4 matrix used in equation (1) is called a transformation matrix M. This conversion matrix M has elements such as the following equation (2), for example.

As an example, if yellow light having a size of "2" enters one pixel 13, each light receiving portion 13 in the pixel 13
1 to 134 output values Aiw, Aiy.

Aig and Aic are Aiw=2. Aiy=2. Aig=l, Aic=1. Using these values, R1
When ゜Gi and Bi are calculated, Ri=1. Gitomi1. l3i=0, and it can be seen that the color of the light incident on the pixel 13 is yellow.

However, the image on document 4 is not of one color;
There is always a color boundary. In this case, Ri, Gi, Bi
What happens will be explained in the case where the color boundary is located at a position corresponding to the center line of the pixel 13 in Fig. Δ.

For example, if white with a size of "3" is incident on the light receiving sections 131 and 132, and black with a size of "0" is incident on the light receiving sections 133 and 134, the output of each of the light receiving sections 131 to 134 is The values Aiw, Aiy, Aig, Aic are Aiw=2. Ai
y=2. Aig=1. Becomes Aic. Equations (1), (2
When Ri, Gi, and Bi are calculated using 1, R1=3. Gi=O, Bi=1. This shows a bluish red, and it can be seen that the bluish red is mixed into noise on the boundary between white and black.

Therefore, a conventional method for preventing noise at the boundary of a read image will be described.

FIG. 8 fan, (b) φ1. φ2 is the CCD in Figure 6
This is a two-phase clock applied to channels 151 and 152, and is always applied continuously. φT GAI φTGl+ in FIG. 8 (C1, (d)) are transfer gate pulses applied to the transfer gates 142 and 141, respectively, and this transfer gate pulse φTG
The cycles TSA and TSB of AI φTGR are the accumulation time. In Figure 6, the area below the center line is the A channel,
If the upper part is the B channel, the accumulation time of the A channel is T
SA, and the accumulation time of the B channel can be said to be TSB. In this conventional example, the period TSA and T
Although SB is equal, the start time of the accumulation time of each channel differs by TD. Therefore, the output signals DA, D from each channel start from immediately after the transfer gate pulses φT, A, φTGB are applied.
For 4, Alg, Alc, A2g, A2c.

In the order of...AIW for again.

Figure 8 (Q
), output as shown in (f).

Next, the operation of the color boundary portion will be explained. FIG. 9 shows the positions (in the sub-scanning direction) of the document where the light-receiving sections of the A channel and B channel in the color image reading device are located with respect to time, and P is the sub-scanning pinch. Time t shown in FIG. Assume that the A channel light receiving sections 133 and 134 are at positions yO, and the B channel light receiving sections 131 and 132 are at positions yO-(P/2). In the color image reading device, each light receiving section 131 to
134 moves relative to the document 4. 8 in Figure 9
1 is the moving state of the light receiving section 133°134 of the A channel, S
2 shows the moving state of the light receiving sections 131 and 132 of the B channel. Time t.

Assuming that the accumulation time starts from , the light incident on the light receiving sections 133 and 134 of the A channel during the accumulation time TSA is reflected light from a portion from yO to yo+p on the fourth surface of the original.

The light receiving sections 131 and 132 of the B channel are located on the fourth side of the original.
The image from −(P/2) to yo+(P/2) will be photoelectrically converted. This was the main cause of noise occurring at color boundaries. Therefore, the start time of the B channel accumulation time TSB is set to t0+ (TSA/2). As a result, as can be seen from FIG. 9, the portion from yO to yo+p on the fourth surface of the original can be photoelectrically converted for the B channel as well. Therefore, even if the color boundary portion of the document 4 is located between yo and yo+p, only intermediate colors between the colors on both sides of this boundary appear and cannot become noise.

Next, a method of signal processing of the output signals of each channel outputted from the COD image sensor 12 in this manner will be described. FIG. 8 is a time chart 1 showing the storage times TSA and TSB of each channel and the timing at which the output signals DA and D11 are sent out.

Each output signal DA, DI is sequentially outputted immediately after the end of each accumulation time. In FIG. 10, the transmission periods of the output signals DA and DI+ are indicated by diagonal lines.

Therefore, each pixel G, C, W of the output signal Da, Ds.

Since the Y pixel signal is shifted by the time T0, in order to convert it into RGB values using the above-mentioned formula (11), as shown in FIG.
It is necessary to calculate equation (1) using the signals of DA' and DA'.

FIG. 11 is a basic block diagram of a color conversion circuit that converts G, C, W, and Y to R, G, and B. CC of Figure 6
The output signal output from the D channel 151 is the 11th
In the figure, only the output signal component is sampled and held by a sample and hold circuit 71. Thereafter, the G and C time series signals are separated into G and C individual signals by a demultiplexer circuit 81. Reference numerals 91 and 92 are buffer memory circuits for delaying each of the G and C individual signals by a time T. In the figure, G' and C' indicate image signals delayed by a time T from the G and C signals, respectively.

Output signal D! from COD channel 152 in FIG. Similarly for l, the sample hold circuit 7
2 and a demultiplexer circuit 82 to separate W and Y individual signals. Individual signal GJ sampled and held through the above processing.

The C', W, and Y signals are input to the matrix calculation circuit 10 that performs the calculation of equation (1) described above, and are converted into RGB values.

As mentioned above, in the conventional signal processing method, the start time of the accumulation period of each channel of the COD image sensor is set to be different in order to reduce noise at the color boundary. The deviation is compensated for by an external buffer memory circuit.

[Problem to be solved by the invention]

Since the conventional color image reading device is configured as described above, COD is used for the purpose of reducing noise at color boundaries.
The start time of the image sensor's accumulation period is made different,
Since a buffer memory circuit is required to compensate for the resulting output signal deviation, there are problems in that the circuit configuration is complicated and the device cost is high.

This invention was made in order to solve the above-mentioned problems, and its purpose is to provide a color image reading method that has a simple circuit configuration and can realize a high-performance color image reading device at a low price. do.

[Means to solve the problem]

In the color image reading method according to the present invention, each light receiving section 131
134 is stored in the storage means (line shift gate 16) until a predetermined output time, and
1 to 134 are characterized in that the output start times of the accumulated charges are set at the same time.

[Effect]

The charges accumulated in the light receiving sections 131 to 134 are stored in the storage means (line shift gate 16) until a predetermined output time. This allows the charges accumulated in each of the light receiving sections 131 to 134 to be output at the same time. Therefore, a conventional buffer memory circuit for reducing noise at the color boundary portion of the document is not required.

[Embodiments of the invention]

An embodiment of the present invention will be described below with reference to the drawings. 1st
The figure shows the main structure of this embodiment, and is a plan view schematically depicting the vicinity of the light receiving section on the sensor substrate 1 shown in FIG. 4. In FIG. 1, 13 is one pixel, 131~
Reference numeral 134 denotes a light receiving section disposed on the COD image sensor 12. The light receiving section 131 is a light receiving section without a color filter, and the light receiving sections 132, 133, and 134 are light receiving sections with yellow, green, and cyan color filters formed on their surfaces, respectively. This corresponds to one pixel 13.

Further, 141 is a transfer gate for transferring charges generated by incident light in the light receiving sections 131 and 132 to the COD channel 151, and 16 is a transfer gate for transferring charges generated by incident light in the light receiving sections 131 and 132;
4 is a line shift gate (storage means) that captures or stores charges generated by incident light.

142 is a transfer gate for transferring the charge temporarily stored in the line shift gate 16 to the COD channel 152. .

FIG. 2 is a time chart showing the setting of the accumulation time for each row of light receiving sections arranged in two rows as shown in FIG.

Next, the operation in this embodiment will be explained. Figure 2(a),
(b) φ1. φ2 is the COD channel 151 in FIG.
．． This is a two-phase clock applied to 152, and is always applied continuously. φV in FIG. 2(C) is a pulse applied to the line shift gate 16 in FIG. 1. This line shift gate 16 includes each light receiving section 131, 132, 133 .
It has a potential well for temporarily storing accumulated charges every 134 times. In other words, line shift pulse φ■
When is a high voltage (high level), the charges in the light receiving sections 133 and 134 are transferred into the line shift gate 16, while when the line shift pulse φV is a low voltage (low level), the charges are transferred to the line shift gate 16. It is temporarily stored and memorized within the body.

φTGA in Fig. 2 fd), (e), φ'I”am
are transfer gate pulses applied to transfer gates 142 and 141, respectively. Assuming that the area below the center line shown in FIG. 1 is the A channel, and the area above the center line is the B channel, the storage times TSA and TSB of each channel correspond to the cycles of φV and φTGB, respectively, in FIG.

Next, a description will be given of how the charges accumulated in each channel in FIG. 1 move. In FIG. 1, the light receiving sections 133 and 134 of the A channel receive the incident light during the period (period TsA) from the high level (time t) of φV of C1 to the next high level (time t2) in FIG. The accumulated charge enters the potential well in the line shift gate 16 of FIG. 1 at time t2, and the charge generated by
α is done.

After that, when φTGA of FIG. 2 (dl) applied to the transfer gate 142 of FIG. 1 becomes high level (time t
, ) The accumulated charge stored in the line shift gate is the first
It is transferred to the CCD channel 152 in the figure, and after time t4,
Output sequentially.

On the other hand, in FIG. 1, the B channel light receiving section 131.1
32 is the period T between the high level of φTGffi (time L1) and the next high level (time [3) in FIG. 2(e).
sm), the charges generated by the incident light are accumulated. The accumulated charges are transferred to the CCD channel 151 through the transfer gate 141 in FIG. 1 at time t3, and the times are sequentially output thereafter.

Next, the operation of the color boundary portion in this embodiment will be explained. Figure 3 shows how far the light receiving sections of the A channel and B channel in the color image reading device are on the document with time.
(sub-scanning direction), and P is the sub-scanning pitch.

At time t0 shown in FIG.
33 and 134 are at position yO, and the light receiving section 1 of the B channel
Assume that 31,132 is at position yO-(P/2).

In the color image reading device, each light receiving section 131 to 13
4 moves relative to the original 4. In FIG. 3, SL indicates the moving state of the light receiving sections 133 and 134 of the A channel, and S2 indicates the moving state of the light receiving sections 131 and 132 of the B channel.

Time t in FIG. + tI+ t2+ t
3 is t in FIG. 2, respectively. This corresponds to the time indicated by t3.

In this embodiment, each channel's accumulation time TSA, T
The start time of SB is the same as in the conventional example.

That is, the start time 1° of the B channel accumulation time is t.

+(TSA/2). As a result, as in the conventional example, noise occurring at color boundaries is removed.

In this embodiment, furthermore, in the A channel, the period T's
The charge accumulated at A is at the end time t2 of the accumulation time TSA.
is temporarily stored in the line shift gate 16 in FIG.
It is designed to be output from time t in FIG. This time t is the same time as the start time at which the charges accumulated in the A channel are output. In other words, the storage time TSA is calculated by both channels A and B.
.

The charges accumulated in the TSB are output at the same time.

As described above, this embodiment does not require the buffer memory circuit that was conventionally required to reduce noise at color boundaries, so the color image reading device has a simple circuit configuration and is inexpensive. is obtained.

In addition, in the above embodiment, the accumulation time TSA of each channel,
Although the case where the TSBs are equal is shown, they may be at different times. Further, in the above embodiment, the light receiving sections 131 to 1
Although the case where the document 34 moves linearly relative to the original 4 has been described, the method of movement is not limited to this.

〔Effect of the invention〕

As described above, according to the present invention, the charges accumulated in each light receiving section are stored in the storage means until a predetermined output time, and the output start times of the charges accumulated in each light receiving section are made to be the same. This eliminates the need for a conventional buffer memory circuit to reduce noise at the color boundaries of originals, which simplifies the circuit configuration and has the effect of realizing a low-cost, high-performance color image reading device. .

[Brief explanation of the drawing]

FIG. 1 is a time chart shown in a color image reading device employing a color image reading method according to an embodiment of the present invention, and FIG. 3 shows the positional relationship between each channel of a CCD image sensor and a document in this embodiment. graph, 4th
9 is a block diagram showing the main parts of a general color image reading device, FIG. 5 is a plan view showing a sensor board, FIG. 6 is a time chart showing a method of driving a CCD image sensor in a conventional color image sensor, and FIG. The figure is a time chart showing the accumulation time and output time of each channel in this conventional example.
The figure is a block diagram showing a conventional color conversion method. 2... Rondo lens array (imaging means), 4.
. . . Document, 16 . . . Line shift gate (storage means) 131 to 134 . . . Light receiving unit. Agent Masuo Daikun (and 2 others) Figure 3 Figure 4 Figure 5 (b) 12+1]]■1--
]" - Figure 8 (d) φ11 几几-田CHO一" T] - Ｇａｎｔ] ゆ几 臠几凅囅11n Procedural Amendment (Volunteer) 1. Display of the case Japanese Patent Application No. 63-075043 3, representative of the person making the amendment Moriya Shiki 5 Subject of amendment □ Column for detailed description of the invention. a Contents of the amendment (1) rAiw= in the 5th and 6th lines of page 7 of the specification
2. Aiy=2. Aig=1. Aic=1" is replaced by rAiw=3. Aiy”2. Aig=O, Aic=0”
and correct it. that's all

Claims (1)

[Claims] comprising a plurality of light receiving sections that convert color images into electrical signals, and an imaging means that forms an image of the document on the light receiving sections,
The light-receiving sections are arranged in a plurality of rows, and each light-receiving section has a different accumulation start time of charges generated by incident light, and accumulates the accumulated charges for a certain accumulation time, and outputs the accumulated charges. A color image reading device characterized in that, in the reading device, the accumulated charges are stored in a storage means until a predetermined output time, and output start times of the charges accumulated in each of the light receiving sections are set at the same time. method.