JP3416309B2 - Imaging device and driving method of imaging device - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image pickup apparatus and an image pickup element driving method.

[0002]

2. Description of the Related Art In recent years, an image pickup device often used in an image pickup device is an interline CCD. With the spread of movie video cameras, mass production of CCDs for movie video cameras has led to cost reduction and movie video in many fields. CCDs for cameras have become widely used.

FIG. 13 is a block diagram of an interline type CCD generalized for movie video. In the figure, reference numeral 20 denotes a photoelectric conversion unit that transmits light in a specific wavelength range and converts incident light into electric charges, and includes a plurality of photoelectric conversion elements (hereinafter referred to as pixels). Reference numeral 21 is a vertical transfer unit for vertically transferring the charges accumulated in each pixel, 22 is a horizontal transfer unit for transferring the charges transferred from the vertical transfer unit for each horizontal line, and 23 is a horizontal transfer unit. It is a floating diffusion amplifier that converts the transferred charge into a voltage signal and outputs it.

FIG. 14 shows a sectional structure and a potential profile of a pixel portion of the present image pickup device. As shown in the figure, a vertical overflow drain is used as the anti-blooming mechanism. Such an image pickup device uses, for example, a color filter configuration of a complementary color checkered color difference sequential system. First, a signal photoelectrically converted in each pixel is transferred to each transfer stage of the vertical transfer unit during the vertical blanking period, and then, It is designed so that the upper and lower two pixels are added and read. Then, generally, as shown as field reading in FIG. 13, pseudo interlaced reading is performed in which pixels to be added in the odd field and the even field are shifted by one row.

Further, the recent trend of improving memory technology has made memory ICs smaller, higher in capacity and lower in cost, which has promoted digitalization of movie video cameras and still video cameras. In addition, the needs of multimedia demand digital image data that is not restricted by video signals. Here, digital cameras that are not regulated by video signals have also appeared in the world.

The image pickup device used in such a digital camera is not restricted by a video signal, and thus can be freely constructed. However, in order to reduce the product price, it is rather a movie. It is preferable to use the image pickup device for the video camera as it is. Here, it is important to devise a digital camera that can obtain image data of higher image quality by using an image sensor for video movie cameras and moving it with a driving method different from that of video movie cameras. Inventions such as Japanese Patent Application No. 6-137318 (filed on June 20, 1994) have been filed.

Next, the concept of the invention of the above-mentioned prior application will be shown.

FIG. 12 is a block diagram showing the configuration of a general digital camera.

Reference numeral 31 is an optical lens for forming an image of reflected light from a subject, 32 is a shutter for controlling the amount of incident light of an optical image incident from the optical lens, and 33 is an image of the subject formed by the optical lens. An image sensor for converting into an electric signal, which is generally used in movie video, for example, a complementary color checkered color difference sequential system interline CCD, 34
Is a timing signal generation circuit (hereinafter referred to as TG) that generates a timing signal necessary for operating the image sensor, 35 is a drive voltage setting circuit that generates a voltage for driving the image sensor, and 36 is a signal from the TG An image sensor drive circuit for amplifying to a drivable level, 37 is a CDS circuit or an AG
A preprocessing circuit having a C circuit, 38 is an A / D converter for converting an analog signal output from the preprocessing circuit into a digital signal, and 39 is an image pickup signal for processing the digitized signal. Image pickup processing circuit, 40 is a recording medium for recording a signal processed by the image pickup processing circuit, 41 is a recording medium I / F for sending a signal to the recording medium, and 42 is a method of starting photographing of a camera and reading an image pickup element. An operation unit for the photographer to control the (field reading method or frame reading method), and 43 is a control signal for switching the drive voltage and setting of the TG signal timing according to the reading method of the imaging method set by the operation unit. A setting switching circuit for outputting a signal for, an EVF for displaying an image pickup signal on a display, a bus controller for 45, and a temporary digital signal for 46 A buffer memory for 憶.

The digital camera shown here is a camera for taking a still image. In such a camera, an image is usually obtained by capturing an image in only one field by field reading drive, so that only an image having a resolution far exceeding the maximum resolution of the image pickup device is originally obtained. Therefore, a complementary color checkered color difference sequential system interline CCD for movie video, which is generally used, is read out without adding individual pixels as shown in the frame reading of FIG. An even field signal is read in the next field period, each signal is taken into the memory as it is, and an image signal with high resolution is obtained by performing image pickup signal processing on each pixel data stored in the memory (to avoid misunderstandings below). Therefore, the frame reading shown in FIG. 13 is called all-pixel frame reading or all-pixel reading).

However, the dynamic range is reduced to about half in all-pixel frame reading, and therefore the output of the image pickup device is about twice as high as the AGC gain in the normal addition reading (hereinafter referred to as field addition reading or simply field reading). The gain is applied, and as a result, the S / N of the image is significantly deteriorated.

[0012] In Japanese Patent Application No. Hei 6-137318, a method for increasing the dynamic range of an image pickup device at the time of reading all pixel frames while enabling switching between field reading and all pixel frame reading in a digital camera is provided. It was something to do. That is, the image pickup device is used under the same conditions as those normally used in a movie video camera during field reading, and the condition is such that the charge storage capacity of each cell of the photoelectric conversion unit is increased during all pixel frame reading.

This will be specifically described below with reference to FIGS. FIG. 14A is a cross-sectional structure diagram of the pixel portion,
(B) shows the potential of the photoelectric conversion cell in the substrate depth direction and the potential of the vertical transfer unit (VCCD) in the horizontal direction. FIG. 15 is a structural diagram of a pixel viewed from above. FIG. 22 is a time chart showing pulses applied to four charges of the VCCD of the image pickup device, and shows the timing of field reading.

The capacity of the photoelectric conversion cell is determined by the potential of the vertical overflow drain and the potential under the read gate electrode. That is, the potential wall in the depth direction is formed in the p-layer shown in FIG. 14B, the height of the wall is determined by the value of the bias voltage Vsub applied to the silicon substrate, and the higher the Vsub value, the higher the potential. The walls are low. On the other hand, the read gate electrode is common to one electrode of the transfer electrode of the vertical transfer portion (normally, FIG.
Then V1 gate and V3 gate), this V1 gate, V3
Therefore, as shown in FIG. 22, the gate is applied with the voltage Vh for reading out the signal charges from the photoelectric conversion cell to below the vertical transfer electrodes and Vm and Vl for the vertical transfer of the charges.
The m value determines the minimum level of the potential wall of the read gate during the charge storage period, and the lower the Vm value, the higher the potential wall height of the read gate during the charge storage period. The above-mentioned Japanese Patent Application No. 6-137318 pays attention to this, and when reading all pixels, rather than when reading fields,
By lowering the Vsub value and the Vm value, the saturation capacity of the photoelectric conversion cell at the time of frame accumulation is increased more than that at the time of normal use.

Next, an example of the operation of the digital camera according to the embodiment of the invention will be described with reference to FIG.

First, the photographer turns on the first switch of the operation unit 42 to start the image pickup operation in the finder mode, and the shutter 32 is operated by a diaphragm (not shown) and an electronic shutter control pulse from the timing signal generator 34. The exposure of the image sensor is controlled and the output of the image sensor is read. The preprocessing circuit 37 performs signal processing such as CDS processing and gain control on the read imaging output. At this time, the gain of the gain control circuit is determined at the time of manufacturing the image pickup device because it is determined by the sensitivity of the image pickup device. The output of the preprocessing circuit 37 is converted into a digital signal by the A / D converter 38, and is input to the imaging signal processing circuit 39 through the bus controller 45. The signal processed by the image pickup signal processing circuit 39 is output to the EVF 44, and the photographer can confirm the image pickup range and the condition of the subject.

In the operation in the finder mode, it is preferable to perform field reading in order to shorten the reading rate of the image pickup signal. Therefore, the substrate potential Vsub of the image sensor and the intermediate potential Vm of the pulse of the vertical transfer gate have the potential shown in FIG. 14 at level 1.
So that the potential becomes. Here, it is desirable that the Vsub value and the Vm value are set to be the same as those used in a normal video movie camera or the like, but it is not necessarily limited to this.

Subsequently, when the second switch of the operation unit is turned on, the image pickup device enters the all-pixel read-out photographing mode.

When the frame photographing mode is entered, first the exposure of the image pickup device is started. The exposure time starts by temporarily supplying a level 0, which is a high voltage, as a Vsub pulse to the substrate potential of the image sensor, and discharging the signal charges accumulated in the pixels up to that point to the substrate. After the exposure starts, the substrate potential Vsub and the intermediate potential Vm of the vertical transfer pulse
Are changed so that the potentials shown in FIG. 14 become level 2. Here, the level 2 electrode potential is set lower than the level 1 electrode potential. Therefore, the amount of charge that can be accumulated in each pixel is larger than that in the field read mode.

As described above, when all pixels are read out, Vs
By setting the ub potential and the Vm potential to be low (to level 2), the saturation capacity of the photoelectric conversion cell of the image sensor can be increased, and thus the saturation level of the image signal can be increased.

Here, it should be noted that when the Vm value is lowered to the level 2, the V1 electrode also serving as the read gate and the V1 electrode.
It is difficult to reduce the Vm value of only three electrodes. That is, as shown in FIG. 15, since the isolation region under the V2 and V4 electrodes and the photoelectric conversion cell is not necessarily entirely isolated by the channel stop, all the electrodes (V1, V2, V3, V4). It is better to lower both.

As a precondition for such a method, the saturation capacity of the vertical transfer unit (hereinafter referred to as VCCD) may be sufficiently high. In other words, this condition is satisfied in that all pixels are read and only charges for one pixel are handled in contrast to the field reading in which two pixels are added.

The description of the operation will be continued with reference to FIG. Figure 2
3 is a time chart of pulses applied to the electrodes of the VCCD of the image sensor when reading out all pixel frames.

Shooting in the all-pixel reading is started as described above, and the shutter is closed when the desired exposure amount is obtained.
Therefore, the exposure time for reading all pixels is from the time the electronic shutter pulse Vsub of level 0 is once supplied until the mechanical shutter is closed.

When the mechanical shutter is closed, first, for example, the Vh potential is added to the V3 gate electrode to place a color filter as shown in FIG. 13 on the CCD.
Signals from the lines Cy to Ye are read out to the vertical transfer unit. This signal is sequentially output by applying a read pulse to the first to fourth electrodes as in the timing pulse of FIG. 23.

The preprocessing circuit 37 performs signal processing such as CDS processing and gain control on the read image output. Since the saturation voltage of the image sensor has increased, the gain control is set to the same setting as that for field reading or a slightly higher gain. The output of the preprocessing circuit 37 is converted into a digital signal by the A / D converter 38 and the buffer memory 4 is converted by the bus controller 45.
6 is stored. The buffer memory 46 is shown in FIG.
With such a concept, when one line is stored, one line is vacated from the next address area, and one line is stored in the subsequent address area, and so on. When all the signal charges of the odd-numbered lines of the image sensor are output in this manner, the signals of the G and Mg lines are read out by the V1 gate, and the preprocessing is similarly performed.
After the A / D, the data is stored in the memory via the buffer memory.
Each line is stored in the blank address area as in (2).

When the writing to the memory is completed, the image data in the memory is read out in a predetermined procedure, processed by the image pickup signal processing circuit, and stored in the storage medium 40 via the storage medium I / F 41.

[0028]

However, in the above-mentioned conventional digital still electronic camera, there is a problem that when the output of the image pickup device exceeds a specific value, the output value becomes uneven and the image is remarkably deteriorated. It was As the Vm value is lowered, this unevenness occurs at a lower level of the output voltage, which is a problem that cannot be ignored particularly in the all-pixel reading. And Vm where this phenomenon occurs
It has been found that the relationship between the value and the output value of the image sensor greatly differs due to variations in the fabrication of the image sensor.

The mechanism by which this phenomenon occurs will be described below.

The following four points are listed as the main factors that determine the saturation capacity of the interline solid-state image pickup device. 1. Saturation capacity of photoelectric conversion cell. This is Vsub value and Vm
Depending on the value, the lower the potential, the larger the saturation amount. 2. Saturation amount of VCCD. This is determined by the maximum transfer capacity of the VCCD. Therefore, the higher the Vm value, the larger the saturation amount. 3. The amount of saturation of the horizontal transfer unit (hereinafter HCCD). Determined by the maximum transfer capacity of the HCCD. 4. The capacitance of the floating diffusion capacitor. Of these, 3 and 4 are excluded here because they are made sufficiently higher than the saturation capacities of 1 and 2 and are not involved in the occurrence of the above-mentioned unevenness.

Therefore, before explaining the remaining factors 1 and 2, reading from the photoelectric conversion cell to the VCCD and V
The charge transfer operation of the CCD will be described in a little more detail.

FIGS. 17A and 17B show the conventional VCCD potential profile from the read operation from each pixel to the VCCD in the field read operation and the completion of the charge transfer operation for the first VCCD line after that. Is.

FIG. 16 is a timing chart of a pulse applied to each electrode of the VCCD before and after the reading from the photoelectric conversion cell to the VCCD during the field reading which is conventionally used, and the first field and the second field are normal movie videos. This is the drive timing for the interlace used.

In the digital still camera, it is sufficient to read only one field at only one timing. Therefore, here, FIG. 17 shows a change in the potential profile when the driving is performed at the timing of the first field (in the second field, the added pixels are different but there is basically no difference, and all pixels are Since the method of changing the added pixel can be confirmed by referring to the potential profile diagrams of both fields of frame reading, the description is omitted here.).

FIG. 19 is a potential profile diagram at the time of reading the first field of all pixel frame reading, and FIG. 20 is a potential profile diagram of the second field at the time of reading all pixel frames. The potential profile of the VCCD from the reading from the cell to the VCCD to the charge transfer operation of the first VCCD is shown.

FIG. 18 (1) shows the first field at the time of reading all pixel frames, and FIG. 18 (2) shows the second field at the time of reading all pixel frames. It is a timing chart before and after reading.

From these figures, it is understood that the maximum transfer charge capacity of the VCCD is the maximum charge storage capacity when the semiconductor surface under two electrodes of the four electrodes of the VCCD is depleted.

Next, regarding the maximum charge storage capacity of the photoelectric conversion cell, as described above, the potential of the vertical overflow drain of the photoelectric conversion cell in FIG. 14 (that is, it is formed in the p-region of FIG. 14). Barrier height) and the potential under the read gate electrode (ie VC
The height of the barrier between the CD and the PD as the pixel portion). That is, the potential wall in the depth direction is determined by the value of the bias voltage Vsub applied to the silicon substrate, and the higher the Vsub value, the lower the potential wall. The readout gate electrode is shared with one electrode of the transfer electrode of the vertical transfer unit (normally, in FIG. 15, V1 gate and V3 gate are used).
Therefore, as shown in FIG. 18, the voltage Vh for reading out the signal charges from the photoelectric conversion cell to below the vertical transfer electrodes and the voltages Vm, Vl for vertical transfer of the charges are added to the V1 gate and the V3 gate. The Vm value determines the minimum level of the potential wall of the read gate during the charge storage period, and the lower the Vm value, the higher the potential wall height of the read gate during the charge storage period.

Normally, in the image pickup device, the potentials Vh, Vm, and Vl of the vertical drive pulse are designed and built so that they can be driven at a preset fixed potential.

Therefore, although the electric potential may fluctuate within the specified fluctuation, it is not moved by changing it to a value other than the specified value.

The substrate potential Vsub value is set to a potential such that the potential wall in the depth direction is lower than the potential wall below the read gate electrode determined by the Vm value. This is because Vsu is applied to the image sensor so that light leaking from the photoelectric conversion cell to the VCCD is eliminated by irradiating the image sensor with light having a light output of about 100 times the standard value.
The b value can be adjusted.

The saturation charge amount of the photoelectric conversion cell and the saturation transfer charge amount of the VCCD are obtained in this way. The relationship between the two saturation amounts is that the saturation transfer charge amount of the VCCD is the saturation charge amount of the photoelectric conversion cell. If it is not larger, the amount of charges larger than the saturation transfer charge amount of VCCD overflows before and after the transfer path during VCCD transfer, so-called blooming occurs. Therefore, the Vsub value is further adjusted so that blooming does not occur under the light amount of about 100 times the standard light.

Now, the Vsu where this is normally done
As for the adjustment of the b value, the amount of charge that can be used without any problem as the output of the image sensor when the drive voltage of the image sensor is determined in this way =
Think about what the output value will be. At first glance, this is the saturated charge amount of the photoelectric conversion cell at the Vsub value obtained by the above adjustment method, and it is easy to think that this is almost equal to the maximum charge transfer amount of the VCCD.

Here, the potential profile diagram of all pixel frame reading in FIGS. 19 and 20, the time chart of all pixel frame reading in FIG. 18, the potential profile diagram of field reading in FIG. 17, and the time chart of field reading in FIG. 16 are shown. Go back and continue the explanation.

What should be noted in these figures is from t1 to t4, that is, the signal charge reading operation from the photoelectric conversion cell to the VCCD.

Immediately before reading from the photoelectric conversion cell, odd-numbered cells (V1, V
3) is a depleted state (hereinafter also referred to as a potential well, a high potential state, the gate electrode is Vm).
Value), the even cells (V2, V
4 (below) is an inversion layer (hereinafter also referred to as a potential wall, a low potential state, the gate electrode is Vl).
Value) (above t1). Here, a Vh potential is applied to the first electrode V1 or the third electrode V3 of the charge transfer element that also serves as the read gate. Here, the signal charge of the photoelectric conversion cell in contact with the first cell of the VCCD or the photoelectric conversion cell in contact with the third cell of the VCCD is read out to the corresponding cell of the VCCD (t2).

After the reading is completed, the same state as t1 is set again and the state is held for a while (t3). In the case of field reading, a pulse is added so that the other electrode, which has not been previously read, similarly reads out the signal charge to one cell of the adjacent charge transfer element (t2 ′, t).
3 ').

The signal charge of each pixel is transferred to the adjacent one charge transfer cell, and after a while, the potential wall below the second electrode or below the fourth electrode that interrupts the charge transfer cells in which the signal charges before and after the pixel are accumulated is blocked. The electrode potential is switched from Vl to Vm so as to be lowered.

Here, a charge storage space is formed by three adjacent charge transfer cells, and in the case of field reading, signal charges of two photoelectric conversion cells are mixed. here,
At t4 in FIG. 17A, the second electrode is opened or the second electrode is opened.
Whether to leave the electrode is switched to perform pseudo interlacing in the first field and the second field.

After this, as shown at t5 and thereafter, the horizontal 1
It is sent to the HCCD line by line and the signal charges are sequentially read out.

The saturation level of the image pickup device is determined by the maximum charge transfer amount of the VCCD, that is, the maximum charge capacity that can be stored when two adjacent charge transfer cells of the VCCD are depleted. However, it is understood that the maximum accumulated charge amount stored when the first cell and the third cell are depleted is more correct.

To be more precise, in the field addition reading, it is the charge quantity obtained by adding the maximum accumulated charge quantity of the first cell and the maximum accumulated charge quantity of the second cell, which is determined by the Vm value. The maximum accumulated charge amount of each first cell and the maximum accumulated charge amount of each second cell determined by the Vm value.

Attention is paid here. That is, (maximum accumulated charge capacity of the first cell) + (maximum accumulated charge capacity of the third cell) at t3 ′ in FIG. 17A and the first cell and the second cell at t4 in the figure are simultaneously depleted. That is, the total maximum accumulated charge capacity does not become equal, and the latter usually has a larger capacity.

Now, let's consider the situation where the charge transfer cell, which has been read out in the state of t3, has a full amount of charge stored therein. The saturation amount at this time is determined by the height of the potential walls of the second cell and the fourth cell. The problem here is that the heights of these potential walls vary individually. Further, in the case of the area sensor, since the vertical transfer gate is extended around a thin aluminum layer, the potential applied to the electrodes is different between the peripheral portion and the central portion of the image area. Therefore, immediately after the pulse of Vh is applied to the electrode for reading, that is, immediately after Vh is returned to Vm, the potential of the semiconductor surface is in the dynamic change from t1 to t2 in the potential profile diagram, and t3 of When the electrode potential condition is leaned, the state becomes stable as shown in the potential profile chart t3, but at the same time, the second cell,
Due to the unevenness of the height of the potential wall of the fourth cell, some charges flow into the depleted empty charge transfer cells in the front and back or the charge transfer cells that have not reached saturation. Then, it takes several hundred nanoseconds from immediately after the electrode potential returns from Vh to Vm after the read pulse is applied until the flow of charges into the charge transfer cell in the state of the adjacent potential well almost stops.

As for the unevenness seen here, when the Vm value is lowered, the fluctuation of the potential wall does not change and the maximum accumulated charge amount of the charge transfer cell decreases, so that the unevenness occurs from the low output voltage level. And the unevenness level increases. This is the unevenness generation mechanism described above. Even if the Vsub value and the Vm value are reduced to increase the maximum accumulated charge capacity of the photoelectric conversion cell, the maximum charge capacity of one cell of the charge transfer element is reduced by decreasing Vm. However, it was not possible to obtain the expected effect because the output was reduced and large output unevenness was generated.

A simple method of suppressing the occurrence of unevenness is to make the saturated charge amount of the photoelectric conversion cell sufficiently smaller than the maximum accumulated charge capacity of the charge transfer cell which also serves as the read gate. However, in such a method, the initial purpose of increasing the output saturation voltage of all-pixel frame reading by reducing the Vsub value and Vm value at the time of all-pixel frame reading to increase the maximum accumulated charge capacity of the photoelectric conversion cell can be obtained. Instead, it will make it worse. Further, even in normal field reading, the saturation output of the image pickup device is determined by the minimum of the maximum accumulated charge capacities of the charge transfer cells which also serve as the read gates, and therefore, a level much lower than the maximum drive charge capacity of the VCCD is imaged. The saturated charge amount as an element, and eventually, the saturated output value will be obtained.

[0057]

SUMMARY OF THE INVENTION The present invention is to improve the saturation amount of an image pickup device which is regulated by the saturation charge amount of one cell of a charge transfer element which also serves as a readout gate of a VCCD in the conventional image pickup device. It is an object.

Further, by operating the image pickup device so that the saturation charge amount of two adjacent charge transfer cells of the VCCD, that is, the maximum transfer charge amount of the VCCD is regulated, all the frame pixels are read even in the field reading. Also in the above, a high saturation output level of the image pickup device is obtained, and an image pickup apparatus capable of obtaining an image with a small S / N by increasing the dynamic range of the image pickup device is provided. is there.

In particular, in the addition type CCD such as the interline type CCD which is currently mainstream, the saturated charge amount when all pixel frame reading is performed is increased, and the dynamic range as the image pickup device is improved. It is a thing.

The image pickup apparatus according to the first aspect of the present invention is the above-mentioned image pickup apparatus .
In order to achieve the object, a plurality of photoelectric conversion cells, a charge transfer means having a number of charge transfer cells is greater than the light <br/> photoelectric conversion cells, the signal of the photoelectric conversion cells to said charge transfer means At the time of transfer, a potential well is formed in the charge transfer cell at the corresponding position, and the signal of each photoelectric conversion cell is transferred to the potential well at the corresponding position by applying a read pulse to the potential well at the corresponding position. To
The capacity of the potential well is increased by applying a predetermined voltage to the transfer cell adjacent to the potential well within a few hundred nanoseconds after the signal is transferred and within a few hundred nanoseconds after the read pulse has been applied. And a control unit.

According to the method of the second aspect of the [0061] present invention, there a plurality of the photoelectric conversion cell, the driving method of the imaging device having a charge transfer means having a charge-transfer cell number greater than said photoelectric conversion cells Then, when the signal of the photoelectric conversion cell is transferred to the charge transfer means, a potential well is formed in the charge transfer cell at a corresponding position, and a read pulse is applied, so that the signal of each photoelectric conversion cell Transfer to the potential well at the position and turn the potentiometer at the corresponding position.
Capacitance of the potential well by applying a predetermined voltage to a transfer cell adjacent to the potential well within a few hundred nanoseconds after the signal is transferred to the well and within a few hundred nanoseconds after the read pulse is added. Is a method of driving an image sensor.

[0062]

[0063]

[0064]

[0065]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A first embodiment of the present invention will be described below with reference to FIGS. Here, FIG. 1 shows a case where a photoelectric conversion cell for all pixel frame reading according to the first embodiment
FIG. 1 (1) shows the read timing of the first field, and FIG. 1 (2) shows the read timing of the second field in the timing chart of the pulse applied to each gate when reading to the CD. 2 is a potential profile diagram corresponding to FIG. 1 (1), and FIG. 3 is a potential profile diagram corresponding to FIG. 1 (2).

In the first field, first, the line in which Cy and Ye are lined up is read out. Prior to reading, at t1, the first gate electrode (V1) and the second gate electrode (V2) are applied with a Vl potential, which is a low potential at the time of transfer, to the third gate electrode (V3) and the fourth gate electrode (V4). Is applied with an intermediate potential Vm which is a high potential at the time of transfer. At this time, a potential well formed under V3 and V4 is formed on the semiconductor surface under each electrode, and a potential wall is formed under V1 and V2. That is, a potential well having a capacity of two charge transfer cells is formed.

Next, in order to transfer the signal charges accumulated in the photoelectric conversion cells of Cy and Ye to VCCD, a high potential is added to V3 which also has a read gate adjacent to Cy (cyan) and Ye (yellow). Then, the potential under V3 is raised and the signal charges of the photoelectric conversion cells of Cy and Ye are moved to the semiconductor surface under V3 (t2).

When the transfer of the signal charge from the photoelectric conversion cell to V3 below is completed and the potential of V3 is returned to Vm later, the signal charge is stored in the potential well formed below V3 and V4 (t3). ). After being held for a while in this state, the transfer for the first one line is performed in the VCCD (t4 to t7).

In the second field, G (green), M
The signal charges of the g (magenta) photoelectric conversion cell are V1 and V2.
Although the drive electrodes are changed so as to be read out downward, the signal charges are read out to the two charge transfer cells in the same manner as in the first field. That is, reading is performed by the same reading method as in the first field.

By adopting such a reading method, the saturation capacity of the image pickup element, which is limited by the capacity of one charge transfer cell in the conventional reading method, is limited to the capacity of two charge transfer cells. There will be no. Here, the substrate potential Vsub value of the image sensor, which has been conventionally adjusted to be sufficiently lower than the saturation capacity of one photoelectric conversion cell, is a level at which excess charges do not flow into the VCCD side when the read gate is at the Vm value. It is possible to increase the saturated charge amount of the photoelectric conversion cell.

If the saturation charge amount of the photoelectric conversion cell at this time is lower than the maximum transfer charge capacity of the VCCD (capacity of the potential well by the two charge transfer cells), the Vm value is reduced to reduce the read gate. It is also possible to further increase the saturation charge amount of the image pickup device by raising the potential wall of (1) and simultaneously lowering the Vsub potential to raise the potential wall in the depth direction. However, Vm
The amount to be reduced must be within a range that does not reduce the charge transfer efficiency.

Although the state in which the charge from the photoelectric conversion cell is first stored is a combination of V1 and V2, V3 and V4 here, it may be a combination of V1 and V4 and V2 and V3. Of course, the timing after t4 in that case must be changed accordingly.

By doing so, the saturated charge amount of the image sensor, that is, the saturated output value of the image sensor at the time of reading all pixel frames can be significantly increased. The S / N was lowered by increasing the amplifier gain because the value is close to half of that in field reading, but it is possible to make the S / N similar to that of field accumulation. It is possible to realize a video camera capable of taking high-quality images. Needless to say, the present invention can be applied not only to still video cameras but also movie video cameras.

That is, since it is desirable to increase the saturation capacity particularly when reading all pixel frames, the effect of the present invention is great. However, even in normal field reading, the saturation capacity of the pixel also becomes large, so only the field reading is performed. However, the dynamic range of the imaging device can be expanded.

Next, a second embodiment will be described with reference to FIGS. 4 to 6. Here, FIG. 4 is a timing chart of the pulse applied to each gate at the time of reading from the photoelectric conversion cell of the all pixel frame reading according to the second embodiment to the VCCD, and FIG. 4 (1) shows the reading timing of the first field. Reference numeral 4 (2) indicates the read timing of the second field. 5 is a potential profile diagram corresponding to FIG. 4 (1), and FIG. 6 is a potential profile diagram corresponding to FIG. 4 (2).

In the first field, the line in which Cy and Ye are lined up is first read. Electrode V at t1 before reading
Vl potential is applied to 1 and Vm is applied to V2, V3 and V4.
An electric potential is applied. At this time, on the semiconductor surface under each electrode, potential wells are formed under V2, V3, and V4, and potential walls are formed under V1. That is, a potential well having a capacity of three charge transfer cells is formed.

Next, in order to read out the signal charges accumulated in the photoelectric conversion cells of Cy and Ye, when a high potential is applied to V3 which also has a read gate adjacent to Cy and Ye, the potential below V3 is raised. , Ye, the signal charges of the photoelectric conversion cells are moved to the semiconductor surface under V3 (t2). The potential of V3 is returned to Vm after the movement of the signal charge from the photoelectric conversion cell to below V3 is completed,
The signal charges are stored in potential wells formed under V2, V3, and V4 (t3). After being held in this state for a while, the transfer of the first one line is performed in the VCCD (t4 to t8).

In the second field, the electrodes to be moved are changed so that the signal charges of the G and Mg photoelectric conversion cells are read below V1, V2, and V4, but as in the first field, three charge transfer cells are formed. The signal charge is read. That is, reading is performed by the same reading method as in the first field.

By doing so, the factor limiting the saturated charge amount of the image pickup device is the same as in the first embodiment.
It is loosened up to one (the maximum transfer charge amount of VCCD).

Now, here, the state in which the charge from the photoelectric conversion cell is first accumulated is set to V1 and V2 for V1 reading.
And V4, and V2, V3, and V4 are combined for V3 read, but V1, V2, and V for V1 read.
3, a combination of V1, V3 and V4 for V3 reading, or V1, V3 and V4 for V1 reading,
A combination of V1, V2, and V3 may be used for V3 reading. Of course, the timing after t4 in that case must be changed accordingly.

Next, a third embodiment will be described with reference to FIGS. 7, 19 and 20.
Will be described.

Here, FIG. 7 is a timing chart of pulses applied to each gate at the time of reading from the photoelectric conversion cell for all pixel frame reading according to the third embodiment to the VCCD, and FIG. 7 (1) shows reading of the first field. Timing, FIG. 7B shows the read timing of the second field. FIG. 19 is a potential profile diagram corresponding to FIG. 7 (1), and FIG. 20 is a potential profile diagram corresponding to FIG. 7 (2). This diagram, as can be seen from the previous use in the conventional example, The transition of the basic potential profile of the embodiment is not different from that of the conventional example.

The difference from the conventional example of FIG. 18 is t in the conventional example.
The time tw for maintaining the state of 3 is extremely short in this embodiment.

In the prior art example, if the signal charge stored in one cell of the charge transfer cell during the period of tw is the amount before and after the saturation of the charge transfer cell, the potential of the signal amount below the saturation of the preceding and following stages. It has been described that there is an excess charge flowing into the well of the above, and that such an excess charge inflow occurs several hundreds of nanoseconds after the end of the read pulse from the photoelectric conversion cell to the VCCD. This means that, conversely, if tw is within a few hundred nanoseconds, excess charges will not leak to the front and rear stages.

Further, looking a little more closely, it is shown that during this period, there is a process of dynamic change from the potential state of t2 in FIGS. 19 and 20 to the potential state of t3. Therefore, if a potential well is formed by two or three cells adjacent to the charge transfer cell during this transition period, the charge storage saturation capacity is substantially increased over that of one cell. In addition, the unevenness which occurs in the conventional output and which exceeds the specific output value does not occur.

As can be seen from the above principle, tw is several hundred nanoseconds or less, and it is clear that the shorter the tw, the higher the effect.

Here, the method of forming the potential well by the three charge transfer cells immediately after the end of the read pulse from the photoelectric conversion cell to the VCCD is described, but even if the potential well by the two charge transfer cells is formed. It goes without saying that the same effect can be obtained.

The embodiment for increasing the saturated charge amount of the image pickup device at the time of reading all pixel frames has been described above. According to the embodiment as described above, it is possible to significantly increase the saturated charge amount of the image sensor, that is, the saturated output value of the image sensor at the time of reading all pixel frames. The S / N was lowered by increasing the amplifier gain because it is close to half that in field reading, but it is possible to achieve S / N similar to field accumulation, and high quality in both field reading and frame reading. It is possible to realize a digital still camera that can take images of.

Next, a fourth embodiment for increasing the amount of charge saturation in the image sensor during field reading will be described.

A fourth embodiment will be described with reference to FIGS. 8 and 9. Here, FIG. 8 is a timing chart of pulses applied to the respective gates at the time of reading from the photoelectric conversion cell for all pixel frame reading according to the fourth embodiment to the VCCD, and FIG.
(1) is the read timing of the first field, as shown in FIG.
(2) shows the read timing of the second field. FIG. 9 is a potential profile diagram corresponding to FIG. 8 (1). The basic operation of both the first field and the second field, which is the main point of this embodiment, is the same as that for reading all pixel frames, so the potential profile diagram of the second field will be omitted. The second field example will be omitted).

The video movie camera realizes the pseudo interlace by repeating the first field and the second field, and the still camera normally reads only one of the first field and the second field.

Now, the present embodiment is an application of the principle of the third embodiment of reading all pixel frames. Cy, Y
The signal charge on the e side and the signal charge on the G side and the Mg side are connected from each of the photoelectric conversion cells to V, which also serves as a read gate adjacent to each.
1, V2 or V4 is depleted so that the signal charges of the preceding stage and the following stage are added immediately after the reading is finished, and the signal charges are mixed and added and accumulated in three adjacent charge transfer cells. Then, in the reading of the conventional example, V
The read pulses from the photoelectric conversion cells 1 and V3 to the VCCD were sequentially applied, but here, if they are applied sequentially, the electrode side that first applies the pulse is the addition that can be performed first after the end of the read pulse. Since the time tw until the formation of the potential well space can be set within several hundred nanoseconds, the in-phase pulse is added to the read pulses applied to V1 and V3.

By doing so, the saturated charge amount of the image pickup device is not limited by the saturated charge amount of the individual charge transfer cells even in the field addition read,
This is limited by the maximum charge transfer charge amount of the CD, and the saturation charge amount of the image pickup device can be increased.

Next, a fifth embodiment of the present application will be described with reference to FIGS.
Will be described with reference to. Here, FIG. 10 shows a photoelectric conversion cell for all-pixel frame reading according to the fifth embodiment, which is converted from a photoelectric conversion cell to a VCCD.
10A and 10B are timing charts of pulses applied to the respective gates at the time of reading from the first field, and FIG. 10A shows the timing of reading the first field, and FIG. FIG. 11 is a potential profile diagram corresponding to FIG. 10 (1).

The present embodiment has the same principle as the first and second embodiments. The charges of the individual photoelectric conversion cells are read so as to enter the potential wells of the three charge transfer cells. As a result, the saturation charge capacity of the image sensor is not limited by the saturation capacity of one charge transfer cell. However, as is clear from t4 of the potential profile diagram, attention should be paid to this reading method.
The signal of 2 photoelectric conversion cells is input to 1 charge transfer cell during the period of t4. Depending on the structure of the image pickup device, the saturated charge amount of the image pickup device may be limited to the maximum charge storage amount in the situation where the Vh potential of the photoelectric conversion cell on the second readout side in this state is added. Therefore, it is difficult to say that the effects of this embodiment can be obtained reliably as in the fourth embodiment, and it is desirable to carry out the operation after sufficiently confirming the structure of the image sensor to be used. .

In the present embodiment, both the read at V1 and the read at V3 are driven so that the signal charges enter the potential wells of the three charge transfer cells, but the charge obtained by the first charge read is two charges. The data may be read into the potential well of the transfer cell (for example, V3 at t1 and t3 is set as a potential wall).

By carrying out the fourth and fifth embodiments as described above, the saturated charge amount of the image pickup device can be increased even during field reading.

Next, a sixth embodiment will be described without showing it.

The camera in the above embodiment carries out a series of photographing in which field reading is performed in the first photographing and subsequent frame photographing is performed in the second photographing.

The first photographing by the camera of the sixth embodiment is the field reading of the conventional example or the reading by the fourth or fifth embodiment of the present invention, and the second photographing is the actual photographing. The reading is performed according to the first, second, or third embodiment. And at the beginning of the second shooting, Vm value,
The Vsub value lowers the charge transfer efficiency in a range that does not impair the charge transfer efficiency as compared with the first shooting. Of course, at the time of the second charge reading, light does not enter the image sensor due to the mechanical shutter.

By adopting such a driving method of the image pickup device, the digital camera of the present invention can make the gain of the output signal from the image pickup device the same in the first and second photographing, and It is possible to obtain high-quality images with low S / N both in the second time and in the second time.

Although the above embodiments have been described on the premise of the color difference line-sequential color filter, the present invention is not limited to the color filter of the above type. That is, it does not matter if there is no color filter, the so-called all-pixel-frame reading method in which the information of the odd-numbered rows and the information of the even-numbered rows are read separately, after adding the information of the odd-numbered rows and the information of the even-numbered rows in the transfer register. Any reading method such as a field reading method for reading or a method for sequentially reading the signals of each row by non-interlace can be applied. The point is, of course, any type may be used as long as the saturation capacitance is not limited by the potential well of the transfer register when the pixel signal is read by the transfer register.

[0103]

As described above, according to the present invention, it is possible to perform readout for increasing the saturated charge capacity of the image pickup device. This effect is particularly great in all-pixel frame reading in the color difference sequential system sensor, which is currently the mainstream. Therefore, particularly in a digital camera, a high-resolution digital still image can be obtained using a low-cost image pickup device for video movies.

[Brief description of drawings]

FIG. 1 is a pulse timing chart of the first embodiment.

FIG. 2 is a potential profile diagram of the first embodiment.

FIG. 3 is a potential profile diagram of the first embodiment.

FIG. 4 is a pulse timing chart of the second embodiment.

FIG. 5 is a potential profile diagram of the second embodiment.

FIG. 6 is a potential profile diagram of the second embodiment.

FIG. 7 is a pulse timing chart of the third embodiment.

FIG. 8 is a pulse timing chart of the fourth embodiment.

FIG. 9 is a potential profile diagram of the fourth embodiment.

FIG. 10 is a pulse timing chart of the fifth embodiment.

FIG. 11 is a potential profile diagram of the fifth example.

FIG. 12 is a configuration diagram of a digital camera.

FIG. 13 is a configuration diagram of an image sensor.

FIG. 14 is a cross-sectional structural view of an image sensor and its potential profile diagram.

FIG. 15 is a structural diagram of an image sensor.

FIG. 16 is a pulse timing chart of a conventional example.

FIG. 17 is a potential profile diagram of a conventional example.

FIG. 18 is a pulse timing chart of a conventional example.

FIG. 19 is a potential profile diagram of a conventional example.

FIG. 20 is a potential profile diagram of a conventional example.

FIG. 21 is a conceptual diagram of data arrangement of storage in a buffer memory.

FIG. 22 is a timing chart of drive pulses of the image sensor.

FIG. 23 is a timing diagram of drive pulses for an image sensor.

Claims (2)

(57) [Claims] And 1. A plurality of photoelectric conversion cells, a charge transfer means having a charge-transfer cell number greater than said photoelectric conversion cells, when transferring the signal of the photoelectric conversion cells to said charge transfer means, the corresponding By forming a potential well in the charge transfer cell at the position and applying a read pulse, the signal of each photoelectric conversion cell is transferred to the potential well at the corresponding position, and the signal is transferred to the potential well at the corresponding position.
And a control means for increasing the capacitance of the potential well by applying a predetermined voltage to the transfer cell adjacent to the potential well within a few hundred nanoseconds after the application of the read pulse. An imaging device having: 2. A plurality of photoelectric conversion cells, a driving method of an image sensor having a charge transfer means having a number of charge transfer cells is greater than the photoelectric conversion cell, the charge signals of the photoelectric conversion cell When transferring to the transfer means, a potential well is formed in the charge transfer cell at the corresponding position and a read pulse is applied to transfer the signal of each photoelectric conversion cell to the potential well at the corresponding position. Signal is transferred to the potential well of
And increasing the capacitance of the potential well by applying a predetermined voltage to a transfer cell adjacent to the potential well within a few hundred nanoseconds after the read pulse is added. And a method for driving an image sensor.