JPH0683402B2 - Solid-state 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 a charge transfer type solid-state image pickup device used in a video camera or the like.

[0002]

2. Description of the Related Art In recent years, a charge transfer device type (hereinafter referred to as CCD) image pickup device for a video camera has been put into practical use and is becoming widespread. At present, since the CCD system has a smaller number of pixels and a higher price than the image pickup tube, the image pickup tube is predominant as an image pickup element of a video camera. However, as the density of semiconductors increases, the number of pixels of the CCD system increases and the cost is also decreasing. Therefore, it is expected that the CCD type image pickup device, which has advantages such as small size, light weight, long life and low power consumption, will become the mainstream as an image pickup device for a video camera.

There is no problem when the conventional method is used for the current video camera. This is because consumer video cameras are just beginning to spread and automation is not progressing. Currently, the aperture is only automated, but in the future, the second white balance and the third focus will be automated. It will take 10 years for these automations to reach completion. During this time, no problem occurs in the conventional configuration.

However, even if these automations are completed, camera shake of the video camera will remain a big problem in the future for consumers who do not have photographing technology. Although it has not been revealed at all at present, it is predicted that the image stabilization will be automated as the fourth automation technology. The mechanical image stabilization method may be a mechanical method, an optical method, a memory method, or the like, but the appearance of a method of processing in the image sensor can be foreseen. Conventional solid-state image pickup devices for video cameras do not correspond to the problems in the distant future at all.

The conventional system will be described below by using two examples of a typical interline system and a frame transfer system.

FIG. 16 shows a conventional interline CC.
The structure of a D solid-state image sensor is shown. Here, a pair of a light detection pixel portion 71 such as a photodiode for light detection and a transfer pixel portion 72 for information transfer are arranged in a matrix in the horizontal and vertical directions. The actual solid-state image sensor has 400 to 800 rows of pixels in the horizontal direction and 250 to 500 pixels in the vertical direction.
It is composed of pixels in columns. It also has interlace processing and color filters arranged in a mosaic pattern. However, in order to facilitate the explanation, the operation principle will be described with reference to the drawing in which the number of pixels, the interlace process, and the color filter are omitted, and there are 4 rows in the horizontal direction and 5 columns in the vertical direction.

The optical image formed on the upper surface of the image pickup section 5 is photoelectrically converted into pixel charges on each light detection pixel section 71. A vertical synchronizing pulse is applied from the transfer pulse circuit 73 to all the light detecting pixel portions 71 based on the vertical synchronizing signal of the TV signal. With this pulse, all the pixel charges are transferred from the light detection pixel unit 71 to the transfer pixel unit 72 all at once in one field, as shown by an arrow 73a.

The first vertical clock circuit 75a and the second vertical clock circuit 75b of the vertical transfer circuit 75 output vertical transfer clocks which are interlocked with the horizontal synchronizing signal of the TV signal. By this signal signal, each transfer pixel 72 is vertically transferred on the vertical transfer units 74a, 74b, 74c, 74d in the lower direction of the screen. Of the pixels transferred in the downward direction, the charges of one line in the lowermost stage are temporarily stored in the horizontal transfer unit 76 that transfers the charges in the horizontal direction. Next, in accordance with the horizontal transfer clock signals of the first horizontal clock circuit 77a and the second horizontal clock circuit 77b of the horizontal transfer circuit 77, the signals are transferred in the right direction in the figure, reach the signal output circuit 78, and after color demodulation, Finally, it is output as a color TV signal. The above is the basic operation of the conventional interline method.

Another conventional system is a frame transfer CCD. This is one in which a storage unit is provided outside the light receiving region in addition to the image pickup unit in the light receiving region. Basically, all pixel charges of the image pickup section which are photoelectrically converted even in the interline method and during the first field period are simultaneously transferred to the storage section in conjunction with the vertical synchronizing signal of the TV signal. Then, in the next second field period, the pixel charge in the storage portion is
The signals are sequentially output as TV signals by vertical transfer and horizontal transfer. Further, the number of pixels of the light receiving section and the storage section are the same.

[0010]

However, in the above structure, the pixel charges photoelectrically converted from the optical image are fixed to the light detecting pixels 71 during one field period of the TV signal. As described above, when used in the conventional video camera, this configuration does not cause any problems. However, when used in a video camera with an electronic image stabilization function that will appear in the distant future, new problems will occur. When external conditions such as camera shake change significantly, the optical image moves at high speed on the image forming surface of the image sensor due to the shake of the body of the video camera even during a short period of one field. In this case, the position of the optical image at the first time of one field and the position of the optical image at the last time of one field are different. Therefore, since the pixel charge based on the previous optical image and the pixel charge based on the moved optical image are mixed, the resolution of the TV signal is lowered. Due to this image deletion, a problem that a beautiful image cannot be obtained is predicted.

The present invention solves the above-mentioned problems of the prior art, and even during the light receiving period of one field or one frame, the pixel charge being received is bidirectionally in the horizontal and vertical directions according to an external control signal such as a camera shake correction signal. Transfer control to 1
An object of the present invention is to provide a solid-state imaging device that can transfer charges by following an optical image even when the optical image moving frequency per pixel is higher than the frequency of one field and has a small number of elements.

[0012]

In order to achieve this object, a solid-state image pickup device of the present invention transfers a pixel charge to a photo-detecting pixel portion for photoelectrically converting an optical image arranged in a matrix and bidirectionally in a horizontal direction. A horizontal pixel charge transfer unit, a light receiving unit having a vertical direction pixel transfer unit for bidirectionally transferring pixel charges in the vertical direction, a storage unit for storing the pixel charge group transferred from the light receiving unit, and a storage unit The image output unit outputs the pixel charge transferred from the image output unit.

[0013]

With this configuration, the pixel charges that are photoelectrically converted during the photoelectric conversion of the optical image during one field or one frame period according to the external control signal such as the camera shake correction signal are converted into the first horizontal charge of the light receiving portion. By the transfer circuit and the first vertical direction charge transfer circuit, it is possible to transfer and control the pixel charges in a direction that follows a bidirectional optical image in the vertical and horizontal directions. Therefore, even if the optical image moves on the light-receiving surface at high speed due to camera shake or the like, the correspondence relationship between the optical image and the photoelectrically converted pixel charges is maintained during the period of one field. Therefore, even if image correction such as camera shake correction is performed, deterioration of resolution and image quality can be prevented. It is also possible to significantly reduce the number of elements in the storage unit.

[0014]

(Embodiment 1) An embodiment of the present invention will be described below with reference to the drawings.

The block diagram of the image pickup device shown in FIG. 1 shows main blocks such as an image pickup section 5, a vertical / horizontal transfer circuit, and a vertical / horizontal control circuit 11a and a horizontal control circuit 11b. The solid-state image pickup device according to the first embodiment of the present invention is an image pickup unit 5 having a CCD image pickup device having a structure similar to that of a conventional frame transfer method.
Is used. As shown in FIG. 1, the upper half of the image pickup unit 5 is a light receiving unit 90 in which light detection pixels 71 that perform photoelectric conversion are arranged in a matrix. Actual CCD is 500x50
Although it is composed of 0 or more pixels, in order to facilitate the explanation in the drawings, the illustration of pixels of 6 vertical rows and 7 horizontal rows in which the number of pixels is deleted is used. It is also assumed that a camera shake correction circuit is combined as an external circuit. The light receiving unit 90 is connected to a light receiving unit vertical / horizontal transfer circuit 91 that transfers each pixel in four vertical / horizontal directions. The light receiving unit vertical / horizontal transfer circuit 91 detects the pitch direction detection circuit and the yaw direction detection of the camera. By connecting it to the circuit, it is possible to correct the camera shake.

Further, in the lower half of the image pickup section 5, only the pixel information necessary for correction among the pixel information of the light receiving section 90 is stored in the external T.
The light receiving unit horizontal transfer circuit 91 and the vertical transfer circuit 75 are interlocked with the vertical synchronizing signal of the V period signal during the vertical blanking period.
To transfer all at once. In this way, the pixel information for the first field is stored in the storage unit 92. In the figure,
The storage unit 92 has pixels in a matrix of 3 columns horizontally and 3 rows vertically. FIG. 2 shows a state in which the transfer of the first field is completed at t = t1. The hatched circles in the storage unit 92 indicate the accumulated pixel information of the 0th field of the field preceding the 1st field.

Upon completion of the transfer, the transfer section charge removing circuit 66d removes the charge from the light receiving section 90 by the transfer section charge removing circuit 66d so as not to mix into the storage section 92. Until the reading of the first field is completed, the pixel information in the storage unit 92 is transferred by the vertical transfer circuit 75 and the horizontal transfer circuit 77 independently of the transfer control of the light receiving unit 90, and by the output circuit 78. It is output as a TV signal of the first field and finally demodulated into a color TV signal.

At t = t2, as shown in FIG. 3, the vertical transfer circuit 75 transfers the pixel information of the first field vertically downward and sequentially sends the pixel information to the horizontal transfer unit 76.
The horizontal transfer circuit 77 transfers one line to the right, and the output circuit 78 starts outputting the TV signal.

At t = t3, the scanning line at the end of the first field is scanned as shown in FIG. 4, and at t = t4, as shown in FIG. 5, reading of all pixel information is completed and the vertical blanking period starts. After that, the function of preventing the mixture of charges from the light receiving unit 90 to the storage unit 92 by the transfer unit charge removal circuit 66d is stopped, and the charges can be transferred from the light receiving unit 90 to the storage unit 92. At t = t5, as shown in FIG. 6, the pixel information of the second field of the light receiving unit 90 is transferred vertically downward by the light receiving unit vertical transfer circuit 91 and the vertical transfer circuit 75. At t = t6, as shown in FIG. 7, among the pixel information received by the light receiving unit 90 during the second field light receiving period, necessary pixel information of the second field (circle number 1 to 9) is stored in the storage unit 92. To the coordinates (3, 6), (4,
6), (5, 6) to activate the charge removal function of the cell,
The charge inflow from the light receiving unit 90 to the storage unit 92 is prevented. As a result, the state returns to the initial state of the scanning cycle of the first field of t = t1 shown in FIG. 2, and thereafter, similarly to the previous time, the light receiving unit 90 and the accumulating unit 92 independently operate until the next vertical blanking period. , Transfer charges separately.

Therefore, in the case of the third embodiment, the vertical transfer circuit 75 and the horizontal transfer circuit 77 of the accumulating section 92 do not have the function of mainly performing control such as image correction of the oscillation, and do not have them at all. The control of the image correction of the swing is performed by the light receiving unit vertical / horizontal transfer circuit 91 in the light receiving unit 90 during the light receiving period of one field or one frame.

Next, the image correction control in the light receiving section 90 will be described. When it is desired to change the image output control, the electric signal corresponding to the optimum fluctuation correction control amount on the image forming surface on the image pickup section 5 is the vertical control circuit. 11a and the horizontal control circuit 11b,
It is sent to the light receiving unit vertical / horizontal transfer circuit 91.

In this case, the difference between the conventional example and the first embodiment is that the former is that pixel information for vertical and horizontal image correction of pixel information during one field or one frame scanning period. The image is corrected by performing charge transfer for image correction in the vertical direction and the horizontal direction mainly during the vertical blanking period without transferring the charge based on. However, in the case of the first embodiment, it is mainly performed during the light receiving period of the pixel information of one field or one frame.
4 (a) to (e) (described later) in the vertical direction,
In accordance with a horizontal image control signal, a real-time charge transfer system is adopted in which potential wells in the CCD substrate are transferred vertically, vertically in the horizontal direction and horizontally in the horizontal direction. Here, as shown in FIG. 2, there are 7 rows in the horizontal direction and 6 columns in the vertical direction of the light detection pixel 71, and at a certain time t = t1, 7 × 6 = 42 circle marks in the figure are shown. The pixel information indicated by is obtained. Of this pixel information,
It is assumed that pixel information of an object to be obtained is imaged in a range indicated by a dotted rectangle on nine pixels numbered 1 to 9 in a circle. In this case, photoelectric conversion is performed in each of the light detection units 71 of 1 to 9, and charges corresponding to each pixel of the pixel information of the subject are stored in the potential well of each pixel in FIG.
As shown in (b) (described later), it is accumulated during the exposure time. Here, an example in which the CCD of the present invention is combined with a camera shake correction detection circuit of a video camera will be described.

While the time for one field is not completed, the camera oscillates in the yaw direction until the next time t = t2, and the imaged optical image of the target object becomes the horizontal coordinate ( 4-6) When moving to the range indicated by the dotted line of the vertical coordinates (3-5), if no measures are taken, the plural pixel information passed during the light receiving period is mixed into one pixel, and the image is blurred. I will end up. In the present invention, for example, the horizontal correction amount of the camera shake in the horizontal direction is output from the horizontal control circuit 11b. Based on this information, the light receiving unit vertical / horizontal transfer circuit 91 transfers the charges in each pixel in the horizontal direction by the correction amount. Therefore, when t = t1 shown in FIG. 2, the respective charges based on the pixel information of the copy body indicated by the circle numbers 1 to 9 in the horizontal coordinates (3 to 5) and the vertical coordinates (3 to 5) are It is transferred in the horizontal direction without being slid as in the previous embodiment, and as shown in FIG. 3, follows the movement of the optical image of the target object and is successively pulled to the adjacent cell. become. Therefore, even for a fast horizontal signal, it is not necessary to interrupt the photoelectric conversion of light from the subject by using the high-speed shutter as in the above-described conventional example as long as it is within the maximum transfer rate of the transfer element. Then, let us assume that the maximum transfer rate is 100 ns per pixel, one side of the final output screen is 500 pixels, and the worst value of the swing width due to camera shake is 100% of the screen. Then, it follows an ultra-high-speed horizontal control signal of 20000 Hz within 1 second, and the frequency characteristic of correction control can be remarkably improved as compared with other conventional image stabilization systems. Moreover, there is no reduction in sensitivity. Naturally, the frequency characteristics can be improved in the same manner in the vertical direction. Therefore, the correction control frequency characteristics in the vertical and horizontal directions are remarkably improved and are limited only to the frequency characteristics of the correction amount detecting means. Image correction such as tilting oscillation can be performed at high speed by a combination of horizontal correction and vertical correction. In the case of a system that controls this ultra-high-speed response according to the swing detection means,
Various application effects such as camera shake correction of a consumer video camera can be obtained.

[0024] For example, in an electronic camera for which standardization is currently in progress for consumer use, camera shake caused by camera shooting is caused by a conventional solid-state image sensor for one field scanning time, that is, for a shutter speed of 1/60 second. Countermeasures are a problem. In the case of the CCD described in the conventional example, a measure to shorten the exposure time can be considered. However, there is a problem that the sensitivity is lowered. However, by using the image sensor according to the first embodiment of the present invention, it means that the shutter speed is equivalently 10 ⁻⁷ sec at the maximum value with respect to camera shake at the time of shooting a still subject, and camera shake By combining it with a sensor, it is possible to realize an electronic camera with practically no camera shake.

However, as a matter of course, if the one-field exposure method is used, the shutter speed is only 1/60 seconds for a moving subject, and in order to stop the moving subject, the exposure time is shortened by electronic means. If so, another measure such as provision of a moving object detection means becomes necessary. However, most of the blurring is caused by camera shake, and in the case of a still camera, most of the image blurring is due to the movement of the camera body, especially during handheld shooting with a telephoto lens. Can be greatly suppressed by detecting. Therefore, the image sensor according to the present invention provides the possibility of realizing an electronic camera capable of hand-held photography of a still subject using a super telephoto lens due to its high-speed response characteristic. Further, the solid-state image pickup device of this embodiment can be applied not only to a consumer-use video camera but also to a video camera for broadcasting station. For example, as you can see from the slow video image of night game relay, even if you use a sturdy tripod, you can see that each image is flowing at the time of slow playback when you swing it. However, this problem is also corrected by using the solid-state image sensor of the present invention, while a moving object such as a ball is not improved, but a stationary object such as a background such as a ground does not flow and is clearly corrected. Therefore, even when used in a video camera for broadcasting stations, each still image of a static subject at the time of panning is corrected without deterioration of sensitivity, and slow motion or image deletion during still screen broadcasting is prevented. The effect of being able to be obtained is obtained.
As described above, even when used in an electronic camera or a camera for a broadcasting station, this solid-state image pickup element is effective when photographing a static subject such as a landscape, a background, or a building.

By using a separately provided high-speed image recognition means, it is possible to follow a moving object without mechanically moving the camera body or the optical system. There is a possibility in the future that the image of can be corrected at high speed.

Returning to the explanation of the principle of operation, in order to further increase the sensitivity, photoelectric conversion is performed by means such as employing transparent electrodes or lenses on the light receiving portions of the pixels or on the transfer cells between the cells. The function can be constantly added. As a result, the photoelectric conversion of the optical image that moves during one field or one frame is not interrupted at all, and the optical image is successively drawn to the adjacent cells. Therefore, it is possible to obtain the effect that the reduction in the sensitivity due to the control of the image range is reduced.

With the charge transfer as shown in FIGS. 2 to 3 in the horizontal direction, the horizontal coordinate 6 and the vertical coordinates 1 to 6 in FIG.
In FIG. 3, the electric charges that existed are in the horizontal coordinate 7 and the vertical coordinate 1 to
No. 6 is merged with each other, and in some cases overflows to deteriorate image quality due to blooming or the like. Therefore, the charge removing circuit 66 removes the charge in the peripheral portion from the charge removing portion provided in the peripheral pixel portion in accordance with the transfer. As a result, it is possible to prevent the overflow of electric charges in the peripheral portion due to the transfer and prevent the deterioration of the image quality. In this case, this circuit may not be provided, and a charge emitting portion may be provided in the peripheral cell so that the charge is always emitted to the substrate or the like. However, as described above, in the three pixel units with the coordinates (3, 6) (4, 6) (5, 6), the charge from the light receiving unit 90 does not flow into the pixel information storage unit 92 during the light receiving period. As described above, the transfer portion charge removing circuit 66d prevents the transfer of the charges or removes the charges, so that there is an important effect of preventing the deterioration of the output image due to the leakage of the charges.

The image correction in the vertical direction will be described.
When the optical image of the subject moves upward in the vertical direction as shown by the dotted arrow in FIG. 4 from t = t ₂ to t = t ₃ within the scanning time of the same field or the same frame. In the same manner as the horizontal correction, the vertical correction circuit 11a provides the light receiving unit vertical horizontal transfer circuit 91 with information on the optimum correction amount. Number 1 which photoelectrically converted the subject image
The pixel information indicated by circles 9 to 9 is transferred to the corresponding portion of the subject image by the light receiving unit vertical / horizontal transfer circuit 91 as in the case of the horizontal direction, and each pixel of the subject image is transferred as shown in FIG. Information is continuously photoelectrically converted for one field or one frame without interruption. In the vertical direction,
While repeating the control in the horizontal direction, at t = t ₄ when the reception of the pixel information of one field or one frame is completed, the pixel information indicated by circles 1 to 9 as shown in FIG. Regardless of the position of the image part,
The light receiving unit vertical / horizontal transfer circuit 91 performs high-speed horizontal transfer to the horizontal coordinates 3 to 5, and then the transfer unit charge removal circuit 6
Coordinates (3, 6) that blocked charge transfer by 6d
(4, 6) (5, 6) open the pixel portion, from the light receiving portion 90,
Pixel information can be transferred to the storage unit 92, and the light receiving unit vertical / horizontal transfer circuit 91 and the vertical transfer circuit 75 transfer the pixel information indicated by circles 1 to 9 in the downward direction of the drawing in the vertical direction, and t
= T ₅ , as shown in FIG. 6, the pixel information of the subject is transferred from the light receiving unit 90 to the accumulation unit 92, and at t = t ₆ , as shown in FIG. Since the transfer to the storage unit 92 is completed in the same vertical blanking period and the transfer unit charge removal circuit 66d prevents the charge from being mixed from the light receiving unit 90 to the storage unit 92, the charges of the light receiving unit 90 and the storage unit 92 are prevented. Are independently and separately transferred, and the pixel information (circle number 1 to 9) of the previous field or the previous frame in the storage unit 92 is the same as t = t ₁ shown in FIG. Image information is output from the output circuit 78 by reading the pixel information. On the other hand, in the light receiving unit 90, the subject to be imaged is imaged in the range indicated by the rectangular dotted line in FIG. 7, and the light detection pixels (circles 11 to 18) in that portion have the light amount of each pixel of the subject. Accumulation of the corresponding charge is started. Then, as described above, in response to the swing of the camera body, the accumulated charges are transferred and controlled by the light receiving unit vertical / horizontal transfer circuit 91 in the direction for correcting the swing of the image, as in the control in the previous field.

The above is the description of the basic operation of the first embodiment. Next, the operation principle of the light receiving unit vertical / horizontal transfer circuit 91 will be described in more detail with reference to the enlarged view of each cell.

FIG. 8 is an enlarged view of pixels arranged in seven columns in the horizontal direction and six columns in the vertical direction. Each pixel has a target structure as shown in the figure, and is represented by coordinates (7, 1). Like A,
It is composed of nine cells B, C, D, E, F, H, and I. Among them, the cells A, C, G, and I shown by the diagonal lines are P
14 is a charge transfer inhibition region 93 provided with a channel stopper due to diffusion of N-type or N-type impurities, and a cell E is a light detection pixel portion 71 for light detection. As shown in FIG. 7, an electrode 96e is provided on a P-type or N-type semiconductor substrate 94 via a thin insulating film 95 such as SiO ₂ , and the electrode 96e is connected to the common clock circuit 91a. The cells D and F are cells for horizontal transfer, and electrodes 96d and 96f for horizontal transfer are provided on the upper surface as shown in FIG.
Is connected to the first horizontal clock circuit 91b, and the electrode 96
f is connected to the second horizontal clock circuit 91C.
14 (b) to 14 (e) (described later) show the state of the charge well depending on the interface potential at each time point. As shown in FIG.
(B) shows a state in which electric charges are being accumulated in the well of the cell E of the photodetection section by photoelectric conversion as indicated by circles. In the case of a P-type substrate, electrons that are minority carriers are accumulated by photoelectric conversion. The cells B and H are vertical transfer cells for performing vertical transfer vertically.
As shown in the cross-sectional view of (a), electrodes 96b and 96h are provided on an upper portion with an insulating film 95 interposed therebetween, and charges photoelectrically converted by the optical image of the subject are shown in the well.
It is accumulated as shown in (b).

Next, a specific vertical / horizontal four-direction transfer operation will be described. An enlarged view of the state at t = t ₁ shown in FIG. 2 is shown in FIG. 9, and the dotted rectangular portion in the range of horizontal coordinates 3 to 5 and vertical coordinates 3 to 5 includes the background of the subject to be photographed. Each pixel information plate of the subject, which indicates the image range, is photoelectrically converted by the light detection pixels, and circles of numbers 1 to 9 indicate charges corresponding to the pixels. This cross-sectional view is shown in FIG. 14A and FIG. 14B shows the interface potential state in this case. As described above, in the case of the P-type substrate, electrons become transfer charges, and the potential well becomes shallow when a LOW voltage is applied to the electrode 96. Therefore, FIG.
In (b), the electrodes 96f and 96d are LOW and the electrode 96e is
Is High and the E cell and the next E cell are 4,
Pixel information indicated by a circle 5 is accumulated. FIG. 9 is a top view of this state. The electrodes are
When set to W, charge transfer is blocked. Here, in order to facilitate the explanation, as in the previous embodiment, the LOW electrode portions are indicated by square marks. Therefore, the cells in the squares in FIGS. 9 to (f) are meant to mean that the transfer of charges is blocked. Here, the pixel information items 1 to 9 in FIG. 9 are surrounded by cells of electrodes of LOW potential, which are surrounded by square marks, and transfer of accumulated charges is blocked. FIG. 14A is a horizontal cross-sectional view of the substrate of the imaging unit 5 showing the pixel portion with the circled number 4 at the coordinates (5, 4), and FIG. In b), the charge of each pixel is confined in the charge well and cannot move in the horizontal direction. Further, FIG. 15A is a vertical cross-sectional view of the substrate of the same pixel, and FIG. 15B shows a state of potential. Each pixel is confined in a charge well and It is in a state where it cannot move.

From the above description, at t = t ₁ , the charge of each pixel is fixed both in the horizontal direction and in the vertical direction.

Next, by the change of the external condition, the image of the subject moves to the right in the figure by the time t = t ₂ as shown in FIG. 3, and this is detected by the horizontal detecting means. Based on this information, while following the movement to the right of the image, the principle of operation in which each charge is horizontally transferred to the right by changing the voltage applied to each electrode when transferring the charge to the right as shown in the figure. Will be explained. The electrodes are transferred cell by cell, and first move from the state of t = t _{1 in} FIG. 9 by one cell in FIG. During the horizontal charge transfer period, the first vertical clock circuit 91d and the second vertical clock circuit 91 generate a LOW potential in order to prevent the transfer charge from leaking in the vertical direction, as shown in FIG. The first vertical transfer electrode 96b and the second vertical electrode 96e have a LOW potential and the potential state is as shown in FIG. The charges are transferred horizontally. Since this state is maintained during the horizontal charge transfer period, a horizontal transfer section that is continuous with cells, D, E, F, D, E, and F must be electronically formed on the image sensor. become. Therefore, after that, as shown in FIGS. 14B to 14E, the operating voltages of the first horizontal clock circuit 91b, the common clock circuit 91a, and the second horizontal clock circuit 91c are changed to change the cell of each electrode portion, D, The electric potentials of the electrodes at the E and F portions are changed. In the first state, as shown in FIG. 14B, since D = LOW, E = High, F = LOW, the charges based on the pixel information of numbers 4 and 5 are fixed in the well. Next, D = LOW, E = High, F =
When it is set to High, the well becomes
It expands to the right and the charge moves to the right. Next, by gradually lowering the potential of E, the charge moves to the right, and
As shown in FIG. 14D from the state of (c), D = LO
W, E = LOW, F = High, a well is formed only under the cell of F, and the horizontal transfer cycle of the charge of one cell is completed from the cell of E to the cell of F on the right. After that, as shown in FIG. 14E, the potential of D is set to High.
Then, by the configuration of D = High, E = LOW, and F = High, the well is expanded to the right and the horizontal transfer cycle of charges from the cell of F to the cell of D on the right is started and the same transfer is performed as described above. By the principle, the charge is transferred from the F cell to the D cell on the right side. This state is followed by the optical image of the subject shown by the dotted rectangle in FIG. Therefore, not only the E cell originally for light detection, but also the F cell originally for charge transfer and the D cell for light detection are made to have a light detection structure by means of adoption of a transparent structure electrode or the like, so that photoelectric conversion is performed during charge transfer. The important effect of preventing a decrease in sensitivity due to transfer is obtained without interruption, and by adopting this image sensor in a video camera, electronic still camera, etc., a pure electronic image correction effect can be obtained even in a dark place. An imaging device is obtained. Then, in the next horizontal transfer cycle, each charge of each pixel is transferred from the D cell to the E cell on the right side, and the pixel horizontal transfer cycle of the charge for one pixel is completed, as shown in FIG. , The electric charge of each cell portion corresponding to each pixel of the subject is transferred horizontally in the right direction while following the optical image formed by the subject indicated by the dotted rectangle. On the contrary, when horizontal transfer is desired in the leftward direction, each charge is transferred in the leftward direction by performing an operation reverse to the horizontal transfer cycle of the cell charge in the rightward direction. Specifically, when the transfer cycle from the cell of F to the cell of E on the left is shown, first, the state as shown in FIG.
A potential of D = LOW, E = LOW, F = High is created on the signal timing chart, and then, FIG.
Then, the horizontal transfer cycle of charges from the lower cell to the cell E on the left is completed by changing the state to the state shown in FIG. 14B. In this way, the horizontal transfer in the left direction becomes possible in the same manner as the horizontal transfer in the right direction, and the image pickup device of the present embodiment is further left and right for the image correction control signals on the left and right in the horizontal direction. The effect that the image can be corrected without lowering the sensitivity is obtained. During the horizontal transfer cycle, the cells B and H are LOW as shown in the vertical potential diagram of FIG. 32 (b) to prevent the leakage of charges to the cells B and H in the vertical direction as described above. It is a voltage.
As a result, it is possible to prevent the leakage of charges in the vertical direction during the horizontal correction and to obtain the effect of not deteriorating the sharpness of the image even if the image is corrected.

Now, the vertical transfer of charges in the vertical direction for image correction in the vertical direction will be described. First, when the external condition changes from the state of t = t ₂ shown in FIG. 3 to the state of t = t ₃ shown in FIG. 4, that is, when the optical image of the subject moves vertically upward, A vertical transfer cycle in which charges are vertically transferred upward will be described. The basic operation principle is the same as the horizontal charge transfer cycle. In the vertical direction, as shown in the horizontal potential diagram of FIG. 14B, the potentials of the D cell and the F cell are changed to the vertical transfer cycle. During the period of time (3), it is set to LOW to prevent the vertical transfer charge from leaking in the horizontal direction. As a result, a vertical transfer portion that transfers charges vertically and vertically is electronically formed on the image sensor. In this state,
15B shows the potential diagram of the pixels of Nos. 4 and 7 in the enlarged view of the light receiving unit shown in FIG.
As shown in (a), the first vertical clock circuit 91d, the common clock circuit 91a, and the second vertical clock circuit 91e are applied with a potential through each electrode. In this case, B = LOW and E = High. , H = LOW, and the charges of No. 1, No. 4 and No. 7 are fixed in the well of small E. As described above, D = LOW, F =
It becomes LOW, and leakage of electric charges in the horizontal direction is prevented.
Next, as shown in the potential diagram of FIG. 14C, B = H
By setting high, E = High and H = LOW, E
The well in the cell No. 1 is expanded to the cell B adjacent in the upper direction, and accordingly, the respective charges of No. 1, No. 4, and No. 7 move vertically upward. Next, as in the horizontal transfer cycle, by gradually applying a LOW voltage to the E cell, the charges of No. 1, No. 4, and No. 7 are further transferred in the upward direction, as shown in FIG. Then, the cell B is transferred almost completely to the cell B in the upper direction of the original cell E, and the vertical transfer cycle of the cell is completed. FIG. 12 shows an enlarged view of the light receiving portion in this state. Further, as shown in FIG. 15 (e), a vertical transfer cycle from the cell of B to the cell of H adjacent to the upper side is started, and finally, as shown in the enlarged view of the light receiving portion of FIG. To complete the vertical transfer cycle. In the downward vertical transfer cycle, the timing chart of the transfer clock signal is changed, and the applied voltage is applied so that the potential diagram of FIG.
By setting the potential state of (c) and the potential state of FIG. 14 (b), downward vertical charge transfer becomes possible. After that, the charge including the pixel information in the light receiving unit 90 mainly during the vertical blanking period is transferred to the storage unit 92 and is output as an image signal, as described in detail.

In the conventional method, a charge discharging means is provided in the light detection pixel 71 for movement of an optical image having a high movement speed,
As the rate of change increases, the charge emission time is extended. In other words, the charge accumulation time associated with the photoelectric conversion of the light detection pixel 71 is shortened. That is, the faster the movement of the optical image due to rocking or the like, the faster the shutter speed and the clearer correction screen can be obtained, but the exposure time becomes shorter, so the sensitivity is reduced accordingly. This is not a problem when shooting outdoors in the daytime, but it becomes a problem when shooting indoors at night. To explain the principle of the CCD, the charge transfer type image pickup plate changes the applied voltage of an electrode provided with a charge well, that is, a bucket in the substrate, and moves it to transfer the charge in the well, that is, the bucket. It's just like transferring charges sequentially in a bucket. The conventional CCD image pickup device is a method of accumulating charges generated by photoelectric conversion in a fixed bucket during an exposure time of one frame or one field. The bucket position is not moved during the exposure time. Then, in the interline method, during the vertical blanking period, the accumulated charges are transferred to the transfer pixels 72, which are the buckets provided adjacent to the charges of all the buckets, in the buckets of the light detection pixels 71. . On the other hand, in the frame transfer method,
The buckets of all pixels for one frame are transferred all at once to a separately provided frame storage unit.

In the conventional CCD, the normal exposure time is 1
When the camera shake cannot be corrected within 1/30 seconds or 1/60 seconds of the scanning period of one frame or one field, a method of shortening the exposure time according to the swing speed can be considered. That is, if the oscillation is fast, unnecessary charges stored in the bucket of the light detection pixel unit 71 are discarded, and the sensitivity is reduced accordingly. In the system of the third embodiment, in order to prevent this decrease in sensitivity, each pixel stored in the bucket of each light detection pixel 71 in real time in response to a detection signal such as oscillation during one field or one frame scanning period. Information charges are vertically bidirectional, that is, vertically and horizontally, that is, horizontally, that is, horizontally, both vertically and horizontally, as if they were bucket relays. It is a method to transfer by the control circuit. While following the movement of the imaging unit 5 on the image plane due to the swing of the formed optical image, the charge in each bucket moves in a direction to be corrected in one field or one frame in real time. Therefore, even if the formed image moves at a high speed due to the oscillation of a frequency faster than the vertical synchronizing signal, it follows within the range of the charge transfer rate, so that most camera shake of the video camera can be corrected. The electric charge generated by photoelectric conversion of pixel information in one field or one frame is a conventional CCD
However, according to the present invention, it is possible to efficiently store the data for one field or one frame without the need for the processing. Therefore, it is possible to prevent a decrease in sensitivity due to image control for camera shake correction. can get.

As described above, by using the image pickup device and the control circuit of this embodiment, the image formation optical image can be moved far faster than the fastest moving speed of the image formation optical image that can be considered under the normal use condition of the video camera. Follows accurately at high speed.
Therefore, it is possible to obtain a solid-state image pickup device which follows a high-speed screen correction and has a correction effect without lowering the sensitivity. As described in the description of the embodiment, it is possible to obtain the remarkable frequency response and the effect that the sensitivity can be maintained purely electronically. In the embodiment, the example of the pixel charge transfer in the four directions of the vertical bidirectional pixel charge transfer and the horizontal bidirectional pixel charge transfer is shown, but the vertical or horizontal bidirectional pixel charge transfer may be performed. , The same effect can be obtained although it is one-dimensional. Further, in the case of the conventional transfer transfer type CCD, the number of pixels of the light receiving portion and the storage portion is the same. However, by using the present invention, as shown in FIG. 1, the number of pixels in the storage section can be greatly reduced compared to the number of pixels in the light receiving section. Therefore, there is an effect that the chip area can be remarkably reduced. For example, when it is used for camera shake correction, a correction range of 50% is required vertically and horizontally. In this case, if the solid-state imaging device of the present invention is used, the number of pixels in the storage section can be reduced to 1/4 of the conventional method. The numbers of rows and columns of the light receiving unit and the storage unit may be the same.

[0039]

As described in detail above, the present invention is
In accordance with an external control signal such as a camera shake correction signal, the vertical and horizontal bidirectional charge transfer control of the vertical transfer unit and the horizontal transfer unit of the light receiving unit is performed in real time even during the period of one field, and the pixel charge is arbitrarily moved vertically and horizontally. While moving, the pixel information within an arbitrary range can be taken out from all the pixel information of the light receiving unit. Therefore, by providing a camera shake correction circuit externally, correction of camera shake and the like of the image pickup section of an industrial or consumer video camera can be processed in the image pickup device. Therefore, as compared with the conventional camera shake correction processing method using a CCD for a video camera, the pixel charge can be directly moved inside the image pickup section, so that there is a remarkable effect of following the camera shake correction signal at a higher speed than the vertical synchronization signal. Therefore, C for conventional video cameras
It was impossible with the electronic correction method using a CD. Image quality deterioration during high-speed image stabilization. There is an effect of disappearing. In addition, mechanical and optical image stabilization that has been conventionally performed in industrial and consumer video cameras can be performed purely electronically without degrading image quality, which has the effect of improving reliability, reducing size and weight. . In addition, since the number of elements in the charge storage section is smaller than that in the light receiving section, a 1-chip CC
This has the effect of greatly reducing the number of D elements.

[Brief description of drawings]

FIG. 1 is a block diagram of a solid-state imaging device according to a first embodiment of the present invention.

FIG. 2 is a diagram showing the principle of operation of charge transfer in the first embodiment of the present invention.

FIG. 3 is a diagram showing the principle of operation of charge transfer in the first embodiment of the present invention.

FIG. 4 is an operation principle diagram of charge transfer in the first embodiment of the present invention.

FIG. 5 is an operation principle diagram of charge transfer in the first embodiment of the present invention.

FIG. 6 is a diagram showing the operating principle of charge transfer in the first embodiment of the present invention.

FIG. 7 is a diagram showing the operating principle of charge transfer in the first embodiment of the present invention.

FIG. 8 is an enlarged view of a light receiving portion of the solid-state imaging device according to the first embodiment of the present invention.

FIG. 9 is an operation principle diagram of charge transfer in the enlarged light receiving portion of the first embodiment of the present invention.

FIG. 10 is an operation principle diagram of charge transfer in the enlarged light receiving portion of the first embodiment of the present invention.

FIG. 11 is an operation principle diagram of charge transfer in the enlarged light receiving portion of the first embodiment of the present invention.

FIG. 12 is an operation principle diagram of charge transfer in the enlarged light receiving portion of the first embodiment of the present invention.

FIG. 13 is an operation principle diagram of charge transfer in the enlarged light receiving portion of the first embodiment of the present invention.

14A is a cross-sectional view of a horizontal charge transfer portion according to the first embodiment of the present invention. FIG. 14B is an interface potential diagram showing the horizontal charge transfer principle according to the first embodiment of the present invention. c) Interface potential diagram showing the horizontal charge transfer principle of the first embodiment of the present invention (d) Interface potential diagram showing the horizontal charge transfer principle of the first embodiment of the present invention (e) The present invention Interface potential diagram showing the principle of horizontal charge transfer in the first embodiment of

15A is a vertical cross-sectional view of the solid-state imaging device according to the first embodiment of the present invention. FIG. 15B is an interface potential diagram showing the principle of vertical charge transfer according to the first embodiment of the present invention. c) Interface potential diagram showing the vertical charge transfer principle of the first embodiment of the present invention (d) Interface potential diagram showing the vertical charge transfer principle of the first embodiment of the present invention (e) The present invention Interface potential diagram showing the vertical charge transfer principle of the first embodiment of FIG.

FIG. 16 is a block diagram of a solid-state image sensor described as a conventional example.

[Explanation of symbols]

 5 image pickup units 11a and 11b vertical and horizontal control circuits 65 reference time signal unit 66 charge removal circuit 71 photodetection pixels 72 transfer pixels 73 transfer pulse circuits 74a to 74d vertical transfer units 75 vertical transfer circuits 76 horizontal transfer units 77 horizontal transfer circuits 78 signal output circuit 80 horizontal transfer control circuit 81 output control SW 82 reset SW 90 light receiving part 91 light receiving part vertical horizontal transfer circuit 92 accumulating part 94 image pickup device substrate 95 insulating layer 96 electrode

Claims (3)

[Claims] 1. A light detection pixel portion arranged in a matrix of a plurality of rows in a horizontal direction and a plurality of columns in a vertical direction to convert light of pixel information into pixel charges corresponding to each pixel by photoelectric conversion, and the light detection pixel portion. A first vertical charge transfer unit that has a detection pixel unit as a part and transfers the pixel charges in the vertical direction, and a first vertical charge transfer unit that has the light detection pixel unit as a part and transfers the pixel charge group in the horizontal direction. A light receiving portion having a horizontal charge transfer portion, a second vertical charge transfer portion coupled to part or all of the first vertical charge transfer portion, and a second horizontal direction coupled to the second vertical charge transfer portion. A charge transfer unit is arranged and stores and outputs the pixel charge group transferred in the column direction from the light receiving unit, and a row unit and a row direction from the second horizontal charge transfer unit of the storage unit. Image that outputs the transferred pixel charge The pixel charge group, which is composed of an output unit and is transferred in a batch from the light receiving unit to the storage unit once in synchronization with a vertical synchronizing signal of an external TV signal, is transferred to the image output unit to output the image. In the charge transfer device for outputting a continuous TV signal from the unit, the pixel charge groups on the first horizontal direction charge transfer unit and the first vertical direction charge transfer unit are set to the horizontal bidirectional and vertical directions according to an external image correction signal. A solid-state imaging device characterized by bidirectional transfer. 2. A vertical bidirectional and a horizontal bidirectional in the light receiving unit, mainly during a period in which pixel charges are continuously transferred from the storage unit to the image output unit and a continuous TV signal is output from the image output unit. 2. The pixel charge transfer according to claim 1 is performed.
The solid-state imaging device described. 3. A charge removing circuit is provided, and a charge removing section is provided in the peripheral portion of the light receiving section, and a part of the transferred pixel charge is removed by a control signal from the charge removing circuit. 1. The solid-state imaging device according to 1.