JP3366518B2 - Image sensor - 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 sensor that uses a charge transfer device (hereinafter abbreviated as CCD) to transfer an image at high speed.

[0002]

2. Description of the Related Art Image processing technology for measuring the position of an object by processing image information is based on FA (factry automatio).
n) Widely used in various fields. The speeding up of these image processing is desired for speeding up FA devices and developing new applications, and the processing speed of image data is increasing year by year. Therefore, in order to increase the speed of the image processing system, it is essential to improve the processing speed of the image sensor for inputting an image to the system. In order to obtain such an image sensor, it is conceivable to transfer one frame image at a higher speed.

Generally, when a two-dimensional image is transferred by the CCD, the entire image obtained by the CCD is transferred sequentially. The procedure in this case is shown in FIG. First, the photoelectric conversion light receiving element 1 is irradiated with light, and charges are generated according to the obtained light amount. Under the control of the charge transfer signal input to the charge transfer signal line 10, the charges are transferred to the vertical CCD 2 all at once. The transferred charges are controlled by the vertical CCD drive signal input to the vertical CCD drive signal line 11, and one horizontal line from the bottom is transferred to the horizontal CCD 15. The charges transferred to the horizontal CCD 15 are output from the output signal line 16 pixel by pixel. After that, each time the horizontal CCD 15 completes the transfer of one line, the vertical CCD 2 transfers one line, and all the charges are transferred.
Information for the screen is output.

However, assuming that the screen is composed of 640 horizontal pixels × 480 vertical pixels, the number of pixels on the screen is 640 × 480 = 307,200, and when transferring 60 frames of image per second. Then, 307, 2
This means that the horizontal CCD 15 is transferring 00 pixels × 60 = 18,432,000 pixels. In order to transfer the image of one frame at a higher speed, it is necessary to operate the horizontal CCD 15 at a speed higher than this. However, it is difficult to make the operating speed higher than this because there is a problem that heat generation of the horizontal CCD 15 and the like becomes large.

Therefore, Japanese Patent Laid-Open No. 3-124176 discloses a technique for realizing a CCD camera capable of a high frame transfer rate that can be used for slow reproduction. In the above publication, by dividing the horizontal CCD into a plurality of areas, the horizontal 1
The line image is divided into a plurality of images. The images are transferred at high speed by performing the parallel operation of reading the output from each area at the same timing to obtain the image of the entire one screen.

However, in image processing, it is very often necessary to process information of a part of an image rather than processing of information of the entire screen of the image. It is more effective to increase the frame transfer rate by transferring only necessary information than transferring.

For example, as shown in FIG. 37, there is a circular object 101 within the entire image area 100.
When it is known that 01 exists only in the range of the area 102, the image area required for position measurement of the object 101 is only the range of the area 102, and the others are not necessary. Therefore, in such a case, the operation of transferring the area other than the area 102 is useless.

Therefore, in order to shorten the required image transfer cycle, there has been proposed a technique for eliminating unnecessary transfer and transferring only the area 102.

For example, in Japanese Unexamined Patent Publication No. 2-97180, as shown in FIG. 38, when the image transferred to the horizontal CCD 15 is not needed, one horizontal line is not transferred and the image is swept away. An image sensor that transfers data at high speed is disclosed. That is, in this image sensor, the plurality of overflow gates 103 have the horizontal CCD 1
5 and the overflow selection signal line 106 and the sweep signal line 105 are connected to each overflow gate 103. As a result, when a line that does not need to be transferred is input to the horizontal CCD 15, the sweep signal is simultaneously sent to each overflow gate 103 via the sweep selection signal line 106.
The image for one line is discharged from the sweep signal line 105.

Further, in Japanese Patent Laid-Open No. 2-295283, as shown in FIG. 39, one horizontal line to be transferred to the horizontal CCD 15 is arbitrarily selected, so that only the necessary horizontal lines are transferred to increase the speed. An image sensor that does this is disclosed. That is, in this image sensor, one electrode is connected to the photoelectric conversion light receiving element 1, and the other electrode is connected to the vertical CCD.
This is a configuration having a switch element 107 connected to 108. At this time, the other electrode is commonly connected to one vertical CCD 108 in the column direction. As a result, the photoelectric conversion light-receiving element 1 is selected row by row by the switch element 107, and only the necessary lines are transferred to the vertical CCD 108. Then, from the output signal line 16 of the horizontal CCD 15 to 1
The pixels for the line are output.

[0011]

However, the conventional image sensor has a problem that the processing speed cannot be increased.

That is, in the thinning method in units of horizontal lines as described above, the horizontal width 111 and vertical width 112 in FIG.
In the case of an image area having a long vertical width 112 such as the area 110 indicated by, there are few horizontal lines that can be thinned out,
Almost the same time is required as when transferring all images. Further, as another area example, when there are two areas 113 shown in FIG. 41 and two target areas 110 shown in FIG. 42, high-speed transfer is similarly performed to the irregular area 114 shown in FIG. 43. I can't.

In order to perform high-speed transfer, it is conceivable to use an image sensor having a small number of pixels for picking up only the area 110. However, in this case, although one frame can be transferred at high speed, there are problems such as not being able to grasp the entire image and being unable to change the imaging range because it is fixed.

By the way, when trying to transfer only a specific area at high speed while grasping the whole area, an image sensor having a small number of pixels for picking up only the area 110 and a normal image as shown in FIG. 36 for picking up the whole area. 2 with sensor
It is conceivable to provide two image sensors. In this case, in order to grasp the whole, a normal image sensor is used to shoot at a normal frame transfer rate, and when only the area 110 is needed, an image sensor with a small number of pixels is used to shoot at a high frame transfer speed. With, you can shoot both whole and part.

However, when a plurality of optical systems are used to use a plurality of image sensors in this way, the angle at which an image is captured and the enlargement ratio are different for each image sensor, so that pixels are associated between different image sensors. Therefore, software for adjustment and calculation of correspondence between pixels is required.
Since such adjustment is required, the processing speed of the image sensor as a whole is reduced. Also,
Since two image sensors are required, the cost is high, and a lot of space for installing the two image sensors is required.

When it is attempted to transfer only a specific area at high speed while grasping the whole with one image sensor without using two image sensors as described above, FIG. 38 and FIG.
A configuration is conceivable in which high speed transfer by partial transfer using the image sensor shown in 9 and transfer at the normal speed of the entire image are switched using a switch or the like. However, since the whole image cannot be transferred during the high-speed partial transfer, it is necessary to perform the high-speed image processing by the partial transfer and at the same time perform the whole inspection by the silent observation using the whole image area. There is no choice but to use two image sensors in parallel and solve the problem.

The present invention has been made to solve the above-mentioned conventional problems, and an object thereof is to improve the processing speed by thinning out pixel units and outputting images at high speed. It is to provide a read image sensor.

Another object of the present invention is to provide a multi-readout image sensor capable of outputting an image by switching between two output systems of partial read-out and whole read-out.

[0019]

In order to achieve the above object, the image sensor according to claim 1 of the present invention comprises a plurality of photoelectric sensors arranged in a matrix and accumulating signal charges according to the amount of incident light. Converter (for example, photoelectric conversion light receiving element)
And a plurality of vertical charge transfer units arranged along each column of the photoelectric conversion units, and a vertical charge transfer unit that sequentially sends out the signal charges taken in from the photoelectric conversion units one pixel at a time from each vertical charge transfer unit ( for example, with a vertical CCD), and a selection transfer means for transferring the signal charges sent from the vertical charge transfer means is arbitrarily selected pixel, the selective inversion
The sending means opens and closes based on the output selection signal .
The signal charge sent simultaneously from each vertical charge transfer
Multiple switching elements for selection and sweep selection
The selection switch is opened and closed based on the selection signal.
Sweep out the signal charges that are not selected for the switching element
It is characterized by being composed of a plurality of switching elements for sweeping .

According to the above structure, the photoelectric conversion section allows
A signal charge corresponding to the amount of light emitted is generated. The signal charge is drawn in by each vertical charge transfer unit adjacent to each column of the photoelectric conversion units, and is simultaneously sent out to the selective transfer unit one pixel at a time. That is, the signal charges sent from the vertical charge transfer means are input to the selective transfer means as pixels for one horizontal line. The selective transfer means selects and transfers only necessary signal charges and discharges unnecessary signal charges to the outside. By performing selective transfer in the same manner for each horizontal line, an image for one frame is transferred.

At this time, it is possible to select only the necessary pixels from the pixels for one horizontal line and output the partial image, so that high-speed transfer is possible for the time required to transfer the unnecessary pixels. Further, since the image sensor of the present invention thins out in pixel units instead of sweeping out in horizontal line units as in the past, it is possible to efficiently and rapidly output any shape region. Furthermore, by changing the way of selecting pixels by the selective transfer means, it is possible to take out an image reduced to 1/2 or an image compressed in the horizontal direction.

According to the above arrangement, the selective transfer means is
The transferred signal charges are transferred by the switching element for selection.
While the required signal charge is selected, the sweep switch
An unnecessary signal charge is swept away by the switching element. this
As a result, the selective transfer means is composed of only multiple switching elements.
Since it is configured, the selective transfer means can be configured with an easy configuration.
It becomes possible to do.

An image sensor according to a second aspect of the present invention is arranged in a matrix form, and has a plurality of photoelectric conversion units for accumulating signal charges according to the amount of incident light, and a plurality of photoelectric conversion units arranged along each row of the photoelectric conversion units. Vertical charge transfer section, which is capable of sending out the signal charge taken in from the photoelectric conversion section one pixel at a time from each vertical charge transfer section, and the signal charge sent out from the vertical charge transfer section. , And a horizontal charge transfer means (for example, horizontal CCD) for transferring the signal charges sent from the vertical charge transfer means in horizontal line units.
And a control means for controlling so that the signal charge from the vertical charge transfer means is input to either the selective transfer means or the horizontal charge transfer means.

According to the above structure, the signal charges generated by the photoelectric conversion section are drawn into the vertical charge transfer means. When this signal charge is controlled by the control means so as to be inputted to the selective transfer means, pixels for one horizontal line are inputted to the selective transfer means. The selective transfer means selects and transfers only necessary signal charges and discharges unnecessary signal charges to the outside. On the other hand, when the control means controls the signal charge from the vertical charge transfer means to be input to the horizontal charge transfer means, pixels for one horizontal line are input to the horizontal charge transfer means. The horizontal charge transfer means transfers and outputs all the signal charges pixel by pixel.

Therefore, it is possible to selectively output whether the partial image is transferred at high speed by the selective transfer means or the entire image is transferred at normal speed by the horizontal charge transfer means. This enables multi-reading in which one image sensor outputs images at two different transfer rates.

As a result, it is not necessary to prepare an optical system or an image sensor for each area, so that the cost and the work for installation can be reduced. Moreover, there is no need for a space for installing them. Further, compared to the case where a plurality of optical systems and image sensors are arranged, complicated processing and adjustment for making correspondence between pixels do not have to be performed, so that the processing speed of the apparatus as a whole can be improved. .

In the image sensor according to a third aspect, in addition to the configuration according to the second aspect, the vertical charge transfer means has output terminals at both ends in the longitudinal direction of each vertical charge transfer section.
Bidirectional vertical charge transfer means (for example, bidirectional vertical CCD) capable of respectively transferring signal charges to one output terminal.
The selective transfer means is connected to one output end of the bidirectional vertical charge transfer means, the horizontal charge transfer means is connected to the other output end, and the control means drives the bidirectional vertical charge transfer means. The drive signal controls which of the two output terminals the signal charge is transferred to.

According to the above structure, the selective transfer means is connected to one output end of the bidirectional vertical charge transfer means, and the horizontal charge transfer means is connected to the other output end. At this time, when the signal charge transferred to the bidirectional vertical charge transfer means is controlled by the drive signal and input to the selective transfer means, a partial image is output. On the other hand, when the drive signal is input to the horizontal charge transfer means, the entire image is output.

This makes it possible to realize a multi-readout image sensor that selectively outputs images at two different transfer rates with a simple structure without providing a complicated circuit or the like.

In the image sensor according to a fourth aspect , in addition to the configuration according to the second or third aspect , the selective transfer means opens and closes on the basis of the output selection signal so that the vertical charge transfer sections simultaneously operate. A plurality of switching elements for selecting the sent signal charges arbitrarily, and a plurality of sweeping elements that open and close based on the sweep selection signal to sweep away the signal charges not selected by the above switching elements. It is characterized by comprising a switching element.

According to the above-mentioned structure, the signal charges transferred to the selective transfer means are selected by the selection switching element while the unnecessary signal charges are swept away by the sweeping switching element. Thus, the selective transfer means is composed of only a plurality of switching elements, so that the selective transfer means can be configured with a simple configuration.

[0032]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiment 1 The following will describe Embodiment 1 of the present invention with reference to FIGS. 1 to 20 and with reference to FIGS. 36 and 38.

The image sensor according to the present embodiment is
As shown in FIG. 1, photoelectric conversion light receiving element 1, vertical CCD
2, a buffer 3, a selection switching element 4, and a sweeping switching element 6.

The photoelectric conversion light receiving elements (photoelectric conversion portions) 1 are arranged in a matrix of n in vertical and m in horizontal to form a light receiving portion. The photoelectric conversion light receiving element 1 accumulates signal charges according to the amount of incident light and transfers them to the vertical CCD 2 at a certain timing. Note that, in the present embodiment, in order to simplify the description, a description will be given by taking as an example a light-receiving unit configured by n = 6 and m = 8 and consisting of vertical 6 pixels × horizontal 8 pixels = total 48 pixels. However, this numerical value does not limit the present embodiment.

The number of vertical CCDs (vertical charge transfer means) 2 is the same as the number of photoelectric conversion light receiving elements 1 in the horizontal direction (here, 8).
Individual vertical CCD sections (vertical charge transfer sections), and each vertical CCD section is arranged vertically adjacent to each vertical column of the photoelectric conversion light receiving element 1. The vertical CCD 2 sequentially sends out the electric charges transferred from the photoelectric conversion light receiving element 1 on the left side in the drawing pixel by pixel from an output end provided at one end. That is, the charges are transferred to the buffer 3 in units of one row by sending the charges from the eight vertical CCD units at the same time.

As shown in FIG. 2, the vertical CCD 2 is provided with the charges generated by the photoelectric conversion light receiving element 1 in the vertical CCD.
The charge transfer signal line 10 for inputting the charge transfer signal C ₁₀ for transferring the charge transfer signal to the inside 2 is connected. In addition, vertical CC
A vertical CCD drive signal line 11 for inputting a vertical CCD drive signal C ₁₁ for driving the vertical CCD 2 and moving charges downward is connected to D2.

The buffer 3 has a vertical C as shown in FIG.
It is connected to the output terminal of CD2 and temporarily holds the charge from the vertical CCD2. Moreover, by adjusting the gain of the buffer 3, it is possible to adjust the difference in sensitivity between the vertical transfer paths.

The switching element 4 for selection is, for example, FE
The drain is connected to the buffer 3, the source is connected to the output signal line 8, and the gate is connected to the output selection signal line 5. The selection switching element 4 selects only the necessary charges in the buffer 3 based on the output selection signal input to the output selection signal line 5 and guides it to the output signal line 8.

The sweep switching element 6 is, for example, FE
The drain is connected to the buffer 3, the source is connected to the sweep signal line 9, and the gate is connected to the sweep selection signal line 7. The sweep signal line 9 is grounded. The sweep switching element 6 selects the electric charge not selected by the selection switching element 4 based on the sweep selection signal input to the sweep selection signal line 7 and guides it to the sweep signal line 9.

The selective transfer means according to claim 1 is
Buffer 3, selection switching element 4, output selection signal line 5, sweep switching element 6, sweep selection signal line 7,
It corresponds to the output signal line 8 and the sweep signal line 9.

The operation of the image sensor having the above structure is
As shown in FIG. 2, a case will be described as an example in which only the effective area 12 formed by 9 pixels of vertical 3 pixels × horizontal 3 pixels in the central portion out of all 48 pixels is output. When performing this operation, a drive clock as shown in FIG. 3 is generated.

In this description, the selection switching element 4 will be described as 4a to 4h from the right in the figure, and the sweep switching element 6 will be described as 6a to 6h.
Further, C _4a to C _4h in FIG. 3 represent output selection signals applied to the switching elements 4a to 4h, and C _6a to C _6h are sweep selection signals applied to the switching elements 6a to 6h. Is represented. Also,
C ₈ indicates a signal output from the output signal line 8.

The photoelectric conversion light receiving element 1 generates charges according to the amount of light applied (see FIG. 5). When the charge transfer signal C ₁₀ shown in FIG. 3 becomes H level, the charges of all 48 pixels are transferred to the vertical CCD 2. It is transferred (see FIG. 6). The transferred charges are transferred to the buffer 3 from the bottom one row when the vertical CCD drive signal C ₁₁ becomes H level. And vertical CC
At the same time that the D drive signal C _{11 is set} to L level, the sweep switching elements 6a to 6h are turned ON. That is,
Since the necessary pixels do not exist in the first row at the bottom, all the switching elements for selection 4a to 4h are turned off,
The sweep switching elements 6a to 6h are simultaneously turned on to sweep all charges at once (see FIG. 7). Next 1
Rows are similarly processed (see FIG. 8).

There are three required pixels in the third to fifth rows from the bottom. Therefore, in this case, the sweep switching elements 6a, 6b, 6f through 6
A total of 5 pixels are swept away by turning on h, and the switching elements for selection 4c to 4e are sequentially turned on one by one to output three pixels one by one (see FIGS. 9 to 11). The last row sweeps out all the pixels like the first row (Fig. 1
2).

In this way, the charges to be transferred are selected by the switching elements 4a to 4h for selection and sequentially output from the output signal line 8, while the charges of unnecessary pixels are swept by the switching elements 6a to 6h for sweeping. It is output to the output signal line 9.

Incidentally, as shown in FIGS. 7 and 12, when all the charges are discharged at one time, the current flowing through the sweep signal line 9 may become large. Therefore, first, the sweep switching elements 6a to 6a. A configuration may be adopted in which only 6d is turned on and half is swept out (see FIG. 13A), and then the sweeping switching elements 6e to 6h are turned on and the other half is swept out (see FIG. 13 (b)). The drive clock in this case is as shown in FIG.

The method for generating the driving clock as shown in FIG. 3 can be easily performed by using the algorithm shown in FIG. First, for comparison, the operation of the conventional image sensor will be described.

The conventional CCD image sensor shown in FIG. 36 shows a flow chart for transferring an image of vertical 6 pixels × horizontal 8 pixels as shown in FIG.

In step (hereinafter abbreviated as S) 101, electric charges are generated according to the amount of light received by the photoelectric conversion light receiving element 1. This is transferred to the vertical CCD 2 and 1 is assigned to the variable I of the vertical counter that counts the number of rows from the bottom (S102). After that, the vertical CCD 2 is driven, charges representing the pixels on the I-th line from the bottom are transferred to the horizontal CCD 15 (S103), and 1 is substituted for the variable J of the horizontal counter that counts the number of columns from the right (S104). And horizontal C
The CD 15 is driven, and information for one pixel is output from the output signal line 16 (S105). Add 1 to variable J (S10
6), if the variable J does not exceed 8 in S107, S10
Return to 5 and repeat the process. If variable J exceeds 8, S
Proceed to 108. That is, the output is repeated until all the image information of one horizontal line (8 pixels) transferred to the horizontal CCD 15 is output.

At S108, 1 is added to the variable I, and S10
If the variable I does not exceed 6 at 9, the process returns to S103, and if the variable I exceeds 6, the process ends. That is, all the pixels are output by repeating the processing of S103 to S109 for the vertical lines. After all, regardless of the number of pixels required, the number of horizontal pixels x the number of vertical pixels (8 x 8 in this case)
6 = 48 times).

Next, in the conventional image sensor shown in FIG. 38, a flow chart in the case of carrying out partial transfer by thinning out an image of vertical 6 pixels × horizontal 8 pixels in units of horizontal 1 line,
As shown in.

S201 to S203 are the above S101 to S
Similarly to 103, when the generated charges are transferred to the vertical CCD 2 and 1 is substituted for the variable I representing the vertical count number, the vertical CCD 2 operates and the charges representing the pixels on the I line from the bottom are transferred to the horizontal CCD 15. To be done. Here, since the variable I indicates which pixel of the horizontal line from the bottom has been transferred to the horizontal CCD 15, the variable I is 3 in S204.
If the range is from 5 to 5 (2 <I <6), S205
Otherwise, the process branches to S211.

In S211, the sweep signal is turned on, and the charges of all the one lines transferred from the sweep signal line 105 to the horizontal CCD 15 are swept together. At this time, the sweep signal is turned off while keeping this state for a sufficient time for sweeping, and the process proceeds to S209.

On the other hand, when the variable I is in the range of 3 to 5 in S204, the processes of S205 to S208 are performed, which is the same as the above-described processes of S104 to S107. That is, the output is repeated until all the image information of one horizontal line (8 pixels) transferred to the horizontal CCD 15 is output.

In S209, 1 is added to the variable I, and S21
If it is 0 and the variable I does not exceed 6, the process returns to S203, and if the variable I exceeds 6, the process ends. That is, by repeating the processing of S203 to S210 for vertical lines, only horizontal lines in which necessary pixels are present are output.
After all, since all the pixels of the 3rd, 4th, and 5th horizontal lines from the bottom, which have the required number of pixels, are output, 3
× 8 (number of horizontal pixels) = 24 times of output. When changing the output range, the numerical value used for the comparison with the variable I in S204 is changed. For example, the effective pixels are 2 from the bottom.
If it is included in the 4th to 4th lines, the comparison expression in S204 may be changed to "1 <I <5".

The partial transfer operation of the image sensor of FIG. 2 in the present embodiment with respect to these conventional examples is as shown in the flowchart of FIG.

Electric charges are generated in the photoelectric conversion light receiving element 1 according to the amount of received light. This is transferred to the vertical CCD 2 (S1), and 1 is assigned to the variable I of the vertical counter that counts the number of rows from the bottom (S2). After that, the vertical CCD 2 is driven, and the charges representing the pixels on the I line from the bottom are transferred to the buffer 3 (S3).

Since the variable I indicates which pixel of the horizontal line from the bottom is transferred to the buffer 3,
In S4, if the variable I is in the range of 3 to 5, the process proceeds to S5, and if not, the process branches to S13. S
In 13, the charges are swept out from the sweep signal line 9 by turning on all the sweep switching elements 6. At this time, all the switching elements 6 for sweeping are turned off while maintaining this state for a sufficient time for sweeping out, and S1
The processing shifts to 1.

On the other hand, in S5, first, the sweeping switching elements 6a, 6b, 6f, 6 corresponding to the 1st, 2nd, 6th, 7th, and 8th pixels from the right, which are the pixels to be swept away.
Only g and 6h are turned on, and the other sweeping switching elements 6c to 6e are kept off (see FIG. 9). Next, 3 is substituted for the variable J of the horizontal counter that counts the number of columns from the right (S6), only the Jth switch from the right of the switching elements 4 for selection is turned on, and output is performed from the output signal line 8 ( S7). After a certain period of time, the switching element 4 for selection is turned off, and S7 ends. 1 is added to the value of the variable J of the horizontal counter (S8), and the variable J is set to 5 in S9.
If it does not exceed S7 and S8, the variable J becomes 5
If it exceeds, proceed to S10. In S10, all the sweep switching elements 6 are turned off. That is, S5
Through the processing from S10 to S10, only the 3rd, 4th, and 5th pixels from the right, which are effective pixels in one horizontal line, are output.

As described above, since only the necessary portion of one horizontal line is output, 3 pixels are required to output 3 pixels.
You need to output only once. This is because, as in the case of the conventional example using the algorithm of FIG. 15, all the lines including effective pixels are output only a small number of times as compared with the case of outputting all (eight horizontal lines output eight times). It is possible to operate at a higher speed than the conventional one.

After the processing of S10 or S13, 1 is added to the variable I of the vertical counter (S11). If the variable I does not exceed 6 in S12, the process returns to S3. If the variable I exceeds 6, the processing is performed. finish.

Through the above processing, among the imaged pixels,
The number of vertical lines including valid pixels is S4 to S4.
By repeating the process of 11 and performing the processes of S7 to S9 only for effective pixels among them, only necessary pixels are output. In this case, only the required 3 vertical pixels × 3 horizontal pixels = 9 pixels are output. This is much smaller than the output of the above-mentioned conventional example of 48 pixels or 24 pixels, and it can be seen that the speed is increased.

When changing the output range, it is necessary to change the numerical value used for comparison with the variable I / J in S4 or S9, the pixel to be swept out in S5, and the value assigned to the variable J in S6. Good. For example, if valid pixels are included in the second to fourth lines from the bottom and the second and third lines from the right, S4
To the comparison expression of "1 <I <5" and the comparison expression of S9 to "J>"
3 ”, the switching element 6 for sweep of S5 is set to“ 6a · 6
d.6e.6f.6g.6h ", and the value assigned to the variable J in S6 may be changed to" 2 ".

Considering the time required for the operation up to the time when the vertical CCD 2 is driven, from FIG. 5 to FIG.
The number of clocks required to output an image for one frame is 19 clocks. However, when all the conventional pixels are output, the remaining sweep-out is 48-9 = 39.
Since pixels also have to be transferred, another 39 clocks are required. At this time, there is no need for the three clocks for sweeping out the horizontal three lines as in the present embodiment.
After all, 19 + 39-3 = 55 clocks are required. Further, even when sweeping out in units of horizontal lines, since all the images of the 3 to 5 lines from the bottom are transferred, 5 pixels are additionally transferred per line, and 19 + 3 × 5 =
34 clocks are required. Since the actual number of pixels of the image sensor is much larger, the larger the difference between the required number of pixels and the total number of pixels is, the faster the partial output speed is as compared to the case of the whole output.

Further, in the above embodiment, the effective area 12
Although a rectangular area or a square area has been described, it is possible to output an area of any shape by using the data table shown in the following sweep data table and output data table and the algorithm shown in FIG.

Table 1 shows the sweep data,
The line number indicates the horizontal line number from the bottom. In Table 1, the values of row numbers 1, 2, and 6 are all "1", indicating that all pixels are swept out. Line number is 3
The values of Nos. 4 and 5 indicate that pixels are not swept out and the rest are swept out because the 3rd, 4th, and 5th from the right are “0”. The area of "0" indicates the range of the effective area 12 shown in FIG.

[0067]

[Table 1]

Table 2 shows the data to be output. The row number corresponds to the row number in Table 1. The column numbers sequentially indicate output positions, and the values of row numbers 1, 2, and 6 are all “0”, indicating that all pixels are not output. In addition, the values with the 3rd, 4th, and 5th line numbers are "3
"4.5", which is 34.5 from the right of the 34.5th horizontal line from the bottom in FIG.
It indicates that the th pixel is output in this order. That is,
Numerical values other than "0" are also shown in the order of outputting the range of the effective area 12 in FIG.

[0069]

[Table 2]

FIG. 17 uses Table 1 and Table 2 below.
The algorithm for outputting an arbitrary area is shown.

In S21 to S23, the charges generated in the photoelectric conversion light receiving element 1 are transferred to the vertical CCD 2 and 1 is substituted into the variable I, as in S1 to S3 described above. After that, the vertical CCD 2 operates to transfer the charges representing the pixels on the I-th line from the bottom to the buffer 3.

Next, the data whose row number is I-th in the sweep data table of Table 1 is input to the variable B (S24). Variable B
The data input to is output to the sweep selection signal line 7 to drive the sweep switching element 6 (S25). Set the first timer to count the time to hold the output (S2
6) Then, 1 is assigned to the variable J of the horizontal counter (S27), and the data whose column number in the output table of Table 2 is J is assigned to the variable C (S28). That is, the data in the I-th row and the J-th column of the output table is substituted into the variable C. In S29, if the value assigned to the variable C is 0, the process proceeds to S30, and if the value of the variable C is not 0, the process proceeds to S34. That is, when the value of the variable C is 0, the process proceeds to the next line. Therefore, the numerical value to the right of "0" in Table 2 is invalid because it is not referenced.

Here, if the value of the variable C is not 0, it means that there is a pixel to be output, so only the C-th output from the right of the selection switching element 4 is turned on (S
34), set a second timer that counts the time for holding the output (S35), and waits for the second timer to elapse (S36). After the elapse of the second timer, the selection output of the switching element 4 for selection which has been turned on is turned off (S3
7), 1 is added to the variable J of the horizontal counter (S38), and if this value exceeds 8, the process proceeds to S30, and if it does not exceed 8, the process returns to S28 and repeats the processing.

In S30, waiting for the first timer to elapse, the data to be swept out is sufficiently swept, and thereafter, all the sweep outputs of the sweep switching element 6 are turned off (S31). Add 1 to the variable I of the vertical counter (S3
2) If the variable I does not exceed 6 in S33, the process returns to S23, and if the variable I exceeds 6, the processing of all 6 horizontal lines is completed.

The first and second timers are for counting the time for sufficiently sweeping out the data. The first timer counts time independently,
It shall have a function to judge whether the set time has elapsed. Note that such a timer function is general and will not be described in detail.

By the above operation, high speed driving which outputs only necessary pixels can be performed. When the data of Tables 1 and 2 and the algorithm of FIG. 17 are used for the effective area 12, the charges move as shown in FIGS.
It produces the same result as the execution of the algorithm of 6. However, the algorithm of FIG. 17 can output not only a simple rectangular area like the algorithm of FIG. 16 but also areas of any shape at high speed.

For example, even when information on only the image area shown by the dark portion shown in FIG. 18 is required,
If the sweep data shown in Table 3 and the output data shown in Table 4 are used to execute the algorithm of FIG. 17, partial output can be performed at high speed. It should be noted that in the speed-up method by selection in units of horizontal lines shown in FIG.
When all the horizontal lines include the necessary pixels as in 8, the speed cannot be increased at all.

[0078]

[Table 3]

[0079]

[Table 4]

Considering the operation in this case, the areas of FIG. 18 are [2.3] and [3.
4], [4.5], [5.6], [6.7], [7.
8], information for every two pixels is required. Since each part other than the necessary part is swept out, the swept data in Table 3 is "0" for the necessary part and "1" for the other parts,
"11111001, 11110011, 111001
11, 11001111, 10011111, 0011
1111 ". Further, in the output data of Table 4, necessary pixels may be arranged as they are, and “2, 3, 3 ,.
4,4,5,5,6,6,7,7.8 ".

Using the algorithm of FIG. 17 for these data, <sweep (11111001), output (2), output (3), sweep (1111001)
1), output (3), output (4), sweep (11100
111), output (4), output (5), sweep (110)
01111), output (5), output (6), sweep (1
0011111), output (6), output (7), sweep out (00111111), output (7), output (8)>. The number of outputs in this case is only 12 which is the required number of pixels. From the above, it can be seen that even if the required pixels are included in all the horizontal lines as shown in FIG. 18, only the required pixels can be output.

Next, consider the case as shown in FIG. This shows an output pixel in the case of performing an output in which a region of 2 pixels in the vertical direction × 2 pixels in the horizontal direction is represented by one pixel, which means that an image reduced to ½ by the representative of the pixels is output. To do. Even in such a case, if the sweep data shown in Table 5 and the output data shown in Table 6 are used to execute the algorithm of FIG. 17, partial output can be performed at high speed.

[0083]

[Table 5]

[0084]

[Table 6]

Considering the operation in this case, the areas in FIG. 19 are respectively [none] and [2.4.6] from the lower horizontal line.
・ 8], [None], [2,4,6,8], [None],
Since the information of [2, 4, 6, 8] is required, the sweep data of Table 5 is "11111111, 1010101.
0, 11111111, 10101010, 11111
111, 10101010 ". In addition, the output data of Table 6 is “none, 2/4/6/8, none, 2/4/6
・ 8, none, 2, 4, 4, 6, 8 ".

Using the algorithm of FIG. 17 for these data, <sweep (11111111), sweep (10101010), output (2), output (4), output (6), output (8), sweep ( 1111
1111), sweep (10101010), output (2), output (4), output (6), output (8), sweep (11111111), sweep (1010101)
0), output (2), output (4), output (6), output (8)> in this order. The number of outputs in this case is only 12 which is the required number of pixels. From the above, it can be seen that only the necessary pixels can be output even when the pixels shown in FIG. 19 are not continuous.

Next, consider the case shown in FIG. This only outputs an area of vertical 6 pixels × horizontal 4 pixels in the central portion. In this case, if the sweep-out data shown in Table 7 and the output data shown in Table 8 are used to execute the algorithm of FIG. 17, it is possible to output a horizontally inverted image at high speed.

[0088]

[Table 7]

[0089]

[Table 8]

Considering the operation in this case, since the information of [3, 4, 5, 6] is required for all the areas in FIG. 20, all the sweep data in Table 7 is "11000011". Also, the output data in Table 8 are all "3, 4, 5,
It is supposed to be 6 ", but in order to reverse left and right, it is set to" 6, 5, 4, 3 "in reverse order.

Using the algorithm of FIG. 17 for these data, <sweep (11000011), output (6), output (5), output (4), output (3), sweep (11000011), output ( 6), output (5),
Output (4), output (3), sweep (1100001
1), output (6), output (5), output (4), output (3), sweep (11000011), output (6),
Output (5), Output (4), Output (3), Sweep (11
000011), output (6), output (5), output (4), output (3), sweep (11000011),
The operation is performed in the order of output (6), output (5), output (4), output (3)>. Here, the output order of horizontal pixels is 3.4.
Since it is not 5/6 but the reverse 6/5/4/3, it can be seen that a horizontally reversed image is output. Further, the number of outputs in this case is only 24, which is the required number of pixels. As a result, it can be understood that even when the output order is changed such as left-right inversion in the horizontal line, only the necessary pixels can be output.

As described above, since the image sensor according to the present embodiment can select only the necessary pixels from the pixels for one horizontal line and output the partial image,
High-speed transfer is possible for the time required to transfer the unnecessary pixels.
As a result, the processing speed of the image sensor can be improved.

Further, since this image sensor thins out in pixel units, even if there is an irregular area in which the output portion is different in horizontal line units or there are a plurality of target areas, all image information is output. A sufficiently high-speed transfer is possible. Further, even when compared with the conventional image sensor that performs thinning in units of lines, it is possible to output at high speed regardless of the shape of the target area, especially when the number of pixels in the target area is small.

Further, the pixel selection by the selective transfer means is set to 1
1 by setting the number of pixels to be transferred every other half
It is also possible to output an image reduced to / 2. Further, it is also possible to simultaneously select a plurality of pixels and output an average value, that is, to simultaneously select n adjacent pixels to extract an image compressed to 1 / n in the horizontal direction.

Embodiment 2 The following will describe Embodiment 2 of the present invention with reference to FIGS. 21 and 22. For the sake of convenience of explanation, the same members as those shown in the drawings of the above-described embodiment are designated by the same reference numerals, and the description thereof will be omitted.

The image sensor of this embodiment has the configuration shown in FIG.
As shown in FIG. 3, a bidirectional vertical CCD 13 is provided instead of the vertical CCD 2 of the first embodiment, and the horizontal CCD 1
5 and a vertical CCD drive signal line 14 are added, and other configurations are the same as those in the first embodiment.

The bidirectional vertical CCD (bidirectional vertical charge transfer means) 13 is capable of moving charges up and down bidirectionally.
It is a CD, and is composed of the same number of vertical CCD units as the photoelectric conversion light receiving elements 1 (here, eight). Each vertical C
The CD unit is arranged vertically adjacent to each one vertical column of the photoelectric conversion light receiving element 1. Each vertical CCD section has an output end at both ends in the longitudinal direction.
A horizontal CCD 15 is connected to the output end of the lower part of D13,
Selective transfer means is connected to the upper output terminal.

The selective transfer means comprises a buffer 3, a selection switching element 4, an output selection signal line 5, a sweep switching element 6, a sweep selection signal line 7, and an output signal line 8. Also, the selective transfer means and the horizontal CCD 1
It goes without saying that the upper and lower sides of 5 may be reversed.

A charge transfer signal line 10 for inputting a charge transfer signal C ₁₀ for transferring the charges generated by the photoelectric conversion light receiving element 1 into the bidirectional vertical CCD 13 is connected to the bidirectional vertical CCD 13. In addition, bidirectional vertical CCD
A vertical CCD drive signal line 14 for inputting a vertical CCD drive signal C ₁₄ for controlling the transfer of electric charges in the upward direction is connected to 13 and a vertical CCD for controlling the transfer of electric charges in the downward direction. The vertical CCD drive signal 11 for inputting the CCD drive signal C ₁₁ is connected.

The horizontal CCD (horizontal charge transfer means) 15 is
The input charges for one line are sequentially output from the output signal line 16 pixel by pixel. Each time the horizontal CCD 15 finishes outputting the charge for one line from the output signal line 16, the bidirectional vertical C
One line is transferred from the CD 13, and all the charges are transferred, so that information for one screen is output. In addition,
The horizontal CCD 15 is driven by the horizontal CCD drive signal C ₂₁ input to the horizontal CCD drive signal line 21.

The drive signal described in claim 3 is the vertical C
It corresponds to the CD drive signal C ₁₁ and the vertical CCD drive signal C ₁₄ .

According to the above configuration, the charges generated by the photoelectric conversion light receiving element 1 are transferred to the adjacent vertical CCD section. In the bidirectional vertical CCD 13, the charges are transferred to the horizontal CCD 15 in units of one row or the vertical CCD is transferred upward by the vertical CCD drive signal C ₁₁ transferred downward.
The drive signal C ₁₄ transfers the charges to the selective transfer means in units of one row.

That is, the image sensor of the present embodiment does not use the vertical CCD drive signal line 14, but the bidirectional vertical CCD 13, the vertical CCD drive signal line 11, and the horizontal C line.
When a normal full screen output is obtained from the output signal line 16 at the CD 15 at a normal speed and when the vertical CCD drive signal line 11 is not used, the bidirectional vertical CCD 13, the vertical CCD drive signal line 14, and the selective transfer means. Thus, the image sensor has two types of output systems, that is, a case where an output in an arbitrary region is obtained at high speed from the output signal line 8.

Therefore, since the information sent to either output system is obtained by the same light-receiving surface, the difference in pixel shift and the enlargement ratio (area expressed by one pixel) between both outputs are basically the same. Does not exist. Therefore, no matter which output system the image is obtained, the same result can be obtained even if the area or the like is measured by image processing or the like.

As a result, as compared with the case where another optical system or image sensor is prepared, a complicated calculation for obtaining a correlation between two optical systems or image sensors and a correlation between two image outputs is required. You don't have to. Further, there is no need to worry about the deviation between the two after adjustment due to temperature or vibration. That is, it is possible to easily and highly accurately obtain images having different aspect ratios and transfer cycles. Further, since only one image sensor is required, it is possible to reduce cost, installation space, and installation work.

The area to be transferred is the same when the entire area is selectively output by the upper selective transfer means and when the entire area is transferred by the lower horizontal CCD 15, but is transferred from each output system. There are differences such as the order of pixels being different. However, having two output systems separately than that has the following major features.

That is, since obtaining an output from the horizontal CCD 15 is exactly the same as obtaining an image with a normal CCD image pickup device, the NTSC (National Televis
It is possible to specialize in outputting signals at a timing that conforms to standards such as the Ion System Committee). On the other hand, the image output in an arbitrary range by the partial read operation using the selective transfer means is not transferred in a fixed form such as NTSC, but at a timing and speed convenient for transfer to the connected image processing apparatus. It is possible to freely design a specialized configuration or a specialized configuration for speeding up.

Even if it is designed freely as described above, the performance of each of the two outputs is not deteriorated, so that it is possible to realize a multi-function that can be freely selected and used according to each purpose.
In the present embodiment, a normal screen is obtained without affecting the high-speed operation of partial reading using the bidirectional vertical CCD 13, the vertical CCD drive signal line 14, and the selective transfer means whose purpose is to increase the speed. be able to.

In FIG. 21, the two output systems target all pixels, but the present invention is not limited to this. For example, as shown in FIG. 22, the number of pixels of the light receiving portion (the number of photoelectric conversion light receiving elements) is 6 pixels in the vertical direction × 16 pixels in the horizontal direction, and all pixels of the light receiving portion connected to the upper portion of the bidirectional vertical CCD 13 are targeted. A horizontal CCD 26 having a width of 16 pixels and a selective transfer means connected to the lower part of the bidirectional vertical CCD 13 for the central vertical 6 pixels × horizontal 8 pixels as a high-speed partial reading unit and the left / right vertical 6 pixels × horizontal 4 pixels It may be configured to include charge sweeping units 29 and 30 for sweeping charges from each of the two portions. The horizontal CCD 26 is driven by the horizontal CCD drive signal C ₂₈ input to the horizontal CCD drive signal line 28. Such an image sensor can be used when the area where high-speed partial transfer is to be performed is limited to the central portion.

According to this, when the charges in the bidirectional vertical CCD 13 are moved upward by the vertical CCD drive signal C ₁₄ and input to the horizontal CCD 26, all the charges are output from the output signal line 27 of the horizontal CCD 26. It On the other hand, the charges in the bidirectional vertical CCD 13 are moved downward by the vertical CCD drive signal C ₁₁ and input to the selective transfer means and the charge sweeping units 29 and 30. Then, as for the electric charges for the central 6 vertical pixels × 8 horizontal pixels input to the selective transfer means, only the necessary electric charges are selectively output from the output signal line 8.
The charges of 6 pixels in the vertical direction and 4 pixels in the horizontal direction, which are input to the charge sweeping units 29 and 30, are swept away.

As a result, similar to the image sensor having the structure shown in FIG. 21, an image sensor having two types of output systems for normal speed transfer and high speed transfer can be obtained. In this case, by limiting the number of pixels in the horizontal direction targeted by the high-speed partial reading unit of the lower selective transfer unit, compared to the case where the circuit configuration of the high-speed reading unit is adapted to correspond to all pixels in the horizontal direction. The circuit is simplified, which is advantageous in terms of cost.

In this embodiment, the bidirectional vertical CC
The horizontal CCD 15 or the horizontal CCD 26 is connected to D13, but the electric charge at the end of the bidirectional vertical CCD 13 does not flow into these horizontal CCDs or vice versa.

In this embodiment, bidirectional vertical CC
Although the electric charges are moved to the two output systems by using D13, it is also possible to adopt a configuration in which the two output systems are switched by using the unidirectional vertical CCD and the switch circuit. However, in this case, since the configuration is complicated because the switch circuit is provided, the configuration using the bidirectional vertical CCD 13 is preferable.

In the first and second embodiments, the output of the vertical CCD 2 or the bidirectional vertical CCD 13 is transferred via the buffer 3. However, the output of the vertical CCD 2 is switched to the selection switching element 4 and the sweep switching. It may be directly connected to the element 6.

Reference Embodiment 1 Reference embodiment 1 will be described below with reference to FIGS. 23 to 25. For the sake of convenience of explanation, the same members as those shown in the drawings of the above-described embodiment are designated by the same reference numerals, and the description thereof will be omitted.

In the image sensor of the present embodiment , as shown in FIG. 23, the number of pixels of the light receiving portion is 6 vertical pixels × 16 horizontal pixels. Therefore, the number of vertical CCD units in the bidirectional vertical CCD 13 is 16. A vertical CCD (charge storage transfer means) 24 is connected to the upper part of the bidirectional vertical CCD 13 via a connecting means 23, and a horizontal CCD 26 is further connected to the vertical CCD 24. In the lower part of the bidirectional vertical CCD 13, the horizontal CCD 15 is connected to the central eight vertical CCD units, and the charge scavenging units 29 and 30 are connected to the left and right four vertical CCD units, respectively.

A vertical CCD drive signal line 2 for inputting a vertical CCD drive signal C ₂₅ for driving the vertical CCD 24 to move electric charges upward in the solid is input to the vertical CCD _24.
5 is connected. Then, the vertical CCD drive signal C ₂₅ becomes O
In the FF state, the electric charge does not move and is held in the solid, and when it is in the ON state, it starts moving. That is, the vertical CCD 24 temporarily stores image information.

The horizontal CCD 26 corresponds to a width of 16 pixels for all the pixels of the light receiving portion, and the vertical CCD 2
The charges of one line from 4 are sequentially output from the output signal line 27 pixel by pixel. This horizontal CCD 26 is a horizontal CCD
It is driven by the horizontal CCD drive signal C ₂₈ input to the drive signal line 28.

The connecting means 23 inputs the output from the bidirectional vertical CCD 13 into the vertical CCD 24 while the electric charge is stored in the vertical CCD 24, and outputs the output from the bidirectional vertical CCD 13 after the electric charge is stored in the vertical CCD 24. Is horizontal C
Connection control is performed so that the data is input to the CD 15 and the charge sweeping units 29 and 30.

The connecting means 23 may be composed of a switching element such as an FET. The vertical CCD 24 itself may be provided with the function of the connecting means 23.
This is because the CCD is an element that moves electric charges according to changes in the electric potential of the element, and therefore, by keeping a certain point at a high potential, it is possible to prevent the electric charges from moving to that point and beyond.

Further, the image sensor includes a shutter 3 for controlling the amount of light applied to the photoelectric conversion light receiving element 1.
1 is provided. This shutter 31 has a shutter drive signal C connected to a shutter drive signal line 32 connected to it.
Entering ₃₂ controls its operation.

The drive signal is the vertical CCD drive signal C.
It corresponds to ₁₁・ C ₁₄・ C ₂₅ .

According to the above configuration, the bidirectional vertical CCD 13
The charge transferred to is the vertical CCD drive signal C ₁₄ to the upper is transferred row by row, this time, the vertical by operating also the vertical CCD drive signal C ₂₅ in synchronism with CCD24
Transfer charge to. By performing this processing only for the number of vertical pixels, all the charges of the bidirectional vertical CCD 13 are
Transfer to 24.

After that, when the connecting means 23 connecting the upper and lower vertical CCDs is cut by an electrical or mechanical means, the portions above and below the connecting means 23 become completely independent operations. At this time, the structure above the connecting means 23 has a structure in which the photoelectric conversion light receiving element 1 is removed from the image sensor. However, since charges have already been transferred and stored, a case where a normal image sensor transfers one frame image An image of one frame can be output from the output signal line 27 by the same operation.

On the other hand, the portion below the connecting means 23 constitutes one image sensor including the photoelectric conversion light receiving element 1. Then, the newly received image is transferred to the lower portion in units of rows by the vertical CCD drive signal C ₁₁ , the charges are swept away by the charge sweeping units 29 and 30, and the horizontal CCD 15 operates in the same manner as an ordinary image sensor. Then, an image of one frame is output from the output signal line 16.

Based on the timing chart of FIG.
The operation of the parallel image sensor will be described. The operation clock of each vertical CCD is slower than the operation of each horizontal CCD. This is because one horizontal CCD operates independently at a relatively high frequency, but a vertical CCD requires a plurality of vertical CCDs to operate in synchronization with each other. It is difficult to drive at high speed.

When the shutter drive signal C ₃₂ becomes H level, the shutter 31 operates for the first time and the photoelectric conversion light receiving element 1 is irradiated with light. When the converted charges are transferred to the bidirectional vertical CCD 13 under the control of the charge transfer signal C ₁₀ , the connecting means 23 is turned on (indicated by C ₂₃ in FIG. 24). At this time, the vertical CCD driving signal lines 14 and the vertical CCD driving signal lines 25 are synchronized with each other.
By supplying the drive signal C ₁₄ and the vertical CCD drive signal C ₂₅ , all the charges are transferred to the vertical CCD 24. During this time, the shutter 31 performs the second operation in parallel,
New charges are generated in the photoelectric conversion light receiving element 1.

The connection means 23 is turned off in synchronization with the second shutter drive signal C ₃₂ becoming L level, and the operation independent of the constituent elements arranged below the connection means 23 is possible. . Therefore, the charges transferred to the vertical CCD 24 are interlocked with the horizontal CCD 26 to sequentially transfer one frame of an image composed of vertical 6 pixels × horizontal 16 pixels from the output signal line ₂₇ as the output signal C _27. be able to. During this time, the newly generated charges are transferred to the bidirectional vertical CCD 13, and the vertical CCD drive signal C ₁₁ causes the horizontal CCD 15 and the charge scavenging unit 2 located below the light receiving unit.
It is sequentially transferred to 9:30. The horizontal CCD 15 and the bidirectional vertical CCD 13 are interlocked with each other, so that the central vertical 6 pixels ×
8 horizontal output signal is the image of one frame consists of pixels C ₁₆
Be transferred as.

Further, the shutter 31 performs the third operation to generate electric charges in the photoelectric conversion light receiving element 1, and similarly, the horizontal CCD 15 and the bidirectional vertical CCD 13 are interlocked with each other, and the central vertical 6 pixels × horizontal 8 pixels An image of another frame composed of is output from the output signal line 16.

While the output signal C ₁₆ is being output from the output signal line 16, the shutter 31 is linked again to generate charges to be transferred next to the upper vertical CCD 24, and the above operation is repeated.

FIG. 25 shows only the shutter drive signal C ₃₂ , the charge transfer signal C ₁₀ , the output signal C ₁₆ , and the output signal C _{27 in} the timing chart of FIG. 24 with the time axis enlarged. . From this, it can be seen that the image is output to the output signal line 16 at the cycle T and the image to the output signal line 27 at the cycle 2 × T (= 2T).

As described above, the image sensor according to the present embodiment uses two vertical CCDs 1 by using the vertical CCD 24 capable of temporarily holding charges.
Images can be output in parallel from 5.26.

Therefore, since the images output in parallel are images taken by the common light receiving section, the common pixels between the outputs are output without any deviation. Therefore, even if each output obtained here is processed by using a different image processing device, there is no displacement in area or position. Therefore, as in the case of using two image sensors, there is no need for camera alignment, magnification, etc. It eliminates the need for physical adjustments and complicated correlation calculations. That is, even in parallel processing in a plurality of image processing devices having different numbers of pixels to be handled, a single image sensor can output an accurate and efficient parallel image with no positional deviation or area difference. As a result, compared to the case where two image sensors are used, processing such as adjustment is not required, so that high-speed processing can be performed.

Further, the two horizontal CCDs in this image sensor can output periodic images of different regions. At this time, it is possible to output in various cycles by using a horizontal CCD having an aspect ratio other than this reference embodiment or by changing the vertical and horizontal pixel ratios or the number of pixels.

Further, an operation such as an asynchronous operation can be realized within a range in which the operation of one horizontal CCD does not hinder the operation of the other horizontal CCD. For example, above
In the reference mode , partial transfer of 6 pixels in the vertical direction × 8 pixels in the horizontal direction can be performed in a short time. However, it is not necessary to periodically perform partial transfer, and partial transfer can be freely operated as long as it does not hinder the whole transfer.

As a result, the partial transfer can be performed frame by frame with a small waiting time by operating the partial transfer only when necessary while periodically performing the entire transfer. That is, partial transfer can be performed at high speed while performing normal image transfer. As a result, it is possible to perform high-speed partial image processing at a fast cycle in parallel by performing small image transfer at a short cycle while performing image processing at a slow cycle in which a large area needs to be viewed at once. it can.

Reference Embodiment 2 Reference embodiment 2 will be described below with reference to FIGS. 26 to 29. For convenience of explanation, the same members as those shown in the drawings of the above-described embodiments or reference embodiments are designated by the same reference numerals, and the description thereof will be omitted.

The image sensor of this reference embodiment is the same as the reference embodiment.
1 bidirectional vertical CCD13 instead of vertical CCD 2, ginseng
Like the first embodiment, the image sensor having two output systems is provided.

As shown in FIG. 26, a vertical CCD 37 is connected to the lower part of the vertical CCD (unidirectional vertical charge transfer means) 2 via the side a of the switching means 35, and a horizontal CCD 38 is further connected to the vertical CCD 37. It Also, the switching means 3
The horizontal CCD 15 is connected to correspond to the vertical 6 pixels × horizontal 8 pixels in the middle through the b side of 5, and the charge sweeping units 29 and 30 are connected to correspond to the horizontal 6 pixels × horizontal 4 pixels. To be done.

The switching means 35 controls whether the output of the vertical CCD 2 is transferred to the vertical CCD 37 or the horizontal CCD 15 and the charge sweeping units 29 and 30.

Like the vertical CCD 24 of the first embodiment , the vertical CCD 37 can hold the information of one frame which has been picked up, and the vertical CCD drive signal C ₃₆ input to the vertical CCD drive signal line ₃₆ makes the horizontal CCD ₃₇ horizontal. The charge is transferred to the CCD 38.

The horizontal CCD 38 transfers charges from the output signal line 39 under the control of the horizontal CCD drive signal C ₄₀ input to the horizontal CCD drive signal line 40.

Based on the timing chart of FIG. 27,
The operation of the image sensor will be described.

When the shutter drive signal C ₃₂ becomes H level, the shutter 31 operates and the photoelectric conversion light receiving element 1 is irradiated with light. When the converted charges are transferred to the vertical CCD 2 by the charge transfer signal C ₁₀ , the switching means 35 becomes a side and the vertical CCD 2 is connected to the vertical CCD 37 (FIG. 2).
7 is designated as C ₃₅ ). Vertical CCs synchronized with the vertical CCD drive signal line 11 and the vertical CCD drive signal line 36.
By supplying the D drive signal C ₁₁ and the vertical CCD drive signal C ₃₆ , all the charges are transferred to the vertical CCD ₃₇ .
During this time, the shutter 31 operates in parallel, and new charges are generated in the photoelectric conversion light receiving element 1.

In synchronism with the second shutter drive signal C _{32 going} to L level, the switching means 35 is turned to the b side, and the vertical CCD 2 has the horizontal CCD 15 and the charge sweeping section 29.3.
Connected to 0. On the other hand, the components on the side of the vertical CCD 37 cut by the switching means 35 are in a state where they can operate independently. Therefore, the electric charges previously transferred to the vertical CCD 37 are sequentially transferred as an output signal C ₃₉ from the output signal line 39 to one frame of an image composed of vertical 6 pixels × horizontal 16 pixels in cooperation with the horizontal CCD 38. It During this time, the newly generated charges are transferred to the vertical CCD 2 and are transferred to the vertical CCD 2.
The horizontal CCD 15 and the charge sweeping unit 29 are driven by the drive signal C ₁₁ .
・ Sequentially transferred to 30. By the interlocking operation of the horizontal CCD 15 and the vertical CCD 2, an image of one frame composed of vertical 6 pixels × horizontal 8 pixels is output from the output signal line 16.

Further, the shutter 31 is operated again to generate charges in the photoelectric conversion light receiving element 1, and similarly, another frame image at the center is output from the output signal line 16.

While the output signal C ₁₆ is being output from the output signal line 16, the shutter 31 is re-engaged and the vertical C
The charges to be transferred next to the CD 37 are generated, and the above operation is repeated.

FIG. 28 is a timing chart showing only the shutter drive signal C ₃₂ , the charge transfer signal C ₁₀ , the output signal C ₁₆ , and the output signal C _{39 in} the timing chart of FIG. . It can be seen that the image is output to the output signal line 16 in the cycle T, and the output signal line 39 in the cycle 2T.

As described above, the image sensor according to this embodiment has a vertical C that can temporarily hold electric charges.
By using CD37, the same as in Reference Embodiment 1
Images can be output in parallel from the two horizontal CCDs 15 and 38.

Differences from the first embodiment are as follows. three
In Consideration 1 , the output of the image obtained from the vertical CCD 24 and the horizontal CCD 26 is sequentially transferred from the upper line of the light receiving portion, and the output of the image obtained from the bidirectional vertical CCD 13 and the horizontal CCD 15 is sequentially transferred from the lower line. Therefore, the transfer order of the lines is reversed at each output. On the other hand, in this reference embodiment , 2
Both outputs are transferred from the lower line.

Further, in the present embodiment , since the transfer destination is switched by the switching means 35 to the vertical CCD 37 or the horizontal CCD 15, a relatively complicated wiring including the switching means 35 is required. On the other hand, in the reference embodiment 1 , the connection means 23 that can be configured by a simple switch that is only connected or not connected to the upper portion of the bidirectional vertical CCD 13 is provided, and thus the configuration of the reference embodiment 1 is preferable.

The image sensor of this embodiment can also be configured as shown in FIG. In this configuration, the horizontal CCD 45, the vertical CCD 37, and the horizontal CCD 38 are connected in series in this order to the vertical CCD 2.
The driving of the horizontal CCD 45 is controlled by the horizontal CCD drive signal input to the horizontal CCD drive signal line 46. That is, the horizontal CCD drive signal is the vertical CCD2.
Charge from the output signal line 47 or a vertical CCD
It controls whether to transfer to 37.

According to this structure, the charges from the vertical CCD 2 are first transferred to the horizontal CCD 45. The transferred charges are transferred to the lower vertical CCD 37 when the horizontal CCD drive signal is OFF. The charges temporarily held by the vertical CCD 37 are transferred to the horizontal CCD 38 and output from the output signal line 39. On the other hand, when the horizontal CCD drive signal is in the ON state, the charge transferred from the vertical CCD 2 is output from the output signal line 47 of the horizontal CCD 45 without being transferred to the vertical CCD 37.

Also in this case, similarly to the image sensor of FIG. 26, the vertical CCD 37 and the horizontal CCD 38 can output in parallel while the vertical CCD 2 and the horizontal CCD 45 output while outputting.

However, because of the following reasons, FIG.
26 is more preferable than the image sensor of FIG. That is, in the case of the configuration of FIG. 26, since the charge transfer direction is switched by the switching means 35, the vertical CCD 2
The horizontal CC while transferring the electric charge of
Although D15 can be freely operated, in the case of the configuration of FIG. 29, since it is transferred to the vertical CCD 37 through the horizontal CCD 45, the horizontal CCD 45 cannot perform the output operation during this period. Therefore, from the configuration of FIG.
The image sensor configured as described above can perform parallel operation at higher speed.

Further, in the configuration of FIG. 29, since the horizontal CCD 45 which can also be transferred to the lower portion replaces the switching means 35, basically the vertical CCD 2, the horizontal CCD 45 and the vertical C
The number of pixels in the horizontal direction of the three CCDs and the CD 37 must match, but in the configuration of FIG. 26, the number of pixels in the horizontal direction is freely determined by providing the switching means 35 and the charge sweeping units 29 and 30. can do.

Reference Embodiment 3 Reference embodiment 3 will be described below with reference to FIGS. 30 to 32. For convenience of explanation, the same members as those shown in the drawings of the above-described embodiments or reference embodiments are designated by the same reference numerals, and the description thereof will be omitted.

The image sensor of this reference embodiment is the same as the reference embodiment.
The configuration of the reference form 2 is applied to the configuration of 1 . That is, as shown in FIG. 30, the vertical CCD 24 and the horizontal C are provided above the bidirectional vertical CCD 13 through the connecting means 23.
The CD 26 is connected in this order. On the other hand, bidirectional vertical CC
A vertical CC is provided under the D13 via the a side of the switching means 35.
The D37 and the horizontal CCD 38 are connected in this order, and the horizontal CCD 15 and the charge sweeping units 29 and 30 are connected via the side b. As a result, the image sensor operates as a 3-parallel image sensor whose output is output from the output signal lines 16, 27, 39.

Based on the timing chart of FIG. 31,
The operation of the image sensor will be described.

When the shutter drive signal C ₃₂ becomes H level, the shutter 31 operates and the photoelectric conversion light receiving element 1 is irradiated with light. When the converted charges are transferred to the bidirectional vertical CCD 13 by the charge transfer signal C ₁₀ , the switching means 35 is in an OFF state in which it is not connected to either the a or b side, and the portion below the bidirectional vertical CCD 13 is independent. On the other hand, the connection means 23 is turned on while the vertical CCD on the upper side
24 and the bidirectional vertical CCD 13 are brought into conduction. Then, by supplying the vertical CCD drive signal line 14 and the vertical CCD drive signal line 25 with the vertical CCD drive signal C ₁₄ and the vertical CCD drive signal C ₂₅ which are synchronized with each other, all the charges are transferred to the vertical CCD 24. During this time, the shutter 31 operates in parallel, and new charges are generated in the photoelectric conversion light receiving element 1.

In synchronization with the second shutter drive signal C ₃₂ becoming L level, the connecting means 23 is turned off, and the components above the vertical CCD 24 are the bidirectional vertical CCs.
Independent of D13. The electric charge temporarily held by the vertical CCD 24 is output by the output signal line 27 of the horizontal CCD 26.
It is transferred as an image for one frame composed of vertical 6 pixels × horizontal 16 pixels.

While the image is being output from the output signal line 27, newly generated charges are generated by the bidirectional vertical CCD 13
Transferred to. The switching means 35 becomes the side b in synchronization with the connection means 23 being turned off, and the bidirectional vertical CCD
Reference numeral 13 is connected to the horizontal CCD 15 and the charge sweeping units 29 and 30. Then, charges are transferred to the horizontal CCD 15 and the charge sweeping units 29 and 30 by the vertical CCD drive signal C _11, and an image of one frame composed of 6 vertical pixels × 8 horizontal pixels at the center is output from the output signal line 16. It

During this time, the shutter 31 operates, and the photoelectric conversion light receiving element 1 generates electric charges. Then, the switching means 35 is set to the side a, and the vertical CCD drive signals C ₁₁ and C ₃₆ operate in synchronization with each other, whereby the newly generated charges are transferred to the vertical CCD 37 as an image for one frame. During this time, the shutter 31 is also operated to generate charges in the photoelectric conversion light receiving element 1.

After the transfer to the vertical CCD 37 is completed, the switching means 35 becomes the side b again, and the output signal line 1 of the horizontal CCD 15 is output.
From 6, another one consisting of 6 vertical pixels x 8 horizontal pixels
The frame image is output. Further, during this time, the shutter 31 also operates to generate electric charges to be transferred to the vertical CCD 24, and the above operation is repeated.

FIG. 32 is a timing chart showing only the shutter drive signal C ₃₂ , the charge transfer signal C ₁₀ , the output signal C ₁₆ , the output signal C ₂₇ , and the output signal C _{39 in} the timing chart of FIG. It is a representation. It can be seen that the image is output to the output signal line 16 in the cycle T, and the output signal lines 27 and 39 in the cycle 2T.

As described above, the image sensor according to this embodiment can output three or more kinds of images in parallel with one image sensor at different periodic intervals. At this time, the book
Using a horizontal CCD with an aspect ratio other than the reference form ,
Depending on the vertical and horizontal pixel ratios, the number of pixels, etc., it is possible to output at various cycles.

Further, an operation such as an asynchronous operation can be realized within a range where one operation does not interfere with the other operation. For example, it is possible to perform an operation of outputting a specific portion of the image in units of one frame only when necessary, while constantly outputting all the images periodically. 31 and 32
In the example shown in the timing chart, vertical 6 pixels x horizontal 1
Output signal line 2 in which an image is output with a period of 2 pixels of 6 pixels
7.39 is operating with a period T offset. Therefore, by alternately outputting the outputs of the output signal lines 27 and 39, it is possible to output an image of vertical 6 pixels × horizontal 16 pixels and an image of vertical 6 pixels × horizontal 8 pixels at a cycle T. This is 1
It is shown that images with different aspect ratios can be taken from one light receiving surface even in the same cycle. As a result, in addition to the effects of the configurations of the first and second embodiments , one image sensor can simultaneously obtain both a high-definition image and a normal image.

[ Reference Mode 4 ] The reference mode 4 will be described below with reference to FIGS. 33 to 35. For convenience of explanation, the same members as those shown in the drawings of the above-described embodiments or reference embodiments are designated by the same reference numerals, and the description thereof will be omitted.

As shown in FIG. 33, the image sensor of the present embodiment has a configuration in which selective transfer means is arranged instead of the horizontal CCD 15 and the charge sweeping units 29 and 30 of the third embodiment . That is, the selective transfer means is connected to the bidirectional vertical CCD 13 via the b side of the switching means 35. It should be noted that pixels of vertical 6 pixels × horizontal 16 pixels are transferred to the selective transfer means.

The electric charge obtained through the switching means 35 is selected by the selection switching element 4 controlled by the output selection signal and output from the output signal line 8, or is swept by the sweep selection signal. It is selected by the output switching element 6 and swept to the sweep signal line 9. At this time, the output signal C ₈ passing through the output signal line 8 is further selected by the signal passing filters 41 and 43 only in the necessary portions, and output from the output signal lines 42 and 44, respectively. This is to transfer the two regions to different outputs at exactly the same time.

Based on the timing charts of FIGS. 34 and 35, the operation of only the part of the operation of the image sensor different from that of the third embodiment will be described.

FIG. 34 shows the timing when the image is periodically output from the output signal lines 27 and 39 of the horizontal CCD 26 and the horizontal CCD 38 without the output by the selective transfer means. . Images in the same range are output from the two output signal lines 27 and 39 in the same cycle by the same operation as in the third embodiment .

In this figure, the shutter drive signal C ₃₂
And the time zone indicated by F of the vertical CCD drive signal C ₁₁ indicates the time zone in which the operation is possible regardless of the operation of these two outputs. That is, in these time zones F, in order to obtain the output from the output signal lines 8, 42, 44 using the selective transfer means, the shutter 31 and the bidirectional vertical CCD 1 are freely selected.
3 or the like can be used.

FIG. 35 shows the timing when the shutter 31 and the bidirectional vertical CCD 13 are used to simultaneously output two relatively small images in an arbitrary range under the same conditions as in FIG. Separate output signals C ₄₂ and C ₄₄ are output from the two output signal lines 42 and 44 at the cycle T, and a total of 4 output signals C ₂₇ and C ₃₉ of the CCD of the cycle 2T are combined. It can be seen that two outputs are obtained from the same light receiving part.

As a result, the image sensor according to this embodiment can output images taken by the common light-receiving section in parallel, as in the third embodiment . In addition to this, since the selective transfer means is provided, the area of the image to be output can be arbitrarily selected. Therefore, an image with a small number of pixels can be output at high speed.
It can be used for input systems that perform real-time visual processing. Further, by providing the signal pass filters 41 and 43, it is possible to simultaneously output the images of the areas of different sizes which are required for various kinds of processing.
It is possible to realize an input device of a parallel image processing system that performs a plurality of image processes in parallel with a single image sensor.

In FIG. 35, the output signal C ₄₂ · C
₄₄ indicates that the image of a small area is periodically output at a fixed cycle, but if the area of the area to be transferred is the same, the cyclic operation is possible even if the transfer area is changed every time. Is.

Further, as shown in FIG. 34, the time zone F can be freely used even when a periodic image is output from the output signal lines 27 and 39, so that any time zone F can be used. It is also possible to change and obtain an image of an arbitrary area required for each frame.

Further, if the period of the output from the output signal lines 27 and 39 is made longer or variable, and the degree of freedom is given, an image of a more complicated pattern can be obtained at a more complicated timing as a whole. it can.

Next, consider a case where the image sensor of this embodiment is applied to a high-definition photographing device. Since high-definition video has approximately 2 million pixels, which is far more than the 200,000-400,000 pixels of conventional TV images, it is possible to output extremely high-definition images when shooting the same area. Is. On the other hand, if the resolution is the same, a much wider area (about 5-10
Fold) can be taken. In other words, by using a high-definition camera in an industrial field such as FA, it is possible to make an image processing application system for many objects at once.

On the other hand, however, it takes time to image-process all high-definition images due to the large number of pixels. Further, since the amount of pixels for one screen output from a high-definition camera is large, it is much more difficult to shorten the time required to output an image for one screen than with conventional cameras. In other words, even if an image application device is realized by simply combining an HDTV camera and an image processing device that targets the number of pixels, the processing speed does not correspond to the speeding up currently required for FA devices. It is possible, and as a result, it can be said that there is not much merit in using a high-definition system.

Further, as described above, in many cases, the part requiring image processing is not the whole but local.
It can be said that transferring all the areas projected by the high-definition camera in order to transfer this local portion is a more useless operation than in the case of using the conventional camera system. However, the application of the high-definition system is 1
It is also true that there is a merit that you can take a very large number of areas each time.

If an image sensor that can realize high-speed partial transfer while transferring the entire area of this embodiment is applied, for example, the following operation can be performed.

The entire area (high-definition entire area) is transferred at a fixed cycle, and the image is transferred to the image display device for night vision or image processing for overall processing. At this time, the book
In the reference mode , since the entire image can be output in parallel, the high-definition image can be transferred at a faster cycle. In this image processing apparatus for overall processing, processing such as determining where a target object is is performed only in a relatively slow cycle or when necessary. Then, in parallel, the partial images of only the target region are transferred at an early cycle, and high-speed image processing is performed with only a few pixels to perform partial high-speed image feedback control. In the present reference mode , the range in which partial transfer is possible is arbitrary, and thus when the target object is displaced, it can be used even when the target object moves by changing the partial transfer range according to the target. In particular, when a high-definition camera is used, the whole area is wide, so it is possible to follow an object wider than ever before without moving the camera.

As described above, by shooting a high-definition image with all pixels and performing partial transfer at the same time, it is possible to obtain a wide-range image of high-definition and a high-speed cycle image of a necessary and efficient part at the same time. There is. For example, a wide range of parts of a factory line is recognized by the whole image, what is where, etc. is recognized, and assembly and adjustment work is performed by high-speed feedback using partial vision obtained from the same camera. It becomes possible. Moreover,
By being able to freely change the partial transfer area, it is possible to deal with variations in adjustment parts and assembly parts, and even with multiple adjustment parts and assembly parts.

The ability to capture a wide range of images means that one camera can perform what was previously realized only by using a plurality of cameras. Therefore, as described in the reference embodiment , in addition to the effect that the installation area of the camera is reduced, complicated operations such as coordinate adjustment between the cameras and physical arrangement of the cameras when performing with a plurality of cameras are deleted. It becomes possible.

In the present embodiment , the two signal passing filters 41 and 43 are connected to the selective transfer means in order to output the two areas at the same time, but the present invention is not limited to this. That is, if there is one area to be output, no filter is required. If there are multiple areas to be output, the number of filters may be changed according to the number. The same can be said for the selective transfer means in the first and second embodiments.

In the first to fourth embodiments , the shutter 31 and the shutter drive signal line 32 for controlling the shutter 31 are used to limit the light receiving time .
In some cases, it is not necessary to use these means when performing a periodic operation.

[0188]

As described above, the image sensor according to claim 1 of the present invention has a plurality of photoelectric conversion units arranged in a matrix and accumulating signal charges according to the amount of incident light, and the photoelectric conversion units. And a vertical charge transfer means for sending out the signal charges taken in from the photoelectric conversion part one pixel at a time from the vertical charge transfer parts at the same time, and a plurality of vertical charge transfer parts arranged along each column. Selection transfer means for arbitrarily selecting and transferring the signal charge sent from the pixel unit in pixel units, wherein the selection transfer means is based on the output selection signal.
By opening and closing each vertical charge transfer unit simultaneously
Multiple selection switches that arbitrarily select the transmitted signal charge.
The switching element and open / close based on the sweep selection signal.
Is not selected as the above switching element due to
Multiple switching elements for sweeping out the signal charge
It consists of and.

As a result, only the necessary pixels can be selected from the pixels for one horizontal line and the partial image can be output, so that high-speed transfer can be performed for the time required to transfer the unnecessary pixels. As a result, the processing speed of the image processing system using this image sensor can be improved.

Further , the selective transfer means is a plurality of switching devices.
Selective transfer with an easy configuration as it is composed of only elements
The effect that it becomes possible to constitute a means is produced.

An image sensor according to a second aspect of the present invention is arranged in a matrix and has a plurality of photoelectric conversion units for accumulating signal charges according to the amount of incident light, and a plurality of photoelectric conversion units arranged along each column of the photoelectric conversion units. Vertical charge transfer section, which is capable of sending out the signal charge taken in from the photoelectric conversion section one pixel at a time from each vertical charge transfer section, and the signal charge sent out from the vertical charge transfer section. Are selectively selected and transferred in pixel units, horizontal charge transfer means for transferring the signal charges sent from the vertical charge transfer means in horizontal line units, and signal charges from the vertical charge transfer means. This is a configuration including control means for controlling to input to either one of the selective transfer means and the horizontal charge transfer means.

Thus, it is possible to select and output whether the partial image is transferred at a high speed by the selective transfer means or the entire image is transferred at a normal speed by the horizontal charge transfer means, so that one image sensor can be used. It is possible to perform multi-reading in which images are output at two different transfer rates. As a result, compared to when using multiple image sensors,
It is possible to improve the processing speed of the image processing system.

According to the image sensor of claim 3, in addition to the structure of claim 2, the vertical charge transfer means has output ends at both ends in the longitudinal direction of each vertical charge transfer part.
It is a bidirectional vertical charge transfer means capable of respectively transferring signal charges to one output end, and the selective transfer means is connected to one output end of the bidirectional vertical charge transfer means and a horizontal charge transfer means is provided. The control signal is connected to the other output terminal, and the control means drives the bidirectional vertical charge transfer means to control which of the two output terminals the signal charge is transferred to.

This makes it possible to realize a multi-readout image sensor which selectively outputs images at two different transfer rates with a simple structure without providing a complicated circuit or the like.

According to the image sensor described in claim 4 , in addition to the configuration described in claim 2 or 3 , the selective transfer means simultaneously opens and closes each vertical charge transfer part by opening and closing based on an output selection signal. A plurality of switching elements for selecting the sent signal charges arbitrarily, and a plurality of sweeping elements that open and close based on the sweep selection signal to sweep away the signal charges not selected by the above switching elements. It is composed of a switching element.

As a result, since the selective transfer means is composed of only a plurality of switching elements, the selective transfer means can be constructed with a simple structure.

[Brief description of drawings]

FIG. 1 is a configuration diagram showing a configuration of an image sensor according to a first embodiment of the present invention.

FIG. 2 is an explanatory diagram showing an effective pixel area of the image sensor.

FIG. 3 is a timing chart showing a drive clock of the image sensor.

FIG. 4 is a timing chart showing another drive clock of the image sensor.

FIG. 5 is an explanatory diagram showing a state in which electric charges are accumulated in a photoelectric conversion light receiving element.

FIG. 6 is an explanatory diagram showing a state where charges are transferred to a vertical CCD.

FIG. 7 is an explanatory diagram showing a processing state when charges in the lowermost line are transferred to a buffer.

FIG. 8 is an explanatory diagram showing a processing state when charges in the second line from the bottom are transferred to a buffer.

FIG. 9 is an explanatory diagram showing a processing state when the charges of the third line from the bottom are transferred to the buffer.

FIG. 10 is an explanatory diagram showing a processing state when the charges on the fourth line from the bottom are transferred to the buffer.

FIG. 11 is an explanatory diagram showing a processing state when the charges of the fifth line from the bottom are transferred to the buffer.

FIG. 12 is an explanatory diagram showing a processing state when the charges of the sixth line from the bottom are transferred to the buffer.

13A and 13B are explanatory diagrams showing another processing state when the charges of the lowermost line are transferred to the buffer, where FIG. 13A is a first half process and FIG. 13B is a second half process.

FIG. 14 is a flowchart showing the operation of the conventional image sensor of FIG.

15 is a flowchart showing the operation of the other conventional image sensor of FIG.

16 is a flowchart showing an operation when partial transfer of a rectangular area is performed by the image sensor of FIG.

FIG. 17 is a flowchart showing an operation when partial transfer of an area having an arbitrary shape is performed by the image sensor of FIG.

FIG. 18 is an explanatory diagram showing an example of a transfer area.

FIG. 19 is an explanatory diagram showing another example of a transfer area.

FIG. 20 is an explanatory diagram showing another example of a transfer area.

FIG. 21 is a configuration diagram showing a configuration of an image sensor according to a second embodiment of the present invention.

FIG. 22 is a configuration diagram showing another configuration of the image sensor.

FIG. 23 is a configuration diagram showing a configuration of an image sensor according to the first reference embodiment .

FIG. 24 is a timing chart showing a drive clock of the image sensor.

FIG. 25 is a timing chart showing a periodic operation of the drive clock.

FIG. 26 is a configuration diagram showing a configuration of an image sensor according to a second embodiment .

FIG. 27 is a timing chart showing a drive clock of the image sensor.

FIG. 28 is a timing chart showing a periodic operation of the drive clock.

FIG. 29 is a configuration diagram showing another configuration of the image sensor.

FIG. 30 is a configuration diagram showing a configuration of an image sensor according to a third embodiment .

FIG. 31 is a timing chart showing a drive clock of the image sensor.

FIG. 32 is a timing chart showing a periodic operation of the drive clock.

FIG. 33 is a configuration diagram showing a configuration of an image sensor according to Embodiment 4 .

FIG. 34 is a timing chart showing an operable time zone of the operation of the image sensor.

FIG. 35 is a timing chart showing a drive clock of the image sensor.

FIG. 36 is a configuration diagram showing a configuration of a conventional image sensor.

FIG. 37 is an explanatory diagram showing an imaging area.

[Fig. 38] Fig. 38 is a configuration diagram illustrating a configuration of a conventional image sensor for thinning out in units of horizontal lines.

[Fig. 39] Fig. 39 is a configuration diagram showing a configuration of another conventional image sensor that performs thinning in units of horizontal lines.

FIG. 40 is an explanatory diagram showing an example of an imaging region that cannot be transferred at high speed by the image sensor of FIGS. 38 and 39.

FIG. 41 is an explanatory diagram showing another example of the imaging region.

FIG. 42 is an explanatory diagram showing another example of the imaging region.

FIG. 43 is an explanatory diagram showing another example of the imaging region.

[Explanation of symbols]

1 photoelectric conversion light receiving element (photoelectric conversion section) 2 vertical CCD (vertical charge transfer means / unidirectional vertical charge transfer means) 3 buffer 4 selection switching element 6 sweeping switching element 13 bidirectional vertical CCD (bidirectional vertical charge transfer) Means) 15.26.38.45 horizontal CCD (horizontal charge transfer means) 23 connection means 24.37 vertical CCD (charge storage transfer means) 29.30 charge sweeping section 35 switching means

─────────────────────────────────────────────────── ─── Continuation of front page (58) Fields surveyed (Int.Cl. ⁷ , DB name) H04N 5/335 H01L 27/148

Claims (4)

(57) [Claims] 1. A plurality of photoelectric conversion units arranged in a matrix and accumulating signal charges according to the amount of incident light, and a plurality of vertical charge transfer units arranged along each column of the photoelectric conversion units, Vertical charge transfer means for sequentially sending out the signal charges taken in from the photoelectric conversion portion one pixel at a time from each vertical charge transfer portion, and selection for arbitrarily selecting and transferring the signal charges sent out from the vertical charge transfer means in pixel units and a transfer means, said selecting transfer means, line-off based on the output selection signal
Signal transmitted from each vertical charge transfer unit simultaneously
Switching elements for arbitrarily selecting signal charges
By opening and closing based on the sweep selection signal,
Signal charges not selected by the switching element for selection
An image sensor comprising: a plurality of sweeping switching elements that are swept away . 2. A plurality of photoelectric conversion units arranged in a matrix and accumulating signal charges according to the amount of incident light, and a plurality of vertical charge transfer units arranged along each column of the photoelectric conversion units, Vertical charge transfer means capable of sequentially sending out the signal charges taken in from the photoelectric conversion part one pixel at a time from each vertical charge transfer part, and the signal charges sent out from the vertical charge transfer means are arbitrarily selected in pixel units. Transfer means for transferring the signal charges from the vertical charge transfer means, horizontal charge transfer means for transferring the signal charges sent from the vertical charge transfer means in horizontal line units, and signal charges from the vertical charge transfer means for the selective transfer means and the horizontal charge transfer means. An image sensor, comprising: a control unit that controls to input to either one of the above. 3. The vertical charge transfer means has bidirectional vertical charge transfer means having output terminals at both longitudinal ends of each vertical charge transfer section and capable of transferring signal charges to the two output terminals respectively. The selective transfer means is connected to one output end of the bidirectional vertical charge transfer means, the horizontal charge transfer means is connected to the other output end, and the control means drives the bidirectional vertical charge transfer means. The image sensor according to claim 2, wherein the drive signal is a drive signal for controlling which of the two output terminals the signal charge is transferred to. 4. The selection transfer means is based on an output selection signal.
By opening and closing each vertical charge transfer unit simultaneously
Multiple selection switches that arbitrarily select the transmitted signal charge.
The switching element and open / close based on the sweep selection signal.
Is not selected as the above switching element due to
Multiple switching elements for sweeping out the signal charge
4. The method according to claim 2 or 3, characterized in that
Image sensor.