JP2000261728A - Solid-state image pickup device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To add only pixels in a part desired to be read by horizontal and vertical selection circuits and to obtain the pixels as a needed signal by outputting added value of signals of plurality of optoelectric conversion elements in the same column among plural optoelectric conversion elements or outputting added value of signals of plurality of optoelectric conversion elements in the same row. SOLUTION: When signals for all pixels are successiveiy read, a read row selection circuit 11 successively selects a row one each at a time, and while one row is selected, a read column selection circuit 12 successively selects a column one each at a time. Then, a pixel signal is immediately and also successively read to an addition circuit 13 including a reading circuit one pixel after another. When signals are desired to be read only from a prescribed area, the circuits 11 and 12 select prescribed row and column, and a desired pixel signal is immediately and also successively read to the circuit 13 with one pixel each at a time in the area. Due to this read scanning, the rapidity of the reading time of pixel signals only in a prescribed area is guaranteed.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a solid-state image pickup device for picking up an image of a subject such as a monitoring camera.

[0002]

2. Description of the Related Art Conventionally, incident light from a subject is imaged,
Image sensors that obtain an electric signal corresponding to incident light after photoelectric conversion are widely known. The image sensor is required to have a high-definition output image year by year, and the number of pixels of the image sensor has been increased.

With the increase in the number of pixels of such an image sensor, in recent years, from the conventional method of simply outputting the signals of all the pixels of the image sensor simply and sequentially, if necessary, some pixels from the image sensor can be used. There has been an increasing demand for a method of outputting only a signal or processing a signal of a pixel to some extent before outputting the signal from an image sensor.

For example, by adding signals between adjacent pixels,
Outputting from an image sensor has often been performed in terms of color processing, resolution conversion, and the like. It is also possible to add signals of a larger number of pixels in the image sensor.

FIG. 7 shows a flow of signals when only one specific vertical column is read from a conventional CCD image sensor. The charge accumulated in each pixel is a vertical CCD
The data is transferred to the register VCCD.
After being transferred in the D register HCCD, it is output to an output terminal via an output amplifier. When it is desired to add the signals of the pixels for one column, the signal output from the output terminal may be added externally. However, for example, a capacitor is provided at the output unit of the VCCD, and the addition is performed. It is also possible to add in the VCCD or HCCD inside the CCD image sensor.

FIG. 9 shows a signal flow when only one specific horizontal line is read from the CCD image sensor. The electric charge accumulated in each pixel is similarly calculated by the vertical CCD.
The data is transferred to the register VCCD.
After being transferred in the D register HCCD, it is output to an output terminal via an output amplifier. When it is desired to add the signals of the pixels arranged in one column, the signal output from the output terminal may be externally added. However, for example, a capacitor is provided at the output unit of the HCCD to sequentially add the charges from the VCCD. As a result, H inside the CCD image sensor
It is also possible to add in the CCD.

[0007]

However, in the above-mentioned conventional example, almost the same operation as reading out all pixels is performed even when reading out only a part of pixels in the image sensor, signals for one column or one row. It had to be done and wasted much. For example, in the case of reading out one vertical column of pixels as shown in FIG. 8, if the pixel column to be read is near the left end, the remaining right side must be sequentially scanned, and the HCCD which is longer by that length.
It cannot be output unless it is transferred inside. Horizontal 1 as shown in FIG.
When reading out pixels of a row, if the pixel row to be read is near the lower end, the remaining upper side must be sequentially vertically scanned,
The data cannot be output unless the data is transferred through each of the longer VCCDs and further transferred within the HCCD. As described above, even in the case where only a part of the conventional CCD image sensor is read, high-speed reading is difficult, and the power consumption is not much reduced as compared with the normal reading.

An object of the present invention is to provide a CCD image sensor in which a part of a pixel is read at high speed and power consumption corresponding to the read part is achieved.

[0009]

According to the present invention, in order to solve the above-mentioned problem, only pixels of a portion to be read out using a CMOS image sensor are selected by a horizontal and vertical selection circuit, and an addition process is performed in a readout circuit. In this way, a necessary signal is obtained.

According to the present invention, in a solid-state imaging device having a plurality of photoelectric conversion elements, an added value of signals of a plurality of photoelectric conversion elements in the same column or a plurality of signals in a same row among the plurality of photoelectric conversion elements is provided. It has an adding means for outputting an added value of the signals of the photoelectric conversion elements.

In the solid-state imaging device, a means for outputting an added value of signals of a plurality of photoelectric conversion elements in the same column at a plurality of locations or an added value of signals of a plurality of photoelectric conversion elements in the same row is provided. It is characterized by having.

Further, in the solid-state imaging device, among the plurality of photoelectric conversion element groups at which the added value is calculated,
At least one photoelectric conversion element group is in a different column or row from other photoelectric conversion element groups.

[0013]

(First Embodiment) FIG. 1 is a diagram showing a configuration of a CMOS image sensor according to a first embodiment of the present invention. In FIG. 1, a read column selection circuit 12 is a circuit for selecting columns of pixels 14 arranged in a matrix or two-dimensional pattern to be read, and a read row selection circuit 11 is a circuit for selecting pixels 14 to be read. This is a circuit for selecting a row. The addition circuit 13 is a circuit for performing addition between the signals of the plurality of pixels 14 that have been read, and has a configuration in which a normal signal reading operation without performing the addition processing can be performed.

In the above configuration, when sequentially reading the signals of all the pixels, the readout row selection circuit 11 selects the rows one by one sequentially, and while one row is selected, the readout column selection circuit 12 sequentially selects the rows. The columns are selected one by one, and the pixel signals are read out one by one immediately and sequentially to the addition circuit 13 including the readout circuit.

When signals are read only from the odd-numbered rows or the even-numbered rows, every other row is sequentially selected by the read-out row selection circuit 11, and while one row is selected, the read-out column selection circuit is selected. The columns are sequentially selected one by one by 12 and pixel signals are read out one by one immediately and sequentially to an addition circuit 13 including a readout circuit. By this scanning, every other row is sequentially selected by the read-out row selection circuit 11, so that the scanning time is halved with respect to the reading of all pixels, and the signal reading time of only the odd-numbered rows or even-numbered rows is faster. Secured.

When it is desired to read out signals only from pixels in a specific area, a specific row and column are selected by a readout row selection circuit 11 and a column selection circuit 12, and a signal of a desired pixel is read out of the area. The data is read out to the readout circuit immediately and sequentially for each pixel. By this read scanning, necessary rows and necessary columns are sequentially selected by the read row selection circuit 11, so that the scanning time for reading all pixels is equivalent to the ratio of the specific area to all pixels, Only the signal read time ensures high speed.

Next, an addition process between a plurality of pixel signals performed in the image sensor having such a configuration is described. Various specific configurations are conceivable. JP-A-4-
As described in Japanese Patent No. 4682, an example in which connection and drive timing in a readout circuit are changed and addition processing is performed in an image sensor is different from a configuration in which all pixels are normally read out sequentially. It is possible.

FIG. 2 shows a circuit diagram of a simple addition circuit of the solid-state imaging device. As shown in the figure, the signal components from the vertical output lines connected to the respective pixels 14 are stored in the storage capacitors 43 and 49 via the transfer switches 41 and 48, and at the next timing, the signal components of the pixels 14 of the other rows are stored. Transfer switch 42
, And then stored in the storage capacitor 45, and then transferred by the transfer switch 46,
By turning on 47 and 50, an added component of two pixel signals of the vertical output line can be obtained on the horizontal output line 62. At this time, in FIG. 2, the signal component of the storage capacitor 45 is weighted together with the signal component of the storage capacitor 43, and the signal component of the pixel of the other row is also weighted, so that an added signal component of at least three pixels can be obtained.

FIG. 3 shows a first embodiment of the present invention.
This shows the arrangement of pixels for which addition processing is to be performed in the image sensor, and is a one-row pixel group 22 located substantially at the center of the screen with respect to the entire pixel area 21. As an example of an actual subject, for example, as shown in FIG.
Consider a case where 3 moves horizontally from left to right.

The read column selection circuit 12 selects one column to be subjected to the addition processing which is specified in advance, and the read row selection circuit 11 sequentially or simultaneously selects all the rows, whereby the pixels of the pixel column to be added are selected. The group signals are read out to the adder circuit. The addition value of the signal of the pixel group selected from the input signal is added and output from the addition circuit 13. FIG. 5 shows the change with time. Only when the white sphere 23 is on the selected pixel group on the image sensor, the addition signal increases greatly, and when passing, a change appears to return to the original signal level. Here, the column to be subjected to the addition processing is set at the center of the screen, but the white sphere 2 shown in FIG.
If a column is set on the left side (23a) of No. 3, the temporal change of the added value moves to the left, and if a column is set on 23c, the temporal change of the added value moves to the right.

As described above, by observing the added value of the pixel signals in one column, it is possible to detect whether or not the image picked up by the image sensor has moved from its temporal change.

Here, for the sake of simplicity, the case where the movement of a white sphere in a uniformly dark background is detected has been described.
Even if there is some background image, if it is stationary as a background, only the offset value of the added value is added and it does not affect the change of the added value. Can be detected.

When such an addition process of pixel signals for one column is performed, the signal level of the added value is higher than the signal level of a single pixel by an amount added, so that the charge accumulation time of each pixel is short. Time is fine. When detailed image data of the entire screen is not required, and only information as to whether or not the subject in the image has motion is needed, for example, if such a single-column addition signal is handled, all pixels of the image sensor can be used. Necessary information is directly compared to the case where signal processing is performed by reading signals.
Obtained in a short time. Further, in the CMOS image sensor, even if a signal is read from each pixel, the signal stored in each pixel is not affected, so that charge accumulation can be continued without resetting each pixel and read again. .
That is, it is possible to observe the addition signal of the pixel group for one column, make some judgment from the addition value level and its change, and read out the signals of all the pixels again.

When the same operation is performed by the CCD image sensor, the driving of the vertical CCD register VCCD and the horizontal CCD register HCCD is not much different from the case of reading out all pixels. In a CMOS image sensor, since only the pixel column to be read is read, the CCD
Compared with the image sensor, the pixel readout of the area other than the specific area is not performed in a short time, so that a desired addition signal can be obtained with low power consumption.

(Second Embodiment) FIG. 6 shows an arrangement of pixels to be subjected to addition processing in an image sensor according to a second embodiment of the present invention. In the figure, a pixel group to be subjected to addition processing is divided into three by four rows and four columns by an image sensor, and arranged on a mesh like a row selection pixel group 32A to 32d and a column selection pixel group 33a to 33d. Have been.
The configuration of the image sensor is the same as that shown in FIG.

As in the first embodiment, as an example of an actual subject, consider a case where a white sphere moves horizontally from left to right in a uniformly dark background as shown in FIG.

First, pixel groups 32A to 32A to be added in the row direction
The rows in which 2d are arranged are sequentially read out by the read row selection circuit 1
Select by 1. While the first row to be subjected to the addition processing is selected, the readout column selection circuit 12 causes the first pixel group 3
By sequentially or simultaneously selecting the columns in which 3a-33d are arranged, the signal of the first pixel group is added to the addition circuit 13
To be read. When the signals of all the first pixel groups are input, the addition circuit 13 outputs the addition signal to an output terminal. Subsequently, similar signal readout and addition processing are performed for the second and third pixel groups to obtain addition signals of the three pixel groups in the first row.

Further, the same operation is performed for the second to fourth rows where the addition processing is performed, and three addition signals for each of the four rows are obtained.

Next, the columns in which the pixel groups to be added in the column direction are arranged are sequentially selected by the read column selection circuit 12. While the first column to be subjected to the addition processing is selected, the row in which the first pixel group is arranged is sequentially or simultaneously selected by the readout row selection circuit 11 so that the signal of the first pixel group is added to the addition circuit. 13 is read. Addition circuit 1
When the signals of all the first pixel groups are input, 3 outputs the added signal to the output terminal. Subsequently, similar signal readout and addition processing are performed for the second and third pixel groups to obtain addition signals of the three pixel groups in the first column.

Further, the same operation is performed for the second to fourth columns in which the addition process is performed, and three addition signals for each of the four columns are obtained.

FIG. 6 shows the temporal change in the added value of the second pixel group of the four columns, that is, the pixel group signal arranged at the center portion, of the pixel group for performing the addition processing of the image sensor in FIG. This is shown in FIG. Only when a white sphere is on each of the selected groups of pixels on the image sensor does the addition signal increase significantly, with a change back to the original signal level appearing after passing. The four added value signals show the same change except for the timing at which the change occurs.

As in the first embodiment, it is possible to detect whether or not the image picked up by the image sensor has moved from the temporal change of each of the added values. The direction in which the subject is moving can be known. In addition, the moving position of the subject in the screen can be determined based on which of the pixel groups has changed, and the direction of the movement can be determined from the changes in the pixel groups arranged in the row direction and the column direction.

Even if there is some background image as in the first embodiment, if the background is stationary, only the offset value of the added value is added and the change of the added value is not affected. The waveform can be observed, and the motion of the subject can be detected.

When such pixel signal addition processing is performed, the signal level of the added value is higher than the signal level of a single pixel by the amount added, so that the charge storage time of each pixel is short. When detailed image data of the entire screen is not required, for example, only information such as where and in which direction the subject in the image is moving is needed, by using such a pixel group addition signal, Necessary information can be obtained directly in a short time as compared with the case where the signals of all the pixels of the sensor are read and the signal processing is performed.

Further, in the CMOS image sensor, even if a signal is read from each pixel, the signal stored in each pixel is not affected, so that charge accumulation is continued without resetting each pixel and read again. Is also possible. That is, it is also possible to observe the added signals of several pixel groups arranged in the image sensor, make some judgment from the added value level and its change, and read out the signals of all the pixels again.

When the same operation is performed by the CCD image sensor, the driving of the vertical CCD register VCCD and the horizontal CCD register HCCD is not much different from the case of reading all pixels. In a CMOS image sensor, since only the pixel column to be read is read, the CCD
A desired addition signal can be obtained in a shorter time and with lower power consumption than the image sensor.

As an example of applying this embodiment,
It can be operated as means for detecting a moving point in all pixels. Based on the above detection result of 4 rows and 4 columns, for example, after 100 msec, 4 rows and 4 columns corresponding to the same address are detected as the addition signal shown in FIG. 7 and compared with the previously detected value, and there is a change. Next, an area having the change is specified, and a motion area is specified. If there are a plurality of motion areas, the area can be specified as a moving area in a plurality of areas. Also, 100 ms later, the pixel addition signal of the area of 4 rows and 4 columns is detected again with the motion area as the center, and the area having the change is identified by comparing with the previously detected value.
A motion area can be detected. As described above, by scanning the area of all pixels and obtaining an addition signal of only a specific area,
A motion area can be detected, and power consumption at that time can be reduced.

In the above-described embodiment, an example of obtaining the detection result of 4 rows and 4 columns has been described. However, from the nature of the target image, a coarser, for example, 2 rows and 2 columns is specified to read the change of the target image. Alternatively, a finer matrix may be used. For example, when an attempt is made to detect a change in the unevenness of the sea surface, it is only necessary to detect the detection points in one row and n columns, and the setting can be arbitrarily set. Thus, the power consumption of the imaging device depends on the characteristics of the target image.

[0039]

As described above, according to the present invention, the CM
Using an OS image sensor, only a specific pixel group is read out, an addition process is performed, and then an addition signal is output. Therefore, information necessary for reading out all the pixels and performing equivalent signal processing is smaller than that in a case where the same signal processing is performed. There is an effect that it can be directly taken out in a short time. Since a signal is added between a plurality of pixels, a desired signal level is reached in a shorter accumulation time than a normal all-pixel readout, high-speed processing is possible, and a part to be operated is limited. Power consumption can be significantly reduced as compared with operation.

After reading out the pixel group that has been subjected to the addition processing, if the charge stored in the pixel is not reset and the storage is continued, some judgment is made based on the addition signal, and then all the pixels are reset. It is also possible to read out pixel signals or read out only a specific pixel group again to perform addition processing.

[Brief description of the drawings]

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

FIG. 2 is a circuit diagram of an addition circuit of the solid-state imaging device according to the present invention.

FIG. 3 is an explanatory diagram of an operation of detecting a pixel area of the solid-state imaging device of the present invention.

FIG. 4 is an explanatory diagram of an operation of detecting a pixel area of the solid-state imaging device according to the present invention.

FIG. 5 is an explanatory diagram of an operation of detecting a pixel area of the solid-state imaging device of the present invention.

FIG. 6 is an explanatory diagram of an operation of detecting a pixel area of the solid-state imaging device according to the present invention.

FIG. 7 is an explanatory diagram of an operation of detecting a pixel area of the solid-state imaging device according to the present invention.

FIG. 8 is an explanatory diagram of an operation of detecting a pixel area of a conventional solid-state imaging device.

FIG. 9 is an explanatory diagram of an operation of detecting a pixel area of a conventional solid-state imaging device.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 11 Read-out row selection circuit 12 Read-out column selection circuit 13 Addition circuit (read-out circuit) 14 Pixel 21, 31 All pixel area 22 Pixel row of selection read-out area 23 White sphere 32 Row selection area 33 Column selection area 41, 42, 46-48 , 50 transfer switch

Claims (8)

[Claims] 1. A solid-state imaging device comprising a plurality of photoelectric conversion elements arranged in a matrix, wherein, among the plurality of photoelectric conversion elements, an added value of signals of a plurality of photoelectric conversion elements in the same column or a plurality of signals in the same row. A solid-state imaging device having an adding unit that outputs an added value of the signals of the photoelectric conversion elements. 2. The solid-state imaging device according to claim 1, wherein an added value of signals of a plurality of photoelectric conversion elements in the same column at a plurality of locations or an added value of signals of a plurality of photoelectric conversion elements in the same row is provided. A solid-state imaging device comprising output means. 3. The method according to claim 1, wherein at least one of the plurality of photoelectric conversion element groups for which the sum has been calculated is in a different column or row from the other photoelectric conversion element groups. Item 3. The solid-state imaging device according to Item 2. 4. The solid-state imaging device according to claim 2, wherein all of the plurality of photoelectric conversion element groups for which the added value is calculated are in the same column or row. 5. The image processing apparatus according to claim 1, further comprising: the addition unit, and a motion detection unit configured to detect a motion of the subject from a temporal change of the same column addition value or the same row addition value. Solid-state imaging device. 6. A motion detecting means for detecting a motion of a subject from a temporal change of the same column sum or the same row sum at the plurality of locations.
The solid-state imaging device according to any one of the preceding claims. 7. The solid-state imaging device according to claim 3, further comprising a motion detection unit configured to detect a motion of a subject based on a temporal change of the same column addition value or the same row addition value at the plurality of locations. 8. A motion detecting means for detecting a motion of a subject from a temporal change of the same column addition value or the same row addition value at the plurality of locations.
The solid-state imaging device according to any one of the preceding claims.