JPS63212278A - Solid-state image pickup element and range finder using it - Google Patents

Abstract

PURPOSE:To read picture element information again and to directly execute the correlation operation of a picture or the like based on this output signal by storing and maintaining the terminal voltage value of a capacity element as the picture element information. CONSTITUTION:In a solid-state imagepickup element, during the information storage period of a detecting cell 1, a photodiode 2 and the capacity element C1 are made conductive by an MOS transistor FT 2 as a switching element, a photoelectric current according to the incident light intensity of a light incident on the photodiode 2 passes through a closed circuit and the terminal voltage value of the capacity element C1 goes to the picture element information. This terminal voltage value is read through an MOS transistor FT 4 as a current amplifying circuit, even if this terminal voltage value is read again, it is not changed, thereby, the detecting cell in which the picture information can be stored and held and which can be used even by repeatedly reading, the same picture element information is obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a solid-state image sensor used in cameras and the like, and a distance measuring device using the same.

[Conventional technology]

Conventionally, M
OS type or CCD type photoelectric conversion elements are known. This rice solid-state imaging device consists of a plurality of detection cells arranged in one or two dimensions with the same structure, and each detection cell is adapted to capture information of one pixel of an image. Further, output signals from each detection cell, that is, pixel information, are output to a common video line.

FIG. 9 is a configuration diagram of a detection cell in a conventional solid-state image sensor. In FIG. 9, a detection cell 80 includes a photodiode 81 and a switching element FT8 that controls the readout timing of an output signal from the photodiode 81 to a video line 82.

The switching element FT8 is, for example, an N-channel type M
This is an OS transistor, and when it is on, it makes the photodiode 81 and the video line 82 conductive,
When off, photodiode 81 and video line 8
2 is electrically isolated from each other.

In the detection cell 80 having such a configuration, when light is incident on the photodiode 81, charges are accumulated in the junction capacitance C0 of the photodiode 81 according to the intensity of the incident light. Switching element FT8 is off while charges are being accumulated. When the switching element FT8 is turned on, the charge accumulated in the junction field JIC0 of the photodiode 81 is transferred to the switching element FT8. For example, it is output to the video line 82 as one pixel information via an MO3 inversion layer of a MOS transistor.

As a result, by sequentially turning on the switching elements of each of the plurality of detection cells in the solid-state image sensor,
Pixel information can be read out onto video line 82 sequentially or serially.

Furthermore, distance measuring devices using the above-mentioned solid-state image sensor are conventionally known to measure the distance to an object to be measured.

FIG. 10 is a configuration diagram of a conventional distance measuring device disclosed in Japanese Patent Application Laid-Open No. 58-148910. The conventional device shown in FIG. 10 is equipped with a solid-state image sensor 106 that receives two light beams F3 and F4 from the same object that have passed through spatially different locations. One of the light beams F3 from the object to be measured is incident on the detection cells a1 to aS of the original array, and the detection cells b to b of the -dimensional array
The other light beam F4 is made incident on +.

Detection cells a to aH of the solid-state image sensor 106.

The detection cells b to bH have the same structure as the detection cell 80 shown in FIG. 9, and the detection cells a1 to a and the detection cells b1 to bM contain mutually shifted pixel information of the same treasure. is accumulated in analog quantities. In this distance measuring device, these detection cells a to aN, detection cells b1 to b
The pixel information stored in M is sequentially taken out and sent to the detection cells b1 to bs.
Correlation calculation result C(1) between pixel information of and pixel information of detection cells a1 to aH, old..., detection cells b1 to bH
Correlation result C(p) between pixel information of pixel information and pixel information of detection cells a to ap-1+H.

Pixel information of detection cells b to bH and detection cell aN-M+
Correlation result C (N-M+1.) with pixel information from 1 to aH
are obtained sequentially, and the correlation result c(i), old...

The distance to the object to be measured is detected by detecting the shift amount p that gives the maximum correlation among C(p), . . . , C(N-M+1).

In order to obtain these correlation calculation results, the solid-state image sensor 10
The six detection cells a1 to aH, b1 to bH are connected to a common video line 82, and are further connected to a shift register 10 having an M-bit transfer stage via a binarization circuit 108.
It is connected to the.

The shift register 10 is further connected to a shift register 11 having a (N-M+1) bit transfer stage. For convenience of explanation, shift register 1 is shown in FIG.
1 is shown as having 9-bit transfer stages 121-129.

The output of the first transfer stage 112 of the shift register 110 is
Each transfer stage 121 to 129 of the deep shift register 111 is connected to one input terminal of the exclusive OR circuits 131 to 139, and each transfer stage 121 to 129 of the corresponding exclusive OR circuit 131 to 13
9 is applied to the other terminal. The exclusive OR results obtained by the exclusive OR circuits 131 to 139 are added to the counters 161 to 169 via the OR gates 141 to 149 and the AND gates 151 to 159. Correlation calculation result C(1),...,C(p),...
..., C(N-M+1 > (in the example of Fig. 10, C(1), ..., C(+)), ...
, C(9)) are output.

Note that the solid-state image sensor 106, the binarization circuit 108, the shift registers 110, 111, and the counters 161 to 16
9 is controlled by a controller 119, and correlation calculations are controlled by gates 114 to 116.

In a distance measuring device having such a configuration, two light beams F3. When F4 enters the solid-state image sensor 106,
The detection cells a1 to aN in the -dimensional array capture image information of the light flux F3, and the detection cells b to bH in the -dimensional array capture image information of the light flux F4. Note that in FIG. 11(a), the images of the light beams F3 and F4 are shifted from each other by a shift amount p.

The detection cells a1 to J and the detection cells b1 to bM of #i as shown in FIG. 9 have a light flux F3. The image of F4 is accumulated as analog pixel information (as the amount of charge of the junction capacitance) as shown in FIG. 11(b).

Next, the analog pixels ff accumulated in the detection cells a1 to aH and the detection cells b1 to bH of the solid-state image sensor 106
The f information is sequentially output to the video line 82 by sequentially turning on the switching elements of each detection cell similar to the switching element FT 8 in FIG.
11(C) by the binarization circuit 108.
The information is converted into binarized digital pixel information as shown in FIG. 1 and sent to shift registers 10 and 111.

In the correlation calculation, the binarized pixel information from the detection cells a to aH is transferred to each transfer stage 121 to 129 of the shift register 11, and the binarized pixel information from the detection cell b1 is transferred to the first transfer stage 112 of the shift register 10. Starts when pixel information is transferred. Note that the period during which the correlation calculation is performed is the logical sum game) 141 to 149. AND gate 15
1 to 159.

12 to 14 show the correlation calculation results C(1).

It shows how C(El) and C(N-M+1) are obtained. In the exclusive OR circuit 131, the binary pixel information of the detection cells b1 to bN as shown in FIG. Detection cells a as shown in FIG. 12(b) are sequentially sent to the transfer stage 121 of
Exclusive OR with the binarized pixel information of 1 to aH,
Exclusive OR signals as shown in FIG. 12(C) are sequentially output to the counter 61. The counter 61 counts the exclusive OR signals that are sequentially sent, and in the case of FIG. Output.

Similarly, for example, in the exclusive OR circuit 135, the binary pixel information of detection cells b to bM as shown in FIG. Exclusive OR is performed with the binarized pixel information of the detection cell mill or aP-148 as shown in FIG. 13(b), which is sequentially sent to the transfer stage 125 of the shift register 111, ) are sequentially output to the counter 65. The counter 65 counts the exclusive OR signals sent sequentially, and in the case of FIG. 13(C), there is no signal of "1", and the binary pixel information of detection cells b to bH is detected. cell a
Since there is a perfect correlation with the binarized pixel information P P-148 of a to a, "0" is output as the correlation calculation result C(p).

Furthermore, in the exclusive OR circuit 139, the signals shown in FIG.
The binary pixel information of detection cells b1 to bH as shown in a) and the detection cell a as shown in FIG.
Exclusive OR is performed with the binary pixel information of aH to aH, and exclusive OR signals as shown in FIG. 14(C) are sequentially output to the counter 169. The counter 169 counts the exclusive OR signals that can be sent sequentially, and in the case of FIG. 14(C), since two signals of 11" are sent, "2" is taken as the correlation calculation result C (N-M+1). Output.

If we pay attention to the fact that the correlation operation is performed by taking exclusive OR, we get the correlation operation result C(1).

......, C(p) ,...,C(N-
The smallest one among M+1) has the largest correlation.

In the above example, C(p) is the minimum among all the correlation calculation results, so the pixel information of detection cells b1 to b and the image M of detection cells a to a
The correlation with P P-1◆H elementary information is the highest, and it is therefore possible to detect that the image information of the light flux F3 and the image information of the light flux F4 deviate by the amount of deviation p, and from this deviation amount p, The distance to the object to be measured can be detected.

[Problem that the invention seeks to solve]

By the way, in the detection cell 80 of the conventional solid-state image sensor as shown in FIG. 9, when the switching element FT8 is turned on, the charge accumulated in the junction capacitance C0 of the photodiode 81 is transferred to the video line as pixel information. 82, the charge accumulated in the junction capacitance C0 disappears. That is, when pixel information is read out from the detection cell 80 many times, the same pixel information cannot be read out and used again from the same detection cell 80 because the detection cell 80 no longer stores the pixel information. there were.

Therefore, the solid-state image element 1 in which detection cells a to a N and bl to bH having a structure like the detection cell 80 is arranged.
When using 06 in a distance measuring device, read each detection cell a to a, b to b+ sequentially.
Nl After the output analog pixel information is converted into binarized digital pixel information by the binarization circuit 108 as shown in FIG. must be stored in an array suitable for correlation calculation processing, that is,
In the correlation calculation process, pixel information from the same detection cell must be used again, so the detection cells a to aH,
1) Analog pixel information from 1 to bM cannot be used directly for correlation calculation processing. For this reason, shift registers 10 and 111 had to be provided as intermediate buffers for correlation calculation processing. As a result, the detection cells a to aH of the solid-state image sensor.

Since it is not possible to directly perform analog correlation calculation processing based on the analog pixel information from F3. There is a problem in that the amount of deviation P of F4 cannot be detected accurately and it takes time to detect the amount of deviation P.

Furthermore, in the distance measuring device shown in FIG. Exclusive OR circuits 131 to 139, OR gates 141 to 149, AND gates 151 to 1
59, counters 161 to 169 are shift register 1
(N-M+) corresponding to 11 (N-M+1) transfer stages
1) Requires 1) stages. In Fig. 10, for convenience of explanation, the number of stages (N-M+1) is "9", but in applications such as cameras, the number of detection cells is usually about "128" or more, so the number of stages is (, N-M+1) is about "128" or more. For this reason, the circuit scale of the distance measuring device becomes large, and there is a problem in that it is difficult to achieve miniaturization and high speed of the device.

An object of the present invention is to provide a solid-state image sensor equipped with a detection cell that stores and retains pixel information even after the pixel information has been read out once, and that allows the same pixel information to be read out and used again. .

The present invention further provides a simple circuit configuration capable of reading out the same pixel information again from the solid-state image sensor that stores and retaining the pixel information and performing analog correlation calculation processing to accurately and quickly detect the distance of the object to be measured. The purpose is to provide a distance measuring device.

[Means for solving problems]

The solid-state imaging device of the first invention includes: a photodiode into which light enters; a capacitive element; a switching element that brings the photodiode and the capacitive element into a conductive state during an information storage period;
The present invention is characterized in that it includes a detection cell having a current amplification circuit that outputs the terminal voltage value of the capacitive element as pixel information.

A second invention is an invention of a distance measuring device using the solid-state image sensor according to the first invention, in which a plurality of detection cells storing and retaining pixel information in a readable manner are arranged, and the distance measuring device is arranged at spatially different locations. A pair of solid-state image sensors, each of which receives two light fluxes from the same object to be measured, and a detection cell of the pair of solid-state image sensors with readout timings of the detection cells of the pair of solid-state image sensors shifted for each scan. A readout means for scanning the pixel information, and a correlation calculation between the pixel information read out from the pair of solid-state image sensors by the readout means is performed for each scan, and the one giving the largest correlation among the respective correlation calculation results is detected. The present invention is characterized by comprising a detection means for detecting a shift between pixel information.

[Effect]

In the solid-state imaging device of the first invention, the terminal voltage value of the capacitive element of the detection cell is initialized in advance.During the information storage period, the photodiode and the capacitive element are brought into conduction by the switching element. , the switching element and the capacitive element form a closed circuit. As a result, due to the light incident on the photodiode, a photocurrent corresponding to the intensity of the incident light flows through a closed circuit, and the terminal voltage value of the capacitive element changes by this photocurrent amount, and the terminal voltage value at that time becomes pixel information. Become. By the way, the terminal voltage value of the capacitive element is read out via the current amplification circuit, so even if this terminal voltage value, that is, the pixel information is read out again, the terminal voltage value of the capacitive element does not change, so that the image information is stored and retained. can do.

The distance measuring device of the second invention uses the image sensor of the first invention. A pair of solid-state imaging devices of the first invention are provided,
When two beams of light from the same object that have passed through spatially different locations are incident on the pair of solid-state image sensors, each detection cell of the pair of solid-state image sensors stores and holds predetermined pixel information. 0 Then, each detection cell of the pair of solid-state image sensors is scanned by the reading means, and pixel information is read from each detection cell.The image information read from each detection cell of the pair of solid-state image sensors by the reading means is as follows.
A correlation calculation is performed by the detection means for each scan. More specifically, the readout timing of one of the pair of solid-state image sensors is shifted for each number of scans, and the correlation calculation result for each scan is determined.

The detection means detects the deviation between pixel information by detecting the one that gives the maximum correlation among the correlation calculation results, for example, by detecting the number of scans until giving the maximum correlation. Based on this, the distance to the object to be measured can be detected.

〔Example〕

Embodiments of the present invention will be described below based on the drawings.

FIG. 1 is a configuration diagram of an embodiment of a detection cell of a solid-state image sensor according to the present invention.

In FIG. 1, a detection cell 1 is equipped with a photodiode 2 that generates a photoelectric current corresponding to the intensity of incident light. The photodiode 2 is connected to the capacitive element C1 via an n-channel type MOS transistor FT2 as a switching element. MOS) transistor FT2 is
The photodiode 2 and the capacitive element C1 are electrically connected when the voltage of the gate G2 is high level and turned on, and the photodiode 2 and the capacitive element C1 are electrically connected when the voltage of the gate G2 is low level and turned off. When the blocking 0M0S transistor FT2 is turned on and the photodiode 2 and the capacitive element C1 become conductive, the photocurrent I generated in the photodiode 2 closes the closed circuit of the MOS transistor FT2 and the capacitive element C1. As a result, the charge accumulated in the capacitive element C1 is discharged toward the photodiode 2, and its terminal voltage decreases. That is, the capacitive element C1 is a MOS
) When transistor FT2 is on, photodiode 2
Photocurrent flowing through! The configuration is such that the terminal voltage value dropped by , is stored and held as pixel information.

Further, the terminal of the capacitive element C1 connected to the MOS transistor FT2 is further connected to an n-channel type MOS transistor FT3 as a switching element and an n-channel type MOS transistor as a current amplification circuit, that is, an impedance conversion element. It is connected to the gate G4 of FT4. The terminal, that is, the drain of the MOS transistor FT3 on the opposite side to that connected to the capacitive element C1 is held at the reference potential Vref, and the MOS transistor FT3 is turned on when the voltage of the gate G3 is at a high level. A reference potential vre is applied to the terminal of the capacitive element C1.
It is designed to give f. Also, the drain D4 of the MOS) transistor FT4 is at a constant voltage ■. D, while the source S4 is connected to the constant current source 3.

The source S4 of the MOS transistor F4 is further connected to the video line 4 via a switching element, for example, an N-channel MOS transistor FT5.
The 0S transistor FT5 is turned on when the voltage at its gate G5 is at a high level, and at this time, the signal from the source S4 of the MOS transistor FT4 is output to the video line 4 via the MOS transistor FT5. . In other words, the MOS transistor FT4 is used in the form of a source follower with high input impedance and low output impedance.
Since the input impedance of the capacitive element C1 is high, the charge stored in the capacitive element C1 does not leak to the video line 4 via the MOS transistor FT4, and since the output impedance is low, the electric charge stored in the capacitive element C1 added to the gate G of the capacitive element C1 is low.
The terminal voltage value of 1 is impedance-converted and output from the source S to the video line 4, so that even if the video line 4 is long, pixel information can be faithfully transmitted. Note that the constant current source 3 is used to always supply a constant drain current to the MOS transistor FT4.

In the detection cell 1 having such a configuration, the MOS transistor FT as a switching element is used to initialize the terminal voltage of the capacitive element C1 to the reference potential vref before use.
Adding the reset signal RS '1' to the gate G3 of 3,
The voltage of the gate G3 is set to high level, and the MOS transistor FT3 is turned on.

After initializing the terminal voltage of the capacitive element C1 to the reference potentials v and ef, the reset signal R9T is turned off and the operation of storing information based on the intensity of light incident on the photodiode 2 as the terminal voltage value of the capacitive element C1 is started. . This information storage operation is performed using a MOS (MOS) transistor FT as a switching element.
This is done by applying the information storage signal DT to the gate G2 of No. 2. Note that the period during which the information accumulation signal DT is applied is the information accumulation period. During the information accumulation period, Gate G2
The voltage becomes high level and the MOS transistor FT2
is turned on. As a result, as described above, the phototransistor 2, the MOS transistor FT2, and the capacitor C
Since a closed circuit is formed between the photodiode 2 and the photodiode 2 according to the intensity of the incident light, the photocurrent 2, which is generated in the photodiode 2 according to the intensity of the incident light, flows through this closed circuit, and the terminal voltage value of the capacitive element C1 is the initial value. The photocurrent It decreases from the set value "rer. The terminal voltage value of the capacitive element C1 at this time becomes pixel information representing the incident light intensity. When the information storage is completed, the capacitive element C1
Because the input impedance of the MOS transistor FT4 is large, the charges stored in the terminal do not leak to the video line 4 side, so the terminal voltage value is determined by applying the reset signal R to the gate G3 of the MOS transistor F'I'3 again.
It is stored and retained until ST is added.

In order to read out the terminal voltage value stored in the capacitive element C1, that is, pixel information, to the video line 4, the gate G5 of the MOS transistor FT5 as a switching element is
A read signal SP is applied to the MOS transistor FT5, and the voltage of the gate I''Gs is set to high level to turn on the MOS transistor FT5.

As a result, the source S4 of the MOS transistor FT4 and the video line 4 are brought into conduction, and the impedance-converted pixel information from the source S4 of the MOS transistor FT4 is sent to the video line 4. Since the terminal of the capacitive element C1 is connected to the gate G4 of the MOS transistor FT4, even if the MOS transistor FT5 is turned on and pixel information is read, the charge accumulated in the capacitive element C1 is transferred from the MOS transistor FT4 to the video signal. The voltage does not flow to the line 4, and as a result, the terminal voltage value of the capacitive element C1 can be stored and held in the state before reading without changing the terminal voltage value.

In addition, in FIG. 1, the photodiode 2 of the detection cell 1 is further equipped with a switching element (for example, a MOS) transistor FTI.
is connected. The terminal of the MOS) transistor FTI on the side opposite to the side connected to the photodiode 2 is connected to a reference voltage power supply (not shown) that provides a reference voltage ref, for example, and the gate G1 of this MOS) transistor FTI is is a signal DT obtained by inverting the information storage period DT.
is now being added.

In this configuration, the voltage at the gate G1 of the MOS transistor FTI is set to a high level by the signal D'F outside the information storage period, so the voltage at the gate G1 of the MOS transistor FTI is set to high level.
I is turned on, and as a result, the photocurrent generated outside the information storage period in the photodiode 2 is transferred to the transistor FT (MOS).
It can be fed into the reference voltage power supply via I. On the other hand, during the information storage period, the voltage at the gate G1 of the MOS transistor FTI is held at a low level and the MOS transistor FTI is turned off, so that the photocurrent generated in the photodiode 2 does not flow into the reference voltage power supply.

In this way, the incident light outside the information storage period can be used to refocus the image.
- By flowing the photocurrent generated in the diode 2 into, for example, a reference voltage power supply, it is possible to effectively prevent the pluming phenomenon caused by the photocurrent outside the information storage period.

According to the embodiment described above, the incident light intensity is stored and held as a terminal voltage value in the capacitive element C1 in the detection cell 1, and the capacitive element C1 is used as a current amplification circuit, that is, the gate G4 of the MOS transistor FT4 as an impedance conversion element.
Since it is connected to y- of M6S transistor FT5,
Even if readout signal SP is applied again to gate G5, the terminal voltage value of capacitive element C1 does not change, and this terminal voltage value can be read out to video line 4 any number of times as pixel information. This allows processing that requires rereading, such as image correlation calculations and filtering processing, to be performed directly based on the output signal of the detection cell, making it particularly suitable for applications such as distance measuring devices and pattern recognition devices. ing.

Furthermore, since the influence of photocurrent generated outside the information storage period can be blocked by the MOS transistor FTI, high quality pixel information that does not cause the pluming phenomenon can be output.

In the embodiments described above, the switching element FTI.

FT2. FT3. Although FT5 has been explained as an N-channel type MOS transistor, it is an N-channel type MOS transistor.
Other types of elements having a transfer gate function may be used instead of the OS transistor.

In addition, a class A source follower type was used as the current amplification circuit, that is, the impedance conversion element.
Other known impedance conversion elements (e.g. class AB
A type push-pull element) may also be used.

FIG. 2 is a configuration diagram of a distance measuring device using a solid-state image sensor in which a plurality of detection cells as shown in FIG. 1 are arranged.

In FIG. 2, the distance measuring device 11 includes a pair of solid-state image sensors 1
2.13, and the solid-state image sensor 12゜13 has a 10th
Similar to the conventional solid-state image sensing device 106 shown in the figure, two light fluxes Fl and F2 from the same object to be measured, which have passed through spatially different locations, are respectively incident thereon.

A plurality of detection cells 1 as described above are arranged in the solid-state image sensor 12.13, for example, the solid-state image sensor 12.
is N detection cells a1' to a arranged in a -dimensional manner.
The solid-state image sensor 13 includes N detection cells b' to bH' arranged in a -dimensional manner. In addition, each detection cell a' to aN of the solid-state image sensor 12.13
'.

A reset signal R3T and an information storage signal DT as explained in FIG. 1 are added to b1' to bN'. Note that if a blooming prevention switching element FTI as shown in FIG. 1 is provided, it is necessary to further add a signal DT obtained by inverting the information storage signal DT using an inverter (not shown).

In the distance measuring device shown in FIG. 2, detection cells a1' to J',
By simultaneously scanning and reading out the pixel information stored in the detection cells b1' to bN', the correlation calculation results between the pixel information of the detection cells a1' to aH' and the pixel information of the detection cells b1' to bN' are sequentially obtained. At the same time, it is determined whether the pixel information of the detection cells b1' to bN' is shifted to the right or to the left with respect to the pixel information of the detection cells a1' to aN'. Then, scanning is performed by shifting the readout timing of pixel information of detection cells b' to bN' with respect to the readout timing of pixel information of detection cells a1' to a, 4' in the direction of decreasing the amount of deviation, and each time, correlation calculation is performed. It is determined whether the result gives the maximum correlation, and when it is determined that the result gives the maximum correlation, the amount of shift of the pixel information of the detection cells b' to bN' up to that point, that is, the shift amount of the pixel information of the detection cells a' to aN' The amount of deviation P between the image information of the light beam F1 and the image information of the light beam F2 is detected from the number of scans (number of readers), and the distance to the object to be measured is detected.

In order to perform such processing, the distance measuring device shown in FIG.
A pair of shift registers 4.15 and a pair of shift registers 14 sequentially supply read signals SPI and SP2 of Lumi 1' to aN 1 to the detection cells', b' to bN' of a pair of solid-state image sensors 12.13, respectively. ,1
a pair of video lines 16 and 17 from which analog pixel information from detection cells a' to a' and detection cells b' to bN' are sequentially read out when read signals SPI and SP2 from 5 are applied;
A comparator 18 that performs a correlation calculation between analog image information from a pair of video lines 16 and 17, and a judgment that determines whether or not this gives the maximum correlation based on the correlation calculation result from the comparator 18. 19, and a scanning number cumulative counter 20 that counts the number of times the detection cells a' to aN' are scanned until the maximum correlation is given, and detects the shift amount p.

The comparator 18 is composed of a differential amplifier 21 and an integrator 22. The differential amplifier 21 determines the difference between the analog image information from the video line 16 and the analog pixel information from the video line 17, for example, the difference between the analog pixel information in the detection cell a1' and the analog pixel information in the detection cell b1'. The signal is then sent to an integrator 22.

The integrator 22 includes detection cells a' to aH'.

The integrated value of the absolute value of the difference between the analog pixel information from the video line 16 and the analog pixel information from the video line 17, which are sequentially sent from the differential amplifier 21 by scanning b' to bN', and the Code SN indicating whether the analog pixel information is shifted to the right or left with respect to the analog pixel information from the video line 16
This outputs the following. The integral value of the absolute value of the difference becomes the correlation calculation result CR of one scan and is sent to the determiner 19.

The determiner 19 receives the correlation calculation result C sent from the integrator 22.
R is compared with the correlation reference value, and when the correlation calculation result CR is smaller than the correlation reference value, it is determined that the correlation between the image information of the light flux F1 and the image information of the light flux F2 has reached the maximum at that point, Shift register 14.15 and external circuit (
(not shown) outputs a determination end signal DED.

Note that the integrator 22 may be an analog type such as a Miller amplifier, or may be a digital type constituted by an A/D converter or an adder.
Further, the determiner 19 may be an analog type comparator or a digital type comparator depending on the output format of the integrator 22.

Shift registers 14 and 15 each have N stages, (N+1)
The shift register 14.15 receives input signals SPI' and SP2' from the OR circuit 23.24.
are added to each. shift register 1
4.15 is the logical sum circuit 23.
The input signals SPI' and SP2' from 24 are transferred to the respective transfer stages, and are used as read signals sp1. These are sequentially added to the corresponding detection cells as sp2.

In the shift register 14, the input signal SPI' is all transferred to the N transfer stages, and the detection cell a1 of the solid-state image sensor 12 is
When the scanning from ' to aH' is completed, a scanning end signal OTI is output. On the other hand, the shift register 15 outputs the scanning end signal OT2 when the input signal SP2' is transferred to the (N-1) transfer stage, and also outputs the scanning end signal OT2 when the input signal SP2' is transferred to the (N-1) transfer stage.
is transferred to the (N+1) transfer stage, the scanning signal O
It is designed to output T3.

Since the OR circuit 23 supplies the OR of the scan start signal STT and the scan end signal OTI of the shift register 14 to the shift register 14 as an input signal SPI', the shift register 14 performs scanning according to the scan start signal STT. When the first scan is completed, the next scan starts again with the scan end signal OTI, and so on until the determination end signal DEC is applied.
′ is scanned repeatedly.

Scan end signal OT2 of shift register 15. OT3 is
It is added to the switching circuit 25, and one of them is selected based on the sign SN of the integrator 22. The scanning end signals OT3 and OT2 are applied to one terminal of the NAND circuits 26 and 27 of the switching circuit 25, respectively, and the code SN and the signal SN inverted by the inverter 28 are applied to the other terminals of the NAND circuits 26 and 27. The code SN is added to each. Also, the NAND circuit 26.
The output signals NAI and NA2 of 27 are applied to a NAND circuit 29. The output signal N of the NAND circuit 29 is input to the OR port tIB24 together with the scan start signal STT.
Since A3 is added, shift register 1
5 starts scanning in response to a scanning start signal STT, and when the first scanning is completed, the NANDI circuit 29
Detection cells a1' to aN' are repeatedly scanned until the determination end signal DED is applied, such that scanning is restarted by the output signal NA3 of . The output signal NA3 of the NAND circuit 29 is set to "1" by the output signal OT3 from the (N+1) stage of the shift register 15 when the code SN from the integrator 22 is "1", and becomes "1" when the code SN is "0". , becomes 1'' by the output signal OT2 from the (N-1) stage of the shift register 15. As a result, the code SN
is "1'", the read signal SP2 from the shift register 15 is set to 1' in the shift register 1 compared to the previous scan.
The read signal SPI from shift register 15 is output one stage later than the read signal SPI from shift register 15.
The read signal SP2 from the shift register 14 can be output one step earlier than the read signal SPI from the shift register 14 compared to the previous scan. Furthermore, the shift registers 14 and 15 are configured to end their operations upon receiving the operation end signal AED from the OR circuit 30.

The OR circuit 30 includes a reset signal R3T and a determiner 19.
The judgment end signal DED from A is added to the operation end signal A.
ED is a signal obtained by calculating the logical sum of the reset signal R9T and the determination end signal DED. As a result, the shift registers 14 and 15 end their operations in response to either the reset signal R9T or the determination end signal DEC.

The scanning number cumulative counter 20 counts the scanning end signal OTI from the shift register 14, calculates the number of scannings until the judgment end signal DED is outputted, and divides this into the image of the light flux F1 and the image of the light flux F2. The deviation amount p is output from the output terminal OUT to an external circuit (not shown). An upper limit value of the number of scans is preset in the cumulative number of scans counter 20, and when the count value of the scan end signal OTI becomes larger than this upper limit value, the determination unit 1
In step 9, it is determined that there is no correlation calculation result smaller than the correlation reference value and the image of the light flux F1 and the image of the light flux F2 cannot be correlated, and an undeterminable signal DEBL is output from the output terminal 0UTI to the external circuit. It looks like this.

Note that the scanning number cumulative counter 20 receives a reset signal R8.
It is cleared by T.

The operation of the distance measuring device 11 having such a configuration will be explained using the time charts shown in FIGS. 3 and 8.

Note that FIG. 3 is a time chart showing the operation of the distance measuring device 11 when the image information of the light beam F2 is shifted by a distance X to the left with respect to the image information of the light beam F1, as shown in FIG. .

First, a distance detection operation when the image information of the light beam F2 is shifted leftward by a distance X with respect to the image information of the light beam F1 will be described.

Before starting the distance detection operation, the solid-state image sensors 12, 13 . Shift register 14.15. In order to clear and initialize the scanning number cumulative counter 20, etc., a reset signal R9T as shown in FIG. 3(a) is applied to the distance measuring device 11.

Next, 1 and N are applied to each detection cell a' to a' and detection cell b' to bN' of the solid-state image sensor 12.13.
1. In order to store the image information of the light beams Fl and F2, the information storage signal DT is sent to all the detection cells a1' to aH' and the detection cells b' to bN' at the timing shown in FIG. 3(b). Add at the same time. As a result, predetermined pixel information is stored in the capacitive element of each detection cell as shown in FIG. 1, and is stored and retained at the end of the information storage period. Figures 5(a) and (b) show the fourth
When the light fluxes Fl and F2 as shown in Figures (a) and (b) are incident on the solid-state image sensor 12°13, the detection element a1'
to a) 4'.

This figure shows analog pixel information stored and held in detection elements b1' to bN'.

In this way, all detection cells a' to a', b' to 1 N
After predetermined pixel information is stored and stored in 1 bN', a scan start signal SST is given to the distance measuring device 11. This scan start signal STT is inputted to the input signals SPI' and SPI' of the shift register 4.15 via the OR circuits 23 and 24.
It becomes SP2'. Input signal s based on scan start signal STT
When pi' and sp2' are added to the shift register 4.15, each detection cell a1 of the solid-state image sensor 12.13
Reading of pixel information ' to a ' and b ' to bN', that is, scanning starts.

In FIG. 3, phase φ1 means the first scan, and phase φp means the p-th scan. Scan start signal S
When the input signals SPI' and SP2' from TT are applied to the shift register 4.15, scanning of phase φ1 is performed as shown in FIGS. 3(c) and 3(d). In phase φ1, the input signals SP1' and SP2' are input to the shift register 4.15 at the same read timing, and are sequentially sent to each transfer stage of the shift register 4.15 in synchronization with the basic clock CK, and the respective read signals SPI , SP2
Detection cells a' to a of the solid-state image sensor 12.13 as
', N accesses are made to the detected cells i' to btt'. Read signal SP from shift register 4
5 (a) from detection cells a' to aH'.
> As shown in FIG.
2, from the detection cells b' to bN', as shown in FIG.
) is sequentially read out to the video line 17.

The pixel information sequentially read out to the video lines 16 and 17 is applied to the differential amplifier 21 of the ratio amplifier 18, and the difference is calculated and input to the integrator 22. Figure 5(C) is Figure 5(a)
, when the pixel information of (b) is input, the differential amplifier 21
This shows difference information sequentially extracted from . Figure 5 (C
) is applied to the integrator 22, the integrator 22 integrates the difference information as shown in FIG.
), the integral value of the absolute value of the difference information is “22.6”
becomes.

Furthermore, the difference information input to the integrator 22 initially has a positive value and then becomes a negative value, so that
It is determined that the image information of the light beam F2 shown in FIG. 4(b) is shifted to the left with respect to the image information of the light beam F1, and "1" is output as the code SN.

The integral value of the absolute value of the difference information from the integrator 22 becomes the correlation calculation result CR for one scan, and is compared with the correlation reference value in the determiner 19. Now, the correlation standard value is, for example, “5.0”.
”, the correlation calculation result “22°6” in phase φ1 is larger than the correlation reference value, so the determiner 19 uses the pixel information of detection cells a ′ to aN′ and the detection cells b 1 ′ to It is determined that the pixel information of bN' does not give the maximum correlation, and a determination end signal DEC is issued.
is not output.

As a result, the operation of the shift register 4.15 continues, and the shift register 4 outputs the scan end signal OTI at the time when the transfer of N stages is completed, and this scan end signal OT
I is the second input signal SPI through the OR circuit 23
' becomes. This input signal SPI' is input to the shift register 4.
The second readout, ie, scanning, phase starts from the time when the input signal is input. Note that at this time, the integrator 22 is cleared. On the other hand, the shift register 5 outputs the scanning end signal OT2 when the transfer of (N-1) stages is completed, and
+1) When the transfer of the stage is completed, the scan end signal OT3 is output. These scanning end signals OT2. OT3 joins the switching circuit 25 and selects either the scan end signal OT2 or OT3 according to the sign SN from the integrator 22. In this case, since the code SN is "1", the scan end signal OT3 is selected and outputted from the NAND circuit 29 as the output signal NA3. This is the OR circuit 24
The second input signal SP to the shift register 5 via
It becomes 2'.

As shown in FIGS. 3(c) and 3(d), the second input signal SP2' is input to the shift register 15 with a delay of one stage relative to the second input signal SPI' of the shift register 4. In the next phase, detection cells a' to a
1 N for ',b' to bN'
1, reading or scanning is performed as shown in FIGS. 6(a) and 6(b). That is, Fig. 5(a), (
Compared to the scanning shown in b), the amount of deviation between pixel information is “1”
scanned in such a way as to decrease the height. Note that the solid-state image sensor 1
2.13 detection cells a' to aH'.

Since the detection cells b' to bN' have the structure shown in FIG. 1, the pixel information stored in them will not be lost even if readout or scanning is performed again.The detection cells a' to a' , b1' to N to bN' are scanned for the second time using the read signals SPI and SP2 from the shift register 4.15.
Pixel information as shown in FIGS. 6(a) and 6(b) is sequentially input to the differential amplifier 21 via video lines 16 and 17, respectively. The difference amplifier 21 sequentially sends difference information as shown in FIG. 6(C) to the integrator 22.
outputs the refined value of the absolute value of the difference information, that is, the correlation calculation result CR, and the code SN, as described above.

The second correlation calculation result using the difference information in Figure 6(C) is also
The result of the first correlation calculation is "22.6", which is larger than the correlation reference value "5.0", so the determination end signal DED is not output from the determiner 19, and the shift register 4
．． 15 continues to operate.

In this way, when the p-th input signal SPI' is input to the shift register 4, a phase φp is entered in which the p-th scan is performed. Until the phase φP, the integrator 2
The code SN from 2 is "1", and the switching circuit 25
is the output signal O from the (N+1) stage of shift register 5
Since T3 was always selected, the input signal SP2' in phase φp is p with respect to the input signal SPI' of the second shift register 4, as shown in FIGS. 3(c) and (d).
It is input to shift register 5 with a stage delay. As a result, in phase φP, detection cells a' to aH', b
FIG. 7(a) for ' to bN'.

When pixel information as shown in FIGS. 7(a) and 7(b) is read out or scanned as shown in FIG. The amplifier 21 sequentially sends difference information as shown in FIG. 7(C) to the integrator 22, and the integrator 22 outputs the correlation calculation result CR and the code SN. The p-th 41'IrWJ calculation result based on the difference information in FIG. 7(C) is "0", which is smaller than the correlation reference value "5.0", so
It is determined that the maximum correlation is given at this time, and a determination end signal DED is outputted at the timing shown in FIG. 3(e).

This determination end signal DED is output to an external circuit (not shown) and is also applied to the shift register 14.15 as an operation end signal AED via the OR circuit 30, thereby controlling the operation of the shift register 14.15. finish.

At this time, the scanning number cumulative counter 20 counts the number of scannings "p" thus far, and this is outputted to the external circuit from the output terminal OUT as the deviation amount p. The external circuit is
By multiplying the deviation lp from the output terminal 0TJT by the interval between detection cells, calculate the actual deviation distance X between the image information of the light flux F1 and the image information of the light flux F2, and calculate the distance from this distance Detect distance.

Distance detection has been described in the case where the image information of the light flux F2 is shifted by a distance X to the left with respect to the image information of the light flux F1. Even when there is a deviation by wIX, the distance detection operation is performed in the same way.

FIG. 8 is a time chart showing the operation of the distance measuring device 11 when the distance is shifted to the right.

Figures 8(a) and (b) are respectively similar to Figure 3(a).
, is a time chart of the reset signal R9T and the information accumulation signal °DT, which are exactly the same as those in FIG. That is, before starting the distance detection operation, the solid-state image sensing devices 12, 13° shift registers 14, 15 . The scanning number cumulative counter 20 and the like are cleared, and predetermined pixel information is stored in the detection cells a' to a' and bN1' to bN' by the information storage signal DT shown in FIG. 8(b), and then stored. Hold.

Next, when the scan start signal STT is applied to the distance measuring device 11, this scan start signal STT is applied to the shift register 14.1.
5 input signals SPI' and SP2'. The shift register 14.15 uses the input signals -qsPi' and SP2' to control the detection cells a' to a' of the solid-state image sensor 12.13,
The reading or scanning of b1' to bN'N is started, and the detection cells a1' to a
', the pixel information from the detection cells b' to bN' is transferred from the video line 16.17 to the differential amplifier 21.17 of the comparator 18.
It is sent to the integrator 22. The integrator 22 outputs the correlation calculation result CR.
However, if the image information of the light beam F2 is shifted to the right with respect to the image information of the light beam F1, the differential amplifier 21 outputs the positive difference information after the negative difference information. , the code SN becomes "0".

As a result, the switching circuit 25 switches the shift register 15 (
The scanning end signal OT2 from the N-1> stage is selected and outputted as the output signal NA3 of the NAND circuit 29. This output signal NA3 is input to the shift register 15 via the OR circuit 24 as the second input signal SP2' of the shift register 15.

On the other hand, the shift register 14 outputs a scan end signal OTI when the transfer of N stages is completed, and the scan end signal OT1 is
It is input to the shift register 4 via the OR circuit 23 as the second input signal SPI' of the shift register 14. Therefore, the time chart of the input signal SPI' shown in FIG. 8(C) is exactly the same as that in FIG. 3(C).

FIG. 8(d) shows the input signal SP2 of the shift register 5.
Fig. 8(C) is a time chart of '.

(If it deviates to the right as seen from d>, the phase φ1......, as it progresses to φp,
The shift register 5 receives an input signal SP2' that is one stage earlier than the input signal SPI' of the shift register 4.

As a result, the deviation between the pixel information of the detection cells a' to aN' and the pixel information of the detection cells b1' to bN' decreases by one step as the phase progresses, and there is no deviation in the phase φp. is determined to give the maximum correlation at this time. Then, the determination end signal DED is output from the determiner 19, and the shift register 4
．． 15 is completed, the number of scans, that is, the deviation ip is output from the output terminal OUT of the cumulative number of scans counter 20 to an external circuit. In this way, even if the image information of the light beam F2 is shifted to the right by a distance Then, by multiplying the deviation p from the output terminal OUT by the interval between detection cells, the luminous flux F
The distance X of the actual deviation between the image information of F1 and the image information of the light beam F2 can be calculated, and the distance to the object to be measured can be detected from this distance X.

Note that instead of calculating the actual deviation distance X and detecting the distance to the object to be measured using an external circuit, a multiplier is provided in the distance measuring device 11 shown in FIG. Although not shown, it is assumed that the integrator 22 is cleared before the start of each phase.

As described above, according to the distance measuring device 11 of this embodiment, the solid-state image sensor 12.13 includes detection cells a' to a' and bl' to NbN' having the same structure as shown in FIG. As a result, the same pixel information can be read out again from each detection cell a' to a', bl' to NbN' of the solid-state image sensor 12, 13, and the same pixel information can be read out again from each detection cell a' to a' to NbN' of the solid-state image sensor 12, 13. Shift registers 10 and 111 serve as intermediate buffers for correlation calculation processing.
Even without providing the detection cells a' to a', b1' to bn', it is possible to detect the amount of shift p by performing a correlation calculation based on the pixel information N directly read out from each detection cell a' to a' and b1' to bn'. Furthermore, each detection cell a′
Since analog correlation calculation can be performed based on the analog pixel information from aH' to aH', highly reliable correlation calculation results can be obtained. This makes it possible to detect the deviation amount p more accurately and at high speed.

Further, in the distance measuring device 11 of this embodiment, the scanning end signals OT1, OT2, . O'r
3 are used as the input signals SPI' and SP2' of the pair of shift registers 14.15, it is possible to significantly simplify the timing control of the readout signals SPI and SP2 of the solid-state image sensor 12.13 and to reduce the circuit scale. can. Furthermore, the process can be completed at the time when the correlation of Yoshidai is given (two scans) without obtaining N correlation calculation results. This makes it possible to reduce the size and speed of the distance measuring device.

Note that in the above embodiment, the pair of solid-state image sensors 12.1
Although a pair of shift registers 14 and 15 and a pair of video lines 16 and 17 are provided for each of the three solid-state image sensors 12 and 13, the video lines of the pair of solid-state image sensors 12 and 13 may be shared. One shift register scans one of the pair of solid-state image sensors 12 and 13 only once, and pixel information at this time is stored in a memory (not shown). After that, a pair of solid-state image sensors 12゜13
scan the other one. The method of scanning at this time is, for example, by shifting the readout timing for each number of scans, like the shift register 15 described above.In this way, the other side is scanned a predetermined number of times, and the pixel information for each scan is stored in the memory. The deviation amount p may be detected by performing a correlation calculation with the stored pixel information from one of the solid-state image sensors and detecting the one that gives the maximum correlation. Instead of determining whether to give or not, it scans all the data for a predetermined number of scanning days (N times), as with conventional devices,
After all pixel information is stored in memory, the one that provides the greatest correlation may be detected.

〔Effect of the invention〕

As explained above, according to the solid-state image sensor of the first invention, the terminal voltage value of the capacitive element is stored and held as pixel information, and the terminal voltage value of the capacitive element is output as pixel information via the current amplification circuit. Therefore, pixel information can be read out again from this detection cell, and image correlation calculations, filtering processing, etc. can be performed directly based on the output signal of the detection cell.

Further, according to the distance measuring device of the second invention using the solid-state image sensor of the first invention, correlation calculation between analog pixel information read out from a pair of solid-state image sensors by scanning of the readout means is directly performed. Therefore, with a simple circuit configuration, the amount of deviation between images can be detected quickly and accurately, and the distance to the object to be measured can be detected.

[Brief explanation of the drawing]

FIG. 1 is a configuration diagram of a detection cell of a solid-state image sensor showing an embodiment of the present invention, FIG. 2 is a configuration diagram of a distance measuring device using the solid-state image sensor shown in FIG. The image is the luminous flux F
1 is a time chart showing the operation of the distance measuring device shown in FIG. Signal DT, input signal SPI', input signal SP2'
, a time chart showing the determination end signal DED, FIGS. 4(a) and 4(b) are diagrams showing images of the luminous fluxes Fl and F2, respectively, and FIGS. 5(a) to (C) are each the phase φ
Figures 6(a) to 6(C) show the pixel information of the first solid-state image sensor, the pixel information of the second solid-state image sensor, and the difference information in 1. 7(a) to (C) are diagrams showing pixel information of the first solid-state image sensor, pixel information of the second solid-state image sensor, and difference information in phase φp, respectively.
Figure 8 shows the operation of the distance measuring device shown in Figure 2 when the image of the light flux F2 is shifted to the right with respect to the image of the light flux F1. FIGS. 8(a) to 8(e) are time charts, respectively.
Reset signal R3T, information storage signal DT, input signal sp
i', an input signal SP2', and a time chart showing the determination end signal DED; FIG. 9 is a configuration diagram of a detection cell of a conventional solid-state image sensor; FIG. 10 is a configuration diagram of a conventional distance measuring device;
Figure 1(a) shows the light flux F3. which enters a conventional distance measuring device. F4
FIG. 11(b) is a diagram showing analog pixel information read out from a solid-state image sensor of a conventional distance measuring device.
Figure (C) shows the analog pixel information shown in Figure 11 (b).
Figures 12 (a) to (C) showing digital pixel information converted into values are pixel information b to bH and pixel information a, respectively.
l to aH, diagram showing the first exclusive OR result, 1st
3 (a) to (C) are pixel information b to b, respectively.
9 pixel information a to aE1-1+H1pI
HD Figures 14 (a) to (C) showing the exclusive OR results for the first time are pixel information b to bH and pixel information a, respectively.
9 to a, (N-M+1)th exclusion N-841N Alternative OR results are not shown. 1...Detection cell, 2...Photodiode, 3...
・Constant current source, 11... Distance measuring device, 12.13... Solid-state image sensor, 14.15... Shift register, 16.17... Video line, 18... Comparator, 1
9... Determiner, 20... Scanning number cumulative counter, 2
1... Differential amplifier, 22... Integrator, 23.24.
...OR circuit, 25...switching circuit, FTI, FT2
．． FT3. FT4. FT5・MOS+-Rangister Patent Applicant Hamamatsu Photonics Co., Ltd. Agent Patent Attorney Masaharu Uemoto

Claims (1)

[Claims] 1) A photodiode on which light enters, a capacitive element, a switching element that makes the photodiode and the capacitive element conductive during an information storage period, and outputs the terminal voltage value of the capacitive element as pixel information. What is claimed is: 1. A solid-state imaging device comprising a detection cell having a current amplification circuit that 2) The solid-state imaging device according to claim 1, wherein the photodiode is connected to a switching element for causing a photocurrent generated outside the information storage period to flow to the outside. 3) The terminal voltage value of the capacitive element is initially set to a predetermined reference potential.
The solid-state image sensor described in . 4) The solid-state imaging device according to claim 1, wherein the current amplification circuit is further connected to a video line via a switching element that is turned on when reading pixel information. 5) A pair of solid-state image sensors in which a plurality of detection cells that store and retain pixel information in a readable manner are arranged, and two light beams from the same object that have passed through spatially different locations are incident on each of the solid-state image sensors; a readout means for scanning the detection cells of the pair of solid-state image sensors by shifting readout timings of the detection cells of the solid-state image sensors for each scan; What is claimed is: 1. A distance measuring device comprising: detecting means for detecting a deviation between pixel information by performing a correlation calculation for each scan and detecting the one that gives the largest correlation among the correlation calculation results. 6) The distance measuring device according to claim 5, wherein a previous scan end signal is inputted to the readout means as an input signal for the next scan. 7) The readout means includes a pair of shift registers corresponding to the pair of solid-state image pickup devices, and the pair of shift registers read out the detection cells of the pair of solid-state image pickup devices simultaneously by shifting readout timings of the detection cells for each scan. 6. The distance measuring device according to claim 5, wherein the distance measuring device is configured to emit a distance. 8) The reading means consists of one shift register,
The shift register first reads one detection cell of the pair of solid-state image sensors, and then reads the other detection cell of the pair of solid-state image sensors by shifting the readout timing for each scan. A distance measuring device according to claim 5, characterized in that: 9) The detection means includes a comparator that performs a correlation calculation result between the pixel information read out from the pair of solid-state image sensors, and a comparator that provides the maximum correlation among the correlation calculation results for each scan from the comparator. The distance measuring device according to claim 5, further comprising a determining device for detecting. 10) The comparator performs a correlation calculation between pixel information read simultaneously from the pair of solid-state image sensors for each scan,
outputting a sign indicating in which direction the pixel information stored and held in one of the pair of solid-state image sensors deviates from the pixel information stored and held in the other solid-state image sensor;
10. The distance measuring device according to claim 9, wherein the direction in which the readout timing is shifted is determined by this code. 11) The detection means is characterized by comprising a scanning number cumulative counter that detects the number of scanning up to that point as the amount of deviation between pixel information when it is determined by the determining device that the maximum correlation is given. A distance measuring device according to claim 9. 12) The comparator is configured to read pixel information first read from one of the pair of solid-state image sensors, and then to read pixel information shifted from the other of the pair of solid-state image sensors for each scan. 10. The distance measuring device according to claim 9, wherein the determining device detects an amount of deviation from the correlation calculation result until the maximum correlation is given.