Image Sensor

The image sensor addresses noise-induced horizontal stripes by adjusting ramp signal slopes and gain based on digital code ranges, improving ISO sensitivity and image quality in low-light conditions.

JP7593897B2Active Publication Date: 2024-12-03RENESAS ELECTRONICS CORP
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
JP2021119572
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Filing Date
2021-07-20
Publication Date
2024-12-03
Estimated Expiration
2041-07-20

AI Technical Summary

Technical Problem

Existing image sensors suffer from noise superimposed on ramp signals, particularly thermal and flicker noise, which causes horizontal stripes in low-light conditions, limiting ISO sensitivity.

Method used

The image sensor employs a reference signal generation circuit that generates ramps with different slopes in response to digital code ranges, using an attenuator to suppress noise by adjusting gain based on the digital code range, and a comparison circuit to latch the digital code value when the signals match.

Benefits of technology

This approach reduces noise, improving ISO sensitivity and minimizing the visibility of horizontal lines, thereby enhancing image quality in low-light conditions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007593897000001
    Figure 0007593897000001
  • Figure 0007593897000002
    Figure 0007593897000002
  • Figure 0007593897000003
    Figure 0007593897000003
Patent Text Reader

Abstract

To provide an image sensor capable of reducing noise to increase ISO sensitivity.SOLUTION: An image sensor comprises a pixel allay PXA comprising a plurality of imaging elements, and an ADC circuit 1 to which the pixel allay PXA supplies pixel data in parallel. The ADC circuit 1 comprises a reference signal generating circuit for outputting reference signals SD_0 to SD_m corresponding to digital codes DCD, and a comparison circuit CP for comparing the reference signals SD_0 to SD_m with the pixel data. The comparison circuit outputs a digital code DCD of which the reference signals SD_0 to SD_m meet the pixel data. The reference signal generating circuit comprises a ramp signal generating circuit RAMP_G for outputting a first ramp signal RAMP when the digital code DCD is included in a first range, and outputting a second ramp signal RAMP having a gradient different from that of the first ramp signal RAMP with respect to the variation of the digital code DCD when the digital code DCD is included in a second range, and an attenuator ATT to which a ramp signal RAMP is supplied and which outputs the reference signals SD_0 to SD_m having the identical gradient in the first range and the second range.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to an image sensor, for example, an image sensor that includes a pixel array in which a plurality of imaging elements are arranged in an array and an analog / digital conversion (hereinafter also referred to as ADC) circuit, and that converts pixel data from the pixel array into a digital signal and outputs it. [Background technology]

[0002] An image sensor (solid-state image sensor) including an image sensor corresponding to a pixel array, an ADC circuit, and a ramp generator is described in, for example, Patent Document 1. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2014-90325 A Summary of the Invention [Problem to be solved by the invention]

[0004] As an ADC circuit that converts pixel data output from a pixel array into a digital signal, there is one that compares a ramp signal that changes according to a digital code with pixel data, and outputs the value of the digital code when the ramp signal and the pixel data, for example, match as a digital signal corresponding to the pixel data. The inventors' investigations revealed that when noise of, for example, several tens of μV, such as thermal noise or flicker noise, is superimposed on the ramp signal, it may be recognized as a horizontal stripe when the digital signal converted by the ADC circuit is displayed on a screen.

[0005] The human eye is more sensitive to dim light than to bright light, so when pixel data captured under dim light is converted into a digital signal, if noise is superimposed on the lamp signal, horizontal stripes on the screen become noticeable, limiting the improvement of the ISO sensitivity.

[0006] Patent Document 1 does not mention that horizontal stripes become noticeable due to noise superimposed on a ramp signal. [Means for solving the problem]

[0007] A brief outline of a representative embodiment of the present invention will be described below.

[0008] The image sensor includes a pixel array having a plurality of imaging elements, and an ADC circuit to which pixel data is supplied in parallel from the pixel array. The ADC circuit includes a reference signal generation circuit that outputs a reference signal according to a digital code, and a comparison circuit that compares the reference signal with pixel data, and outputs a digital code value when the reference signal and the pixel data have a predetermined relationship. Here, the reference signal generation circuit includes a ramp signal generation circuit that outputs ramp signals with different slopes in response to a change in the digital code when the value of the digital code is in a first range and when it is in a second range different from the first range, and an attenuator that receives the ramp signal and outputs a reference signal with the same slope in the first range and the second range.

[0009] Other objects and novel features will become apparent from the description of this specification and the accompanying drawings. Effect of the Invention

[0010] According to one embodiment, it is possible to provide an image sensor capable of improving ISO sensitivity by reducing noise. [Brief description of the drawings]

[0011] [Figure 1] 1 is a block diagram showing a configuration of an image sensor according to a first embodiment. [Diagram 2] FIG. 2 is a diagram for explaining the operation of the image sensor according to the first embodiment. [Diagram 3] FIG. 2 is a diagram for explaining the operation of the image sensor according to the first embodiment. [Figure 4] 5A to 5C are waveform diagrams illustrating the operation of the image sensor according to the first embodiment. [Diagram 5] 1A to 1C are diagrams illustrating a ramp signal generating circuit according to a first embodiment. [Figure 6] FIG. 13 is a diagram for explaining an image sensor according to a modified example of the first embodiment. [Figure 7] 13(A) to 13(C) are waveform diagrams illustrating an image sensor according to a modification of the first embodiment. [Figure 8] 13A and 13B are waveform diagrams for explaining a problem that occurs in the image sensor according to the first embodiment. [Figure 9] FIG. 11 is a block diagram showing a configuration of an image sensor according to a second embodiment. [Figure 10] FIG. 11 is a diagram for explaining a computing unit according to a second embodiment. [Figure 11] 13A to 13C are waveform diagrams illustrating the operation of the image sensor according to the second embodiment. [Figure 12] FIG. 11 is a block diagram showing a configuration of an image sensor according to a third embodiment. [Figure 13] 13A and 13B are waveform diagrams illustrating the operation of the image sensor according to the third embodiment. [Figure 14] FIG. 13 is a block diagram showing a configuration of an image sensor according to a fourth embodiment. [Figure 15] 13(A) and 13(B) are waveform diagrams illustrating the operation of the image sensor according to the fourth embodiment. [Figure 16] FIG. 1 is a schematic plan view of an image sensor according to a first embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] Hereinafter, the embodiments of the present invention will be described with reference to the drawings. Note that the disclosure is merely an example, and those skilled in the art can easily come up with appropriate modifications that maintain the gist of the invention and are naturally included in the scope of the present invention.

[0013] In addition, in this specification and each figure, elements similar to those described above with respect to the previous figures are given the same reference numerals, and detailed descriptions may be omitted as appropriate.

[0014] In the following description, an image sensor in which a pixel array and an ADC circuit are arranged on one semiconductor chip will be described as an example, but the present invention is not limited to this. For example, the pixel array and the ADC circuit may be arranged on different semiconductor chips.

[0015] (Embodiment 1) <Overall composition> Fig. 16 is a schematic plan view of an image sensor according to the first embodiment. The image sensor 100 is composed of a plurality of circuit blocks arranged on one semiconductor chip, but only the circuit blocks necessary for explanation are depicted in the figure. In Fig. 16, Pix indicates an imaging element that captures an image. A plurality of imaging elements Pix are arranged in an array (matrix) to form a pixel array PXA.

[0016] Although not particularly limited, in the pixel array PXA, multiple imaging elements arranged in the same column (pixel column) are connected to a pixel data line, and multiple imaging elements arranged in the same row (pixel row) are selected simultaneously during readout. In Fig. 16, the pixel rows are exemplified as PIL_0 to PIL_n, and the pixel data lines are exemplified as PID_0 to PID_m.

[0017] The pixel data lines PID_0 to PID_m of the pixel array PXA are connected to the ADC circuit 1. When reading the pixel array PXA, for example, as shown in FIG. 16, pixel data is sequentially read from a plurality of imaging elements Pix arranged in the same pixel row to the pixel data lines PID_0 to PID_m in the direction from the pixel row PIL_0 to PIL_n. The read pixel data is converted into a corresponding digital signal by the ADC circuit 1 and output from the ADC circuit 1 as digital pixel data DPI.

[0018] <Configuration of ADC Circuit> FIG. 1 is a block diagram showing the configuration of an image sensor according to Embodiment 1. In FIG. 1, the pixel array PXA operates with a first voltage V1 as an operating voltage, and the ADC circuit 1 operates with a second voltage V2 having an absolute value lower than the first voltage V1 as an operating voltage. The first voltage V1 is, for example, 5 (V), and the second voltage V2 is, for example, 3.3 (V).

[0019] The imaging element Pix (FIG. 16) outputs pixel data of, for example, 2 (V) when imaging the darkest light, and outputs pixel data of, for example, 1 (V) when imaging the brightest light. That is, the value of the pixel data changes between 1 (V) and 2 (V) according to the luminance of the imaged light.

[0020] By reading, pixel data of 1 (V) to 2 (V) is supplied to the ADC circuit 1 in parallel from the pixel array PXA via the pixel data lines PID_0 to PID_m.

[0021] The ADC circuit 1 includes a ramp signal generation circuit (generation circuit) RAMP_G that generates a ramp signal RAMP, an attenuator ATT, a control circuit CNT, a comparison circuit CP, a bias circuit BSC, a latch circuit LTC, and a code generation circuit DCD_G that generates a digital code DCD.

[0022] The ramp signal generating circuit RAMP_G includes a digital / analog conversion (hereinafter also referred to as DAC) circuit 2 and a voltage follower circuit 3. A digital code DCD is supplied to the DAC circuit 2, which converts the value of the digital code DCD (digital code value) into an analog signal. The converted analog signal is supplied to the voltage follower circuit 3 constituted by an arithmetic unit. An output of the voltage follower circuit 3 is supplied to a ramp signal line RAMP_L as a ramp signal RAMP.

[0023] The ramp signal generating circuit RAMP_G operates using the second voltage V2 as an operating voltage. The digital code DCD is generated by the code generating circuit DCD_G. In the first embodiment, the code generating circuit DCD_G is configured by a 12-bit counter, and the digital code DCD is configured by a 12-bit binary signal. The 12-bit counter counts up or down over time, so that the value of the digital code DCD changes over time. When the value of the digital code DCD is expressed in decimal, the value of the digital code DCD changes, for example, from 0 to 4095 over time.

[0024] As the value of the digital code DCD changes (increases or decreases) over time, the voltage of the analog signal output from the DAC circuit 2 also changes, and the voltage of the ramp signal RAMP also changes over time. In the first embodiment, the DAC circuit 2 is controlled by a ramp control signal RAMP_C from the control circuit CNT. As will be described later with reference to the drawings, in the DAC circuit 2, the reference current is switched by the ramp control signal RAMP_C. By switching the reference current, the slope of the ramp signal RAMP with respect to the digital code DCD is changed.

[0025] The ramp signal RAMP is supplied to the comparison circuit CP as a reference signal SD (SD_0 to SD_m) via the attenuator ATT. The gain of the attenuator ATT is switched by an attenuator control signal ATT_C from the control circuit CNT. That is, the attenuator ATT is an attenuator that attenuates the ramp signal RAMP according to the gain indicated by the attenuator control signal ATT_C. When the control circuit CNT switches the slope of the ramp signal RAMP by the ramp control signal RAMP_C, it also switches the gain of the attenuator ATT by the attenuator control signal ATT_C.

[0026] In the first embodiment, the attenuator ATT is composed of a plurality of unit attenuators AT_0 to AT_m corresponding to the pixel data lines PID_0 to PID_m. The comparison circuit CP is also composed of a plurality of comparators CP_0 to CP_m corresponding to the pixel data lines PID_0 to PID_m. Taking the pixel data line PID_0 as an example, a ramp signal RAMP is supplied from the corresponding unit attenuator AT_0 to the corresponding comparator CP_0 as a reference signal SD_0. The comparator CP_0 compares the supplied reference signal SD_0 with the pixel data supplied via the corresponding pixel data line PID_0. The other unit attenuators AT_1 to AT_m and comparators CP_1 to CP_m are similar to the unit attenuator AT_0 and comparator CP_0.

[0027] A predetermined bias voltage is supplied to the comparators CP_0 to CP_m from a bias circuit BSC. Taking the comparator CP_0 as an example, the configuration of the comparator will be described. The comparator includes capacitive elements C1 to C4, a differential amplifier A1, a comparator A2, and a Schmitt trigger circuit ST. A reference signal SD_0 and pixel data are supplied to the differential amplifier A1 via capacitive elements C1 and C2 and amplified. The output of the differential amplifier A1 is supplied to the comparator A2 via capacitive elements C3 and C4, and the voltage of the reference signal SD_0 is compared with the voltage of the pixel data. The output of the comparator A2 is supplied to the Schmitt trigger circuit ST. When the voltage of the reference signal SD_0 and the voltage of the pixel data have a predetermined relationship, for example, when they match, a latch trigger (trigger) signal LT_C is output from the Schmitt trigger circuit ST to the latch circuit LTC.

[0028] The latch circuit LTC is also composed of a number of unit latch circuits LT_0 to LT_m corresponding to the pixel data lines PID_0 to PID_m. The digital code DCD output from the 12-bit counter constituting the code generating circuit DCD_G is supplied to each of the unit latch circuits LT_0 to LT_m via 12 signal lines, and in response to a trigger signal LT_C from the corresponding comparator, the unit latch circuit latches (holds) the value of the digital code DCD at that time. Taking the unit latch circuit LT_0 as an example, when the trigger signal LT_C is supplied from the corresponding comparator CP_0, this is used as a trigger to latch the value of the digital code DCD supplied at that time. The other unit latch circuits LT_1 to LT_m are similar to the unit latch circuit LT_0.

[0029] With the above configuration, the reference signals SD_0 to SD_m are compared with the pixel data on the pixel data lines PID_0 to PID_m, and when they match, the value of the digital code DCD at that time is latched in the unit latch circuits LT_0 to LT_m. The value of the digital code DCD latched in the latch circuit LTC is output as digital pixel data DPI (FIG. 16) corresponding to the pixel data.

[0030] In Fig. 1, reference signals SD_0 to SD_m according to the digital code signal DCD are generated by a ramp signal generating circuit RAMP_G, an attenuator ATT, and a control circuit CNT. Therefore, it can be considered that a reference signal generating circuit that generates a reference signal is composed of the ramp signal generating circuit RAMP_G, the attenuator ATT, and the control circuit CNT. Also, the comparators CP_0 to CP_m can be considered as 1-bit ADC circuits. Note that the configuration of the comparators CP_0 to CP_m shown in Fig. 1 is an example, and is not limited to this configuration.

[0031] <Operation of the reference signal generation circuit> Next, the operation of the reference signal generating circuit will be described with reference to the drawings. Figures 2 to 4 are diagrams for explaining the operation of the image sensor according to the first embodiment.

[0032] Fig. 2 shows the ramp signal generating circuit RAMP_G, the control circuit CNT, and the unit attenuator AT_0 and comparator CP_0 corresponding to the pixel data line PID_0 from the configuration shown in Fig. 1. Here, the operation will be explained using the unit attenuator AT_0 and comparator CP_0 corresponding to the pixel data line PID_0 as an example, but the operation is similar for the unit attenuators and comparators corresponding to the other pixel data lines. In Fig. 2, SD_G shows the reference signal generating circuit when it is considered that the reference signal generating circuit is composed of the ramp signal generating circuit RAMP_G, the attenuator ATT, and the control circuit CNT.

[0033] In FIG. 2, the unit attenuator AT_0 is composed of four capacitance elements CD1 to CD4 connected in series between the ramp signal line RAMP_L (FIG. 1) and a predetermined voltage. The four capacitance elements CD1 to CD4 are set to the same value. The voltage of the ramp signal RAMP is divided by the four capacitance elements, and the gain is determined depending on which capacitance element the reference signal SD_0 is extracted from. Here, the predetermined voltage corresponds to the voltage value (2(V)) when the voltage of the pixel data output from the pixel array PXA is the darkest. The unit attenuator AT_0 is composed of capacitance elements, and can be regarded as a voltage dividing circuit that divides the difference voltage between the ramp signal RAMP and the predetermined voltage.

[0034] The unit attenuator AT_0 shown in FIG. 2 can set its gain to 1 or 1 / 4 by the attenuator control signal ATT_C. That is, by turning on the switch DW1 by the attenuator control signal ATT_C, the gain can be set to 1, and by turning on the switch DW2 by the attenuator control signal ATT_C, the gain can be set to 1 / 4. The voltage difference between the voltage of the ramp signal RAMP on the ramp signal line RAMP_L and a predetermined voltage (2(V)) is multiplied by the set gain, the divided voltage calculated by the multiplication is subtracted from the predetermined voltage, and the voltage value of the subtraction result is output from the attenuator AT_0. For example, as shown in FIG. 2, when the gain is set to 1 / 4, the difference voltage between the voltage of the ramp signal RAMP and the predetermined voltage is multiplied by 1 / 4, the voltage resulting from the multiplication is subtracted from 2(V), and the voltage resulting from the subtraction is output from the unit attenuator AT_0.

[0035] 2, NZ denotes noise superimposed on the ramp signal RAMP output from the ramp signal generating circuit RAMP_G. As described above, this noise NZ is generated due to, for example, thermal noise, flicker noise, etc., and fluctuates up and down over time, for example, with respect to the ground voltage, as shown in FIG.

[0036] In Fig. 2, the reference current of the DAC circuit 2 is switched to 1x or 4x by the ramp control signal RAMP_C. An example of the DAC circuit 2 and the switching of its reference current will be described later with reference to Fig. 5. Also, the gain of the unit attenuator AT_0 is switched to 1x or 1 / 4x by the attenuator control signal ATT_C as described above.

[0037] The voltage of the ramp signal RAMP varies between 2 (V), which corresponds to the voltage of pixel data of dim light, and 1 (V), which corresponds to the voltage of pixel data of bright light. The pixel data of dim light is converted by the ADC circuit 1 into a digital code DCD value in the range (first range) of 0 to 1023 (decimal), and the pixel data of bright light is converted into a digital code DCD value in the range (second range) of 1024 to 4095 (decimal).

[0038] 3, in the first embodiment, when the value of the digital code DCD is in the first range, the control circuit CNT sets the reference current in the ramp signal generating circuit RAMP_G to 4 times by the ramp control signal RAMP_C, and sets the gain of the unit attenuator AT_0 to 1 / 4 times by the attenuator control signal ATT_C. On the other hand, when the value of the digital code DCD is in the second range, the control circuit CNT sets the reference current in the ramp signal generating circuit RAMP_G to 1 time by the ramp control signal RAMP_C, and sets the gain of the unit attenuator AT_0 to 1 time by the attenuator control signal ATT_C.

[0039] When the value of the digital code DCD is in the first range, the reference current in the ramp signal generating circuit RAMP_G is set to four times, so that the voltage change of the ramp signal RAMP, which changes according to the least significant bit of the digital code DCD, becomes four times larger. Therefore, as shown in Fig. 3, when the value of the digital code DCD changes from 0 to 1023, the voltage of the ramp signal RAMP changes from 2 (V) to 1 (V), and the slope of the voltage change of the ramp signal RAMP with respect to the change in the value of the digital code DCD becomes four times larger than when the reference current is 1 time.

[0040] On the other hand, when the value of the digital code DCD is in the second range, that is, in the range of 1024 to 4095, the reference current is set to 1, so the slope of the voltage of the ramp signal RAMP is 1, which is smaller than when it is 4. In the first range, the voltage of the ramp signal RAMP changes from 2 (V) to 1 (V), and when it switches to the second range, the voltage of the ramp signal RAMP changes to 1.75 (V) determined by the value of the digital code DCD "1024", and then the voltage of the ramp signal RAMP changes toward 1 (V).

[0041] The gain of the unit attenuator AT_0 is 1 / 4 when the value of the digital code DCD is in the first range. Therefore, in the first range, the voltage of the ramp signal RAMP is attenuated and changes from 2 (V) to 1.75 (V) as shown by the solid line reference signal SD_0 in FIG. 3. That is, the ramp signal RAMP, which has a 4-fold slope, is reduced to 1 / 4 by the unit attenuator AT_0 and output as the reference signal SD_0. At this time, the noise NZ superimposed on the ramp signal RAMP is also attenuated to 1 / 4, so that the noise NZ can be suppressed to a small value.

[0042] On the other hand, when the digital code DCD is in the second range, the gain of the unit attenuator AT_0 is 1. Therefore, the unit attenuator AT_0 outputs the ramp signal RAMP as the reference signal SD_0.

[0043] 4A to 4C show the voltage change of the ramp signal RAMP, the gain change of the unit attenuator AT_0, and the voltage change of the reference signal SD_0 in response to the change in the digital code DCD. As shown in FIG. 4A, the slope of the voltage change of the ramp signal RAMP in response to the change in the digital code DCD is different in the first range and the second range. In contrast, as shown in FIG. 4C, the voltage change of the reference signal SD_0 in response to the change in the digital code DCD is the same in the first range and the second range.

[0044] In the first embodiment, the noise superimposed on the ramp signal RAMP corresponding to pixel data of dark light, i.e., the ramp signal RAMP when the value of the digital code DCD is in the first range, is attenuated because the gain of the unit attenuator AT_0 is 1 / 4, and it is possible to suppress the amount of noise superimposed on the reference signal SD_0 to a small amount. Since the amount of noise is suppressed, it is possible to reduce the recognition of horizontal stripes when viewed by humans, and it is possible to improve the ISO sensitivity.

[0045] Note that the 12-bit counter which is the code generating circuit DCD_G according to the first embodiment counts up the value of the digital code DCD from 0 to 4095 over time. Therefore, Fig. 4(A) to (C) can be regarded as showing the voltage change of the ramp signal RAMP, the gain change of the unit attenuator AT_0, and the voltage change of the reference signal SD_0 over time.

[0046] <Example of a ramp signal generating circuit> Fig. 5 is a diagram for explaining the ramp signal generating circuit according to the first embodiment. Here, Fig. 5(A) is a diagram showing the configuration of the ramp signal generating circuit RAMP_G, and Fig. 5(B) is a diagram showing the configuration of a current cell and a reference current circuit used in the DAC circuit 2. Fig. 5(C) is a waveform diagram showing the voltage change of the ramp signal RAMP caused by switching the reference current.

[0047] In FIG. 5(A), the DAC circuit 2 includes a current cell array having a plurality of current cells ICL arranged in a matrix, a row decoder 2_R that selects a row from the current cell array, a column decoder 2_C that selects a column from the current cell array, and a load resistance element R connected between a current cell selected from the current cell array and a voltage 2 (V).

[0048] A digital code DCD is supplied to the row decoder 2_R and the column decoder 2_C. The row decoder 2_R decodes the digital code DCD to select zero, one or more rows. Similarly, the column decoder 2_C decodes the digital code DCD to select zero, one or more columns. The current cell ICL selected by both the row decoder 2_R and the column decoder 2_C is connected to a voltage 2 (V) via a load resistor R, and a voltage of a ramp signal RAMP is output from a connection node VOUT between the load resistor R and the current cell array according to the digital code DCD.

[0049] In Fig. 5, in order to avoid complicating the drawing, a current cell array in which the current cells ICL are arranged in 8 rows x 8 columns is illustrated. When the digital code DCD varies from 0 to 4095, the current cell array has the current cells ICL arranged in 64 rows x 64 columns. According to the value of the digital code DCD, the number of current cells selected from the current cell array increases, and the voltage at the connection node VOUT decreases.

[0050] As shown in Fig. 5B, the current cell ICL is composed of an N-channel field effect transistor (transistor) NT1 and a switch SW1. When the current cell ICL is selected by both the row decoder 2_R and the column decoder 2_C, the switch SW1 of the current cell ICL is turned on.

[0051] Although not limited to this, the gate of the N-channel transistor NT1 is connected to a bias circuit common to all the current cells ICL. As shown in Fig. 5B, this bias circuit includes a diode-connected N-channel transistor NT2, switches SW2 and SW3 that are selectively turned on by a ramp control signal RAMP_C, and two constant current sources RI1 and RI2.

[0052] The constant current source RI1 outputs a reference current i that is 1x, and the constant current source RI2 outputs a reference current 4i that is 4x the reference current i. The gate of the N-channel transistor NT1 and the gate of the N-channel transistor NT2 are connected in common to form a current mirror circuit. By turning on the switch SW2 or SW3 using the ramp control signal RAMP_C, the 4x reference current 4i or the 1x reference current i is supplied to the current mirror circuit, and the 4x reference current 4i or the 1x reference current i flows through the transistor NT1 in the current cell ICL.

[0053] 5(C), in the case of a 4x reference current 4i, the voltage of the ramp signal RAMP reaches 1(V) when the digital code DCD has a value of 1023, and in the case of a 1x reference current i, the voltage of the ramp signal RAMP reaches 1(V) when the digital code DCD has a value of 4095. Note that the resistance value of the load resistance element R and the current value of the reference current i are set so that the connection node VOUT becomes 1(V) when the transistors NT1 of all current cells ICL are turned on.

[0054] Here, a current cell matrix type DAC circuit in which current cells are arranged in a matrix has been described as the DAC circuit 2 capable of switching the reference current, but the configuration of the DAC circuit is not limited to a current cell matrix type DAC circuit.

[0055] <Modification> 2 shows an example in which the gain of the attenuator ATT is switched between 1 / 4 and 1, but the present invention is not limited to this and the number of gains to be switched may be increased. As a modified example, an example in which the gain is switched between 1 / 4, 1 / 2, 3 / 4, and 1 will be described.

[0056] Fig. 6 is a block diagram showing a configuration of an image sensor according to a modification of embodiment 1. Fig. 7 is a waveform diagram for explaining the image sensor according to the modification of embodiment 1. Fig. 6 is similar to Fig. 2, and Figs. 7(A) to (C) are similar to Figs. 4(A) to (C), so the following mainly describes the differences.

[0057] FIG. 6 shows unit attenuators AT_0 to AT_m connected to a ramp signal line RAMP_L, comparators CP_0 to CP_m corresponding to the unit attenuators, and a ramp signal generating circuit RAMP_G. As shown in FIG. 6, the ramp signal generating circuit RAMP_G is common to a plurality of unit attenuators and a plurality of comparators. The difference from FIG. 2 is that the DAC circuit 2 shown in FIG. 6 can switch the reference current to 1x, 4 / 3x, 2x, or 4x by the ramp control signal RAMP_C. Also, in the unit attenuators, the attenuator control signal ATT_C can selectively turn on the switches DW1, DW3, DW4, and DW2, thereby switching the gain to 1x, 3 / 4x, 1 / 2x, or 1 / 4x.

[0058] In the modified example, the value of the digital code DCD is set as a first range of 0 to 1023, a second range of 1024 to 2047, a third range of 2048 to 3071, and a fourth range of 3072 to 4095. In the first range, the reference current of the DAC circuit 2 is set to 4 times by the ramp control signal RAMP_C, in the second range, the reference current of the DAC circuit 2 is set to 2 times, in the third range, the reference current of the DAC circuit 2 is set to 4 / 3 times, and in the fourth range, the reference current of the DAC circuit 2 is set to 1 time. In addition, in the first range, the gain of the unit attenuator is set to 1 / 4 times by the attenuator control signal AAT_C, in the second range, the gain of the unit attenuator is set to 1 / 2 times, in the third range, the gain of the unit attenuator is set to 3 / 4 times, and in the fourth range, the gain of the unit attenuator is set to 1 time.

[0059] Since the reference current of the DAC circuit 2 changes according to the range in which the digital code DCD exists, the voltage of the ramp signal RAMP changes from 2 (V) to 1 (V) in the first range, from 1.5 (V) to 1 (V) in the second range, from 1.33 (V) to 1 (V) in the third range, and from 1.25 (V) to 1 (V) in the fourth range, as shown in Fig. 7(A). That is, the slope of the voltage of the ramp signal with respect to the change in the value of the digital code DCD differs in the first to fourth ranges.

[0060] The gain of the unit attenuator also changes according to the change in the value of the digital code DCD, as shown in Fig. 7(B). That is, in the first range, the gain of the unit attenuator is set to 1 / 4, in the second range, the gain is set to 1 / 2, in the third range, the gain is set to 3 / 4, and in the fourth range, the gain is set to 1. As a result, the potential of the reference signal (e.g., SD_0) changes from 2 (V) to 1 (V) with the same gradient as the change in the digital code DCD, as shown in Fig. 7(C).

[0061] According to the modified example, as shown in FIG. 7(C), in the second range (1024 to 2047) and the third range (2048 to 3071) corresponding to intermediate luminance between bright light and dim light, the gain of the unit attenuator is 1 / 2 and 3 / 4, respectively, so that the amount of noise is reduced and the ISO sensitivity can be further improved.

[0062] In the first embodiment, the ramp signal generating circuit RAMP_G outputs a ramp signal having a different slope of voltage change with respect to the value of the digital code DCD depending on the range (first range to fourth range) in which the value of the digital code DCD exists. The attenuator to which the ramp signal is supplied is controlled so that the gain changes depending on the range (first range to fourth range) in which the value of the digital code DCD exists, and the attenuator outputs a reference signal having a constant slope of voltage change with respect to the change in the value of the digital code DCD. Considering the characteristics of the human eye, the attenuator attenuates the amount of noise in the first range to the third range corresponding to dark pixel data including intermediate luminance, thereby improving the ISO sensitivity.

[0063] (Embodiment 2) In the first embodiment, as shown in Fig. 1, unit attenuators AT_0 to AT_m corresponding to pixel data lines PID_0 to PID_m are connected to the ramp signal line RAMP_L. Since the number of pixel data lines is, for example, as many as 4000, the number of unit attenuators connected to the ramp signal line RAMP_L is also very large. As a result, the parasitic capacitance connected to the ramp signal line RAMP_L becomes very large.

[0064] In the first embodiment, when the value of the digital code DCD changes from the first range to the second range, the voltage of the ramp signal RAMP changes from 1 (V) to 1.75 (V). However, since a very large parasitic capacitance exists on the ramp signal line RAMP_L, this voltage change takes time.

[0065] FIG. 8 is a waveform diagram for explaining a problem that occurs in the image sensor according to the first embodiment. As shown in FIG. 8(A), when the digital code DCD changes from 1023 to 1024, the voltage of the ramp signal RAMP changes from 1 (V) to 1.75 (V), but this voltage change requires a period td. Due to this, as shown in FIG. 8(B), the voltage of the reference signal SD_0 also rises from 1 (V) to 1.75 (V) during the period td. As a result, it takes longer for the voltage value of the reference voltage SD_0 to reach 1 (V) from 2 (V), and it takes longer to convert one pixel row into a digital signal in the pixel array PXA (FIG. 16), resulting in a problem that the frame rate is slowed down.

[0066] Fig. 9 is a block diagram showing the configuration of an image sensor according to embodiment 2. Fig. 9 is similar to Fig. 1, so the following mainly describes the differences. The differences are that in Fig. 9, the output of a code generation circuit DCD_G is supplied to a calculator 4, and the output of the calculator 4 is a digital code DCD, and a control circuit CNT outputs a code control signal DCD_C that controls the calculator 4.

[0067] Fig. 10 is a diagram for explaining a computing unit according to the second embodiment, and Fig. 11 is a waveform diagram for explaining the operation of the image sensor according to the second embodiment.

[0068] The calculator 4 is notified by a code control signal DCD_C from the control circuit CNT whether the value of the digital code DCD is in the first range (0 to 1023) or the second range (1024 to 2095).

[0069] When notified that it is in the first range, the calculator 4 outputs the input N from the code generating circuit DCD_G as it is, as the output N. That is, in the case of the first range, the output of the code generating circuit DCD_G is used as the digital code DCD.

[0070] On the other hand, when the second range is notified, the calculator 4 subtracts the input N from the code generating circuit DCD_G from 5119 (decimal number) and outputs the subtraction result (5119-N). That is, in the second range, the subtraction result is used as the digital code DCD.

[0071] When the code generation circuit DCD_G outputs a code value of "1023" and then outputs a code value of "1024", the calculator 4 outputs the digital code DCD value of "1023" and then outputs the digital code value of "4095" obtained by subtraction. Thereafter, the 12-bit counter which is the code generation circuit DCD_G repeatedly counts up, so that the calculator 4 outputs a digital code DCD that changes from "4095" to "1024".

[0072] As a result, as shown in Fig. 11(A), the ramp signal generating circuit RAMP_G outputs a ramp signal RAMP that changes so as to rise from 1 (V) toward 1.75 (V) in the second range. That is, the ramp signal generating circuit RAMP_G outputs the ramp signal RAMP of 1 (V) in the first range, and then continuously changes the voltage of the ramp signal RAMP from 1 (V) to 1.75 (V). Since the voltage of the ramp signal RAMP changes continuously, it is not necessary to wait for the period td until the voltage of the ramp signal RAMP reaches 1 (V) from 1 (V) as described in Fig. 8.

[0073] As shown by the solid line in Figure 11 (B), the reference voltage SD_0 reaches a voltage of 1.75 (V) corresponding to the digital code DCD value of "1023", then changes to 1 (V) corresponding to the digital code DCD value of "4095", and then rises toward a voltage corresponding to the digital code DCD value of "1024".

[0074] When the value of the digital code DCD is in the second range, the comparator CP_0 compares the pixel data with the digital code DCD from the reference voltage SD_0 corresponding to the digital code DCD value of "4095" toward the reference voltage SD_0 corresponding to the digital code DCD value of "1024" over time. By sorting the digital conversion results when the digital code DCD value is in the second range in the order of the digital code DCD, it is possible to obtain continuous results of the ADC circuit 1, as shown in FIG. 11C.

[0075] Here, the explanation is based on an example in which the reference current of the DAC circuit 2 is switched between 1 and 4 times, and the gain of the attenuator ATT is switched between 1 and 1 / 4 times, but the present invention is not limited to this. For example, as explained in the modified example of the first embodiment, the number of times the reference current is switched and the number of times the gain is switched may be four or more.

[0076] Also, a decrement counter may be provided in the calculator 4, and the digital code DCD in the second range may be formed by decrementing from "4095" to "1024" by the decrement counter.

[0077] According to the second embodiment, it is possible to reduce the amount of noise and improve the ISO sensitivity, and also to solve the problem of a slow frame rate.

[0078] (Embodiment 3) Fig. 12 is a block diagram showing a configuration of an image sensor according to embodiment 3. Fig. 13 is a waveform diagram for explaining the operation of the image sensor according to embodiment 3.

[0079] Fig. 12 is similar to Fig. 6, so differences will be mainly described. In Fig. 12, the unit attenuators AT_0 to AT_m are removed, and a ramp signal line RAMP_L is connected to the comparators CP_0 to CP_m. In addition, a unit attenuator is connected as an attenuator ATT between the DAC circuit 2 and the voltage follower circuit 3. That is, the ramp signal generating circuit RAMP_G includes the DAC circuit 2, the attenuator ATT, and the voltage follower circuit 3.

[0080] The DAC circuit 2 outputs a ramp signal RAMP_CS in accordance with the digital code DCD. The ramp signal RAMP_CS is supplied as a ramp signal RAMP to a ramp signal line RAMP_L via an attenuator ATT and a voltage follower circuit 3. The comparators CP_0 to CP_m compare the ramp signal RAMP with pixel data instead of the reference signals SD_0 to SD_m, and output a trigger signal LT_C.

[0081] Since the attenuator ATT according to the third embodiment is common to the comparators CP_0 to CP_m, the attenuator ATT can be regarded as a common attenuator. As described in the first embodiment, the DAC circuit 2 and the attenuator ATT switch the reference current and the gain by the ramp control signal RAMP_C and the attenuator control signal ATT_C.

[0082] 12, Cp denotes a parasitic capacitance generated by the comparators CP_0 to CP_m connected to the ramp signal line RAMP_L. As described in the second embodiment, when there are 4000 pixel data lines, 4000 comparators are connected to the ramp signal line RAMP_L, so that the parasitic capacitance Cp becomes a very large value.

[0083] In the third embodiment, the output of the DAC circuit 2 is only connected to the attenuator ATT. Therefore, it is possible to reduce the parasitic capacitance connected to the output of the DAC circuit 2. Since the parasitic capacitance is small, as shown in FIG. 13(A), when the digital code value "1024" is supplied after the ramp signal RAMP_CS reaches 1 (V) at the digital code value "1023", the ramp signal RAMP_CS can be quickly increased to 1.75 (V). On the other hand, although the parasitic capacitance Cp is connected to the output of the voltage follower circuit 3, it is possible to smoothly change the ramp signal RAMP (corresponding to SD_0 to SD_m) by the voltage follower circuit 3, as shown in FIG. 13(B).

[0084] According to the third embodiment, it is possible to reduce the amount of noise and improve the ISO sensitivity, while preventing the frame rate from slowing down, and further, by reducing the number of components, it is possible to prevent an increase in the occupied area.

[0085] FIG. 13 shows waveforms when the reference current is switched between 1x and 4x, and the gain is switched between 1x and 1 / 4x, but the switching multiple is not limited to this.

[0086] (Embodiment 4) In the first to third embodiments, an image sensor suitable for a case where the second voltage V2 that operates the ramp signal generating circuit RAMP_G is lower than the first voltage V1 that operates the pixel array PXA has been described. In the fourth embodiment, an image sensor suitable for a case where the voltage value that operates the ramp signal generating circuit RAMP_G is higher than the second voltage V2, for example, the same voltage value as the first voltage V1, has been described.

[0087] Fig. 14 is a block diagram showing a configuration of an image sensor according to a fourth embodiment. Fig. 14 is similar to Fig. 2. That is, Fig. 14 shows only the ramp signal generating circuit RAMP_G, the unit attenuator AT_0 corresponding to the pixel data line PID_0, and the comparator CP_0 in the ADC circuit 1 shown in Fig. 1. Other unit attenuators and comparators not shown in Fig. 14 are also configured in the same manner. Note that the control circuit CNT shown in Fig. 2 is not provided in the fourth embodiment.

[0088] The ramp signal generating circuit RAMP_G according to the fourth embodiment includes a DAC circuit 2 that converts a digital code DCD into a ramp signal VOUT that is an analog signal, and a voltage regulator (voltage dividing feedback regulator) VRG.

[0089] The voltage regulator VRG is supplied with a first voltage V1, which is the same as that of the pixel array PXA, and operates on this first voltage V1. On the other hand, the DAC circuit 2 is supplied with a second voltage V2, as in the first to third embodiments, and operates on this second voltage V2. Although not particularly limited, the second voltage V2 is generated from the first voltage V1 by a step-down circuit (not shown).

[0090] The voltage regulator VRG is composed of a comparator 6, a transistor PT, and resistor elements R1 and R2. The transistor PT is composed of a P-channel transistor. The transistor PT and the resistor elements R2 and R1 are connected in series between a first voltage V1 and a voltage 1 (V). A ramp signal VOUT is supplied to one input (+) of the comparator 6, and the other input (-) is connected to a connection node between the resistor elements R2 and R1, and the output of the comparator 6 is supplied to the gate of the transistor PT. A ramp signal RAMP is output from the connection node between the transistor PT and the resistor element R2.

[0091] The DAC circuit 2 operates with the second voltage V2, and therefore outputs a ramp signal VOUT that changes between 2(V) and 1(V) according to the value of the digital code DCD, as in the first to third embodiments. On the other hand, the voltage regulator VRG operates with the first voltage V1, which is higher than the second voltage V2, and therefore converts the ramp signal VOUT that changes between 2(V) and 1(V) into a ramp signal RAMP that changes between a voltage higher than 2(V) and 1(V) and outputs the ramp signal. If the on-resistance of the transistor PT is ignored, the voltage of the ramp signal RAMP is expressed as (R1+R2) / R1×(VOUT-1(V))+1(V). In the following, an example will be described in which the first voltage V1 is 5(V) and the ratio of the resistive elements R1 and R2 is 1:2. In this case, the voltage of the ramp signal RAMP changes between 4(V) and 1(V) according to the value of the digital code DCD. That is, the voltage regulator VRG functions as a voltage dividing regulator that widens the amplitude of the ramp signal VOUT (2(V) to 1(V)) output from the DAC circuit 2 and outputs it as the ramp signal RAMP (4(V) to 1(V)).

[0092] The unit attenuator AT_0 is composed of three capacitance elements CD1 to CD3 connected in series between the ramp signal RAMP and the voltage 1 (V), and a reference signal SD_0 is output from the connection node between the capacitance elements CD2 and CD3. That is, the gain of the unit attenuator AT_0 shown in FIG. 14 is fixed to 1 / 3.

[0093] The comparator CP_0 operates, for example, with the second voltage V2. As already described, the comparator CP_0 operates to compare the pixel data with the reference signal SD_0.

[0094] Fig. 15 is a waveform diagram for explaining the operation of the image sensor according to the fourth embodiment. Here, Fig. 15(A) shows the voltage waveform of the ramp signal RAMP, and Fig. 15(B) shows the voltage waveform of the reference signal SD_0. In both Figs. 15(A) and 15(B), the horizontal axis indicates the digital code DCD.

[0095] As shown in FIG. 15(A), the voltage of the ramp signal RAMP output from the voltage regulator VRG drops from 4 (V) to 1 (V) at a first slope. The unit attenuator AT_0 attenuates the ramp signal RAMP by 1 / 3 and outputs the reference signal SD_0. Therefore, as shown in FIG. 15(B), the reference signal SD_0 drops, for example, from 2 (V) to 1 (V) at a second slope smaller than the first slope. As shown in FIG. 15(A), if noise NZ is superimposed on the ramp signal RAMP, the noise NZ is attenuated by 1 / 3 by the unit attenuator AT_0.

[0096] According to the fourth embodiment, it is possible to attenuate noise at all luminance levels and improve the ISO sensitivity. In addition, since the control circuit CNT shown in the first to third embodiments is not required, it is possible to suppress an increase in the area required for the control circuit CNT.

[0097] 14 shows a case where the gain of the unit attenuator AT_0 is 1 / 3, but the present invention is not limited to this. Also, as shown in FIG. 12, a common attenuator may be used instead of the unit attenuators.

[0098] In the first to third embodiments, the control circuit CNT outputs the lamp control signal RAMP_C, the attenuator control signal ATT_C, and the code control signal DCD_C based on the digital code DCD, although this is not particularly limited. For example, the control circuit CNT compares the value of the digital code that defines the first range to the fourth range, etc., with the value of the digital code DCD output from the code generating circuit DCD_G, and outputs the lamp control signal RAMP_C, the attenuator control signal ATT_C, and the code control signal DCD_C that correspond to the period in which they match.

[0099] The present invention has been specifically described above based on the embodiments, but the present invention is not limited to the above embodiments and can be modified in various ways without departing from the spirit of the present invention. For example, in the first to third embodiments, the common control circuit CNT outputs the ramp control signal RAMP_C, the attenuator control signal ATT_C, and the code control signal DCD_C, but the present invention is not limited to this. For example, the ramp signal generating circuit RAMP_G, the attenuator ATT, and the calculator 4 may generate their respective control signals based on the output of the code generating circuit DCD_G. [Explanation of symbols]

[0100] 1 ADC circuit 2 DAC circuit 3. Voltage follower circuit 4 Arithmetic unit 100 Image Sensor ATT Attenuator BSC Bias circuit CNT control circuit CP comparison circuit DCD_G code generator PID_0~PID_m pixel data lines RAMP Ramp signal RAMP_G Ramp signal generation circuit SD_0~SD_m Reference signal

Claims

1. A pixel array including a plurality of imaging elements; an ADC circuit to which pixel data is supplied in parallel from the pixel array, the ADC circuit including a reference signal generating circuit that outputs a reference signal according to a digital code, and a comparison circuit that compares the reference signal with the pixel data, the ADC circuit outputting the digital code when the reference signal and the pixel data have a predetermined relationship; Equipped with The reference signal generating circuit includes: a ramp signal generating circuit that outputs ramp signals with different slopes in response to a change in the digital code when the digital code is in a first range and when the digital code is in a second range different from the first range; an attenuator that receives the ramp signal and outputs the reference signal having the same slope in the first range and the second range; An image sensor comprising:

2. 2. The image sensor according to claim 1, the comparison circuit includes a plurality of comparators corresponding to the pixel data supplied in parallel, the ADC circuit includes a plurality of latch circuits corresponding to the plurality of comparators; The digital code is supplied to the plurality of latch circuits, and when a corresponding comparator indicates that the pixel data and a reference signal match, the latch circuit holds the digital code as a digital code corresponding to the pixel data.

3. 3. The image sensor according to claim 2, The attenuator includes a voltage divider circuit that divides a voltage of the ramp signal.

4. 3. The image sensor according to claim 2, an image sensor, wherein the reference signal generating circuit includes a code generating circuit that generates a code, and a calculator that receives the code generated by the code generating circuit and outputs the digital code, and the calculator outputs a digital code value that continuously changes the ramp signal when changing from the first range to the second range.

5. 3. The image sensor according to claim 2, The attenuator includes a plurality of unit attenuators corresponding to the plurality of comparators.

6. 3. The image sensor according to claim 2, The reference signal output from the attenuator is commonly supplied to the plurality of comparators.

7. 2. The image sensor according to claim 1, The image sensor, wherein the pixel array operates on a first voltage, and the ADC circuit operates on a second voltage having a voltage value lower than the first voltage.

8. a pixel array including a plurality of image sensors and operated by a first voltage; an ADC circuit to which pixel data is supplied in parallel from the pixel array, the ADC circuit including a reference signal generating circuit that outputs a reference signal according to a digital code value, and a comparison circuit that compares the reference signal with the pixel data, the ADC circuit outputting the digital code value when the reference signal and the pixel data have a predetermined relationship; Equipped with The reference signal generating circuit includes: a generating circuit that operates by a second voltage lower than the first voltage and outputs a ramp signal having a first slope in accordance with a change in the digital code value; a voltage conversion circuit that is operated by the first voltage and converts a voltage of the ramp signal; an attenuator that receives the ramp signal converted by the voltage conversion circuit and outputs a ramp signal having a second slope different from the first slope as the reference signal; An image sensor comprising:

9. 9. The image sensor according to claim 8, The image sensor, wherein the generating circuit includes a DAC circuit that converts the digital code value into an analog signal, the voltage conversion circuit includes a voltage divider feedback regulator that operates by the first voltage, and the attenuator includes a voltage divider circuit configured with a plurality of capacitive elements.

Citation Information

Patent Citations

  • Voltage generation circuit, digital / analog converter, lamp wave generation circuit, analog / digital converter, image sensor system and voltage generation method

    JP2011041205A

  • Analog digital conversion ramp wave generator unit element

    JP2012005126A

  • Solid-state imaging element

    JP2014090325A

  • Photoelectric conversion device and imaging system

    JP2015162751A

  • Photoelectric conversion apparatus and image capturing system

    US20150244388A1