Image sensor

Abstract

The utility model provides an image sensor which comprises a pixel circuit and a quantization circuit. The pixel circuit comprises a first pixel unit and a second pixel unit, the first pixel unit generates a first signal voltage at least based on photoelectric conversion, and the second pixel unit generates a second signal voltage at least based on photoelectric conversion; the quantization circuit comprises a first quantization unit and a second quantization unit which correspond to the first pixel unit and the second pixel unit respectively and are used for quantizing the first signal voltage and the second signal voltage through the ramp voltage and correcting the first signal voltage and the second signal voltage respectively based on the difference value of the first signal voltage and the second signal voltage so as to obtain a corrected signal. And the deviation condition of an actual value and an ideal value is improved. According to the CMOS image sensor, the problem of color ratio deviation caused by mutual interference between adjacent pixels of an existing CMOS image sensor is solved.

Description

Technical Field

[0001] This utility model relates to the field of image sensor technology, and in particular to an image sensor. Background Technology

[0002] In CMOS image sensors, the color ratio is a very important performance indicator, which reflects whether the quantization ratio of different colors in the CMOS image sensor remains consistent. For example, after blue light (B) and green light (G) pass through the light-transmitting plate, the ratio of the charges stored in the two illumination areas is a fixed value, taking G / B = 5 / 1 as an example.

[0003] In practical applications, due to mutual interference among pixels, there is a parasitic capacitance between the pixels that quantize blue light and the pixels that quantize green light, causing the blue light voltage value and green light voltage value output by the pixels to deviate. This results in the ratio of quantized green light to blue light deviating from the ideal value, causing a color ratio deviation.

[0004] It should be noted that the above introduction to the technical background is only for the purpose of providing a clear and complete explanation of the technical solutions of this utility model and facilitating understanding by those skilled in the art. It should not be assumed that these technical solutions are known to those skilled in the art simply because they have been described in the background section of this utility model. Utility Model Content

[0005] In view of the shortcomings of the prior art described above, the purpose of this utility model is to provide an image sensor to solve the problem of color ratio deviation caused by mutual interference between adjacent pixels in existing CMOS image sensors.

[0006] To achieve the above and other related objectives, this utility model provides an image sensor, the image sensor comprising:

[0007] A pixel circuit includes a first pixel unit and a second pixel unit, wherein the first pixel unit generates a first signal voltage based on photoelectric conversion, and the second pixel unit generates a second signal voltage based on photoelectric conversion.

[0008] The quantization circuit includes a first quantization unit and a second quantization unit, which correspond to the first pixel unit and the second pixel unit, respectively. It is used to quantize the first signal voltage and the second signal voltage through a ramp voltage, and to correct the first signal voltage and the second signal voltage based on the difference between the first signal voltage and the second signal voltage, thereby improving the deviation between the actual value and the ideal value.

[0009] Optionally, the first pixel unit performs photoelectric conversion on the first color light to obtain a first charge and generates the first signal voltage, and the second pixel unit performs photoelectric conversion on the second color light to obtain a second charge and generates the second signal voltage, wherein the first color light and the second color light are different colors.

[0010] Optionally, the transmittance of the first color light is greater than that of the second color light, and the first charge is greater than the second charge; due to the influence of the parasitic capacitance between the first pixel unit and the second pixel unit, the actual value of the first signal voltage is greater than the ideal value, and the actual value of the second signal voltage is smaller than the ideal value.

[0011] Optionally, the first pixel unit and the second pixel unit are two adjacent pixel units in the same pixel row, wherein the first color light includes green light and the second color light includes blue light or red light.

[0012] Optionally, the first quantization unit corrects the first signal voltage based on the product of the difference between the first signal voltage and the second signal voltage and a first correction coefficient; the second quantization unit corrects the second signal voltage based on the product of the difference between the first signal voltage and the second signal voltage and a second correction coefficient; wherein the first correction coefficient and the second correction coefficient are equal or unequal.

[0013] Optionally, if the actual value of the first signal voltage is larger than the ideal value, the first signal voltage is corrected by addition. The first quantization unit includes a first comparator, a first decoupling capacitor, a second decoupling capacitor, a first correction capacitor, and a second correction capacitor. The non-inverting input of the first comparator receives the ramp voltage through the first decoupling capacitor, the inverting input of the first comparator receives the first signal voltage through the second decoupling capacitor, the non-inverting input of the first comparator also receives the second signal voltage through the first correction capacitor, the inverting input of the first comparator also receives the first signal voltage through the second correction capacitor, and the output of the first comparator outputs the first quantization result.

[0014] Optionally, the capacitance values ​​of the first decoupling capacitor and the second decoupling capacitor are equal, and the capacitance values ​​of the first correction capacitor and the second correction capacitor are equal.

[0015] Optionally, if the actual value of the second signal voltage is smaller than the ideal value, the second signal voltage is corrected by subtraction. The second quantization unit includes a second comparator, a third decoupling capacitor, a fourth decoupling capacitor, a third correction capacitor, and a fourth correction capacitor. The non-inverting input of the second comparator receives the ramp voltage through the third decoupling capacitor, the inverting input of the second comparator receives the second signal voltage through the fourth decoupling capacitor, the non-inverting input of the second comparator also receives the first signal voltage through the third correction capacitor, the inverting input of the second comparator also receives the second signal voltage through the fourth correction capacitor, and the output of the second comparator outputs the second quantization result.

[0016] Optionally, the capacitance values ​​of the third decoupling capacitor and the fourth decoupling capacitor are equal, and the capacitance values ​​of the third correction capacitor and the fourth correction capacitor are equal.

[0017] Optionally, the image sensor further includes a ramp circuit for providing the ramp voltage.

[0018] As described above, the image sensor of this invention solves the technical problem of color ratio deviation by compensating for the voltage error caused by mutual interference between two pixel units in the quantization section of the analog circuit. Moreover, the compensation range of this invention is large and not limited to a certain range. Attached Figure Description

[0019] Figure 1 The diagram shows a pixel unit quantized by a comparator.

[0020] Figure 2 Displayed as Figure 1 The diagram shows the quantization result output by the comparator.

[0021] Figure 3 This is a diagram showing the pixel arrangement in the same pixel row.

[0022] Figure 4 The diagram shows the parasitic capacitance between two pixel units due to mutual interference.

[0023] Figure 5 This diagram illustrates the color deviation caused by mutual interference.

[0024] Figure 6 The diagram shown is a structural schematic of the image sensor in an embodiment of this utility model.

[0025] Figure 7 The diagram shown illustrates how a quantization unit corrects color deviation in an embodiment of this invention.

[0026] Component designation explanation

[0027] 10 Image Sensors

[0028] 100-pixel circuit

[0029] 110 First pixel unit

[0030] 120 Second pixel unit

[0031] 200 Quantization Circuit

[0032] 210 First Quantization Unit

[0033] 220 Second Quantization Unit

[0034] 300 ramp circuit Detailed Implementation

[0035] The following specific examples illustrate the implementation of this utility model. Those skilled in the art can easily understand other advantages and effects of this utility model from the content disclosed in this specification. This utility model can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this utility model.

[0036] Please see Figures 1 to 7 It should be noted that the illustrations provided in this embodiment are only schematic representations of the basic concept of this utility model. Therefore, the illustrations only show the components related to this utility model and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the shape, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0037] Figure 1 A schematic diagram of a pixel unit quantized by a comparator is shown. The pixel unit includes a reset transistor M1, a transfer transistor M2, a photosensitive element PD, a source follower transistor M3, a select transistor M4, and a floating diffusion node FD. The pixel unit also has a tail current source vln, and the specific connection is shown in the figure.

[0038] When the pixel unit is working:

[0039] First, a reset operation is performed. Reset transistor M1 is turned on, transfer transistor M2 is turned off, and selection transistor M4 is turned on. The floating diffusion node FD is pulled high to the pixel voltage PIXVDD, and the output node BL is also pulled high. At this time, the voltage of the output node BL is the reset voltage, denoted as Vrst. When light shines on the photosensitive element PD, the photosensitive element PD generates charge, which is transferred to the floating diffusion node FD through the turned-on transfer transistor M2. This causes the voltage of the floating diffusion node FD to drop from the pixel voltage PIXVDD to a certain voltage. The voltage of the output node BL also drops along with the voltage of the floating diffusion node FD. At this time, the voltage on the output node BL is the signal voltage, denoted as Vsig. The difference between the reset voltage Vrst and the signal voltage Vsig represents the light intensity. A larger difference between the reset voltage Vrst and the signal voltage Vsig indicates stronger light.

[0040] When the comparator is quantized:

[0041] When the ramp voltage Vramp begins to decrease, the comparator output is pulled from low to high until Vramp drops to equal the voltage at the output node BL. At this point, the comparator output flips from high to low. The time the comparator holds the high level is the quantization time of the corresponding voltage. For the reset voltage Vrst and the signal voltage Vsig, each has a quantization time during quantization by the comparator, namely, the reset quantization time trst and the signal quantization time tsig, as shown below. Figure 2 As shown, the quantization result is obtained by subtracting the reset quantization time trst from the signal quantization time tsig. It can be seen that the stronger the illumination, the smaller the signal voltage Vsig, and the longer the quantization time of the signal voltage Vsig, that is, the larger the signal quantization time tsig. Therefore, the larger the value of tsig-trst, the larger the quantization result.

[0042] Figure 3 This illustrates a pixel arrangement where each pixel unit in the current pixel row corresponds to two colors arranged alternately, such as green (G) and blue (B). When quantizing each pixel unit in the same pixel row, there is crosstalk between adjacent pixel units. This can be considered as parasitic capacitance between adjacent pixel units.

[0043] Taking any two adjacent pixel units as an example, such as Figure 4As shown, a parasitic capacitance Cpar exists between the floating diffusion nodes of the green pixel unit (PIXEL_G) and the blue pixel unit (PIXEL_B). Assuming the reset voltages of the two floating diffusion nodes are equal, when light shines on the two pixel units, due to the different transmittances of green and blue light, the amount of charge received by the two floating diffusion nodes is also different. The charge of floating diffusion node FD_G is greater than that of floating diffusion node FD_B. At this time, the voltage level of floating diffusion node FD_G is lower than that of floating diffusion node FD_B. However, due to the presence of the parasitic capacitance Cpar, the floating diffusion nodes of the two pixel units interfere with each other, causing the voltage level of floating diffusion node FD_G to increase and the voltage level of floating diffusion node FD_B to decrease. This results in the actual value of the signal voltage of floating diffusion node FD_G being higher than the ideal value and the actual value of the signal voltage of floating diffusion node FD_B being lower than the ideal value.

[0044] When quantizing the above signal voltage, the green pixel unit (PIXEL_G) corresponds to comparator CMPG, and the blue pixel unit (PIXEL_B) corresponds to comparator CMPB. Capacitors C0 and C1 are decoupling capacitors with equal capacitance values. Figure 5 As shown in the figure, due to the interference between adjacent pixel units, the actual value of the signal voltage (denoted as Vsigg) sent to comparator CMPG is higher than the ideal value, and the actual value of the signal voltage (denoted as Vsigb) sent to comparator CMPB is lower than the ideal value. This results in the actual value of the signal quantization time (denoted as tsigg) output by comparator CMPG being smaller than the ideal value, and the actual value of the signal quantization time (denoted as tsigb) output by comparator CMPB being larger than the ideal value. Ultimately, this is reflected in the image as the color ratio deviating from the original predetermined value, causing a color ratio deviation.

[0045] To address the aforementioned technical problems, this embodiment proposes an image sensor that compensates for the voltage error caused by mutual interference between two pixel units in the quantization section of the analog circuit, thereby solving the technical problem of color ratio deviation. Moreover, the compensation range of this embodiment is large and not limited to a certain range.

[0046] like Figure 6 and Figure 7 As shown, this embodiment provides an image sensor 10, including a pixel circuit 100 and a quantization circuit 200, and further includes a ramp circuit 300.

[0047] The pixel circuit 100 includes a first pixel unit 110 and a second pixel unit 120. In practical applications, the pixel circuit 100 may also include other pixel units, such as a third pixel unit, a fourth pixel unit, etc. Of course, the number of each pixel unit is not limited to one, but may be more than one, thus forming a pixel array. In addition, the pixel circuit 100 also includes several tail current sources vln. For example, each tail current source vln corresponds one-to-one with each pixel column in the pixel array, and is used to provide output drive for the column line of each pixel column, so that the corresponding voltage is output through the corresponding column line.

[0048] In this design, the first pixel unit 110 generates a first signal voltage, denoted as Vsig1, based at least on photoelectric conversion, and the second pixel unit 120 generates a second signal voltage, denoted as Vsig2, based at least on photoelectric conversion. During the quantization stage, parasitic capacitance Cpar exists between the first pixel unit 110 and the second pixel unit 120 due to crosstalk, causing the actual values ​​of the first signal voltage Vsig1 and the second signal voltage Vsig2 to deviate from their ideal values. Of course, before generating the first signal voltage Vsig1 and the second signal voltage Vsig2, the first pixel unit 110 can also generate a first reset voltage, denoted as Vrst1, based on a reset operation, and the second pixel unit 120 can also generate a second reset voltage, denoted as Vrst2, based on a reset operation. In practical applications, the first reset voltage Vrst1 and the second reset voltage Vrst2 are usually designed to be equal, thus avoiding the influence of parasitic capacitance Cpar on the first reset voltage Vrst1 and the second reset voltage Vrst2.

[0049] Specifically, the first pixel unit 110 performs photoelectric conversion on the first color light to obtain a first charge Q1 and generates a first signal voltage Vsig1, and the second pixel unit 120 performs photoelectric conversion on the second color light to obtain a second charge Q2 and generates a second signal voltage Vsig2, wherein the first color light and the second color light are different colors.

[0050] In one example, the transmittance of the first color light is greater than that of the second color light. Therefore, the first charge Q1 obtained by the first pixel unit 110 after photoelectric conversion of the first color light is greater than the second charge Q2 obtained by the second pixel unit 120 after photoelectric conversion of the second color light. The first charge Q1 is transferred and stored in the floating diffusion node of the first pixel unit 110, and the second charge Q2 is transferred and stored in the floating diffusion node of the second pixel unit 120. At this time, a first signal voltage Vsig1 is generated at the floating diffusion node of the first pixel unit 110. A second signal voltage Vsig2 is generated at the floating diffusion node of the second pixel unit 120. However, due to the influence of parasitic capacitance Cpar, the level of the floating diffusion node of the first pixel unit 110 and the level of the floating diffusion node of the second pixel unit 120 affect each other. This causes the first charge Q1 at the floating diffusion node of the first pixel unit 110 to decrease and the second charge Q2 at the floating diffusion node of the second pixel unit 120 to increase. As a result, the actual value of the first signal voltage Vsig1 is larger than the ideal value and the actual value of the second signal voltage Vsig2 is smaller than the ideal value.

[0051] In one implementation, the first pixel unit 110 and the second pixel unit 120 are two adjacent pixel units in the same pixel row. Since the first pixel unit 110 and the second pixel unit 120 are in the same pixel row in the layout, there is crosstalk between them, resulting in a parasitic capacitance Cpar. Furthermore, each pixel unit in the pixel circuit 100 typically uses a Bayer color array, i.e., an RGGB color array. In this case, the first color light corresponding to the first pixel unit 110 includes green light (G), and the second color light corresponding to the second pixel unit 120 includes blue light (B), or the first color light corresponding to the first pixel unit 110 includes green light (G), and the second color light corresponding to the second pixel unit 120 includes red light (R). In practical applications, the circuit structure of each pixel unit is the same, for example, including a reset transistor, a transfer transistor, a photosensitive element, a source follower transistor, and a selection transistor. For specific connections, please refer to [reference needed]. Figure 1 Of course, it can also include gain transistors, overflow transistors, overflow capacitors, etc. In fact, this embodiment does not limit the circuit structure of each pixel unit, and any circuit structure that interferes with each other is applicable to this embodiment.

[0052] The quantization circuit 200 includes a first quantization unit 210 and a second quantization unit 220. Of course, the quantization circuit 200 may also include other quantization units. In practical applications, for example, each quantization unit corresponds one-to-one with each pixel column in the pixel array, so that when quantizing row by row, the voltage of each pixel unit in the corresponding pixel row can be quantized and read out through each quantization unit.

[0053] The first quantization unit 210 and the second quantization unit 220 correspond to the first pixel unit 110 and the second pixel unit 120, respectively. When quantizing the first signal voltage Vsig1 and the second signal voltage Vsig2 using the ramp voltage Vramp, the first signal voltage Vsig1 and the second signal voltage Vsig2 are corrected based on the difference between them to improve the deviation between the actual value and the ideal value. Of course, when the first pixel unit 110 and the second pixel unit 120 also generate the first reset voltage Vrst1 and the second reset voltage Vrst2, the first quantization unit 210 and the second quantization unit 220 also quantize the first reset voltage Vrst1 and the second reset voltage Vrst2 using the ramp voltage Vramp. In practical applications, the reset voltage is usually quantized first, followed by the signal voltage.

[0054] Specifically, the first quantization unit 210 corrects the first signal voltage Vsig1 based on the product of the difference between the first signal voltage Vsig1 and the second signal voltage Vsig2 and the first correction coefficient K1; the second quantization unit 220 corrects the second signal voltage Vsig2 based on the product of the difference between the first signal voltage Vsig1 and the second signal voltage Vsig2 and the second correction coefficient K2. The first correction coefficient K1 and the second correction coefficient K2 can be equal or unequal, and should be designed according to the actual situation. In one example, if the actual value of the first signal voltage is larger than the ideal value, then the first signal voltage Vsig1 is corrected by addition; if the actual value of the second signal voltage is smaller than the ideal value, then the second signal voltage Vsig2 is corrected by subtraction.

[0055] In one implementation scheme, such as Figure 7 As shown, for the case where the actual value of the first signal voltage is larger than the ideal value, the first quantization unit 210 includes a first comparator CMP1, a first decoupling capacitor C01, a second decoupling capacitor C02, a first correction capacitor C11, and a second correction capacitor C12. The non-inverting input terminal (i.e., the vin terminal) of the first comparator CMP1 receives the ramp voltage Vramp through the first decoupling capacitor C01, the inverting input terminal (i.e., the vinn terminal) of the first comparator CMP1 receives the first signal voltage Vsig1 through the second decoupling capacitor C02, the non-inverting input terminal (i.e., the vin terminal) of the first comparator CMP1 also receives the second signal voltage Vsig2 through the first correction capacitor C11, the inverting input terminal (i.e., the vinn terminal) of the first comparator CMP1 also receives the first signal voltage Vsig1 through the second correction capacitor C12, and the output terminal of the first comparator CMP1 outputs the first quantization result cmpout1.

[0056] For cases where the actual value of the second signal voltage is smaller than the ideal value, the second quantization unit 220 includes a second comparator CMP2, a third decoupling capacitor C03, a fourth decoupling capacitor C04, a third correction capacitor C13, and a fourth correction capacitor C14. The non-inverting input terminal (i.e., the vin terminal) of the second comparator CMP2 receives the ramp voltage Vramp through the third decoupling capacitor C03, the inverting input terminal (i.e., the vinn terminal) of the second comparator CMP2 receives the second signal voltage Vsig2 through the fourth decoupling capacitor C04, the non-inverting input terminal (i.e., the vin terminal) of the second comparator CMP2 also receives the first signal voltage Vsig1 through the third correction capacitor C13, the inverting input terminal (i.e., the vinn terminal) of the second comparator CMP2 also receives the second signal voltage Vsig2 through the fourth correction capacitor C14, and the output terminal of the second comparator CMP2 outputs the second quantization result cmpout2.

[0057] Specifically, the capacitance values ​​of the first decoupling capacitor C01 and the second decoupling capacitor C02 are equal, i.e., C01 = C02 = Cinput1; the capacitance values ​​of the third decoupling capacitor C03 and the fourth decoupling capacitor C04 are equal, i.e., C03 = C04 = Cinput2; furthermore, the capacitance values ​​of all four decoupling capacitors are equal, i.e., C01 = C02 = C03 = C04 = Cinput. The capacitance values ​​of the first correction capacitor C11 and the second correction capacitor C12 are equal, i.e., C11 = C12 = Cfix1, and this capacitance value determines the value of the first correction coefficient K1; the capacitance values ​​of the third correction capacitor C13 and the fourth correction capacitor C14 are equal, i.e., C13 = C14 = Cfix2, and this capacitance value determines the value of the second correction coefficient K2. It should be noted that the first signal voltage Vsig1 mentioned in the quantization section refers to the actual value of the first signal voltage, and the second signal voltage Vsig2 refers to the actual value of the second signal voltage.

[0058] The ramp circuit 300 is used to provide the ramp voltage Vramp. In practical applications, the ramp circuit 300 is implemented using a ramp generator. Of course, other circuit structures that can generate the ramp voltage Vramp are also applicable to this embodiment, and there is no limitation on them.

[0059] Accordingly, this embodiment also provides a voltage correction method for an image sensor 10, including an exposure stage and a quantization stage; wherein the image sensor 10 is implemented using the structure described above.

[0060] During the exposure readout stage, the first pixel unit 110 generates a first signal voltage Vsig1 based on photoelectric conversion, and the second pixel unit 120 generates a second signal voltage Vsig2 based on photoelectric conversion. Of course, before generating the first signal voltage Vsig1 and the second signal voltage Vsig2, the first pixel unit 110 can also generate a first reset voltage Vrst1 based on a reset operation, and the second pixel unit 120 can also generate a second reset voltage Vrst2 based on a reset operation. In order to avoid the influence of parasitic capacitance on the first reset voltage Vrst1 and the second reset voltage Vrst2, the first reset voltage Vrst1 and the second reset voltage Vrst2 are usually designed to be equal, for example, both being pixel voltage PIXVDD.

[0061] Specifically, a reset operation is first performed on the first pixel unit 110 and the second pixel unit 120. At this time, a first reset voltage Vrst1 is generated at the floating diffusion node of the first pixel unit 110, and a second reset voltage Vrst2 is generated at the floating diffusion node of the second pixel unit 120. Then, the first charge obtained by the first pixel unit 110 through photoelectric conversion of the first color light is transferred to the corresponding floating diffusion node to generate a first signal voltage Vsig1, and the second charge obtained by the second pixel unit 120 through photoelectric conversion of the second color light is transferred to the corresponding floating diffusion node to generate a second signal voltage Vsig2.

[0062] During the quantization phase, the first quantization unit 210 and the second quantization unit 220 quantize the first signal voltage Vsig1 and the second signal voltage Vsig2 respectively using the ramp voltage Vramp, and correct the first signal voltage Vsig1 and the second signal voltage Vsig2 based on the difference between the first signal voltage Vsig1 and the second signal voltage Vsig2 to improve the deviation between the actual value and the ideal value. Of course, before quantizing the first signal voltage Vsig1 and the second signal voltage Vsig2, the first quantization unit 210 and the second quantization unit 220 also quantize the first reset voltage Vrst1 and the second reset voltage Vrst2 respectively using the ramp voltage Vramp.

[0063] Specifically, when the first quantization unit 210 quantizes the first signal voltage Vsig1 using the ramp voltage Vramp, the actual value of the first signal voltage is larger than the ideal value. The first signal voltage is corrected using the formula Vsig1_fixed = Vsig1_real + K1·(Vsig1_real - Vsig2_real); where Vsig1_fixed is the corrected value of the first signal voltage, Vsig1_real is the actual value of the first signal voltage, Vsig2_real is the actual value of the second signal voltage, and K1 is the first correction coefficient. More specifically, the first correction coefficient K1 satisfies the formula... Wherein, Cfix1 is the capacitance value of the first correction capacitor C11 or the second correction capacitor C12, and Cinput1 is the capacitance value of the first decoupling capacitor C01 or the second decoupling capacitor C02.

[0064] When the second quantization unit 220 quantizes the second signal voltage Vsig2 using the ramp voltage Vramp, the actual value of the second signal voltage is smaller than the ideal value. Therefore, the second signal voltage is corrected using the formula Vsig2_fixed = Vsig2_real - K2·(Vsig1_real - Vsig2_real); where Vsig2_fixed is the corrected value of the second signal voltage, Vsig2_real is the actual value of the second signal voltage, Vsig1_real is the actual value of the first signal voltage, and K2 is the second correction coefficient. More specifically, the second correction coefficient K2 satisfies the formula... Wherein, Cfix2 is the capacitance value of the third correction capacitor C13 or the fourth correction capacitor C14, and Cinput2 is the capacitance value of the third decoupling capacitor C03 or the fourth decoupling capacitor C04.

[0065] In practical applications, the values ​​of Cfix1 and Cfix2 are designed to determine the values ​​of K1 and K2, respectively, thereby correcting the first signal voltage Vsig1 and the second signal voltage Vsig2 to varying degrees, making the corrected values ​​close to or even restored to the ideal values. When designing the values ​​of Cfix1 and Cfix2, the values ​​of Cpar / Ccg can be referenced. Considering actual non-ideal factors, the values ​​of Cfix1 and Cfix2 can fluctuate based on the value of Cpar / Ccg. Here, Cpar is the capacitance value of the parasitic capacitance between the first pixel unit 110 and the second pixel unit 120, and Ccg is the capacitance value corresponding to the floating diffusion nodes of the first pixel unit 110 and the second pixel unit 120. Both are determined by the layout, and the values ​​of Cpar and Ccg can be obtained after the local layout is determined.

[0066] Below, please combine Figure 6 and Figure 7 The theoretical basis of the image sensor and its voltage correction method in this embodiment is explained in detail.

[0067] Assuming the first color light corresponding to the first pixel unit 110 is green light, the second color light corresponding to the second pixel unit 120 is blue light, and the first charge is 5Q and the second charge is Q, then:

[0068] Ideally, for the first pixel unit 110, there is the formula 5Q = Ccg * U1, from which U1 = 5Q / Ccg is derived; for the second pixel unit 120, there is the formula Q = Ccg * U2, from which U2 = Q / Ccg is derived; where Ccg is the capacitance value of the node capacitor corresponding to the floating diffusion node of the first pixel unit 110 and the second pixel unit 120, U1 is the ideal voltage drop generated across the node capacitor of the first pixel unit 110 due to photoelectric conversion, and U2 is the ideal voltage drop generated across the node capacitor of the second pixel unit 120 due to photoelectric conversion.

[0069] Due to mutual interference, for the first pixel unit 110, there is the formula 5Q=Ccg*U1'+Cpar*(U1'-U2'), from which U1=5Q / Ccg=U1'+Cpar / Ccg*(U1'-U2'); for the second pixel unit 120, there is the formula Q=Ccg*U2'-Cpar*(U1'-U2'), from which U2=Q / Ccg=U2'-Cpar / Ccg*(U1'-U2'); where U1' is the actual voltage drop across the node capacitor caused by photoelectric conversion in the first pixel unit 110, U2' is the actual voltage drop across the node capacitor caused by photoelectric conversion in the second pixel unit 120, and Cpar is the capacitance value of the parasitic capacitance between the first pixel unit 110 and the second pixel unit 120. It can be seen that the errors generated by photoelectric conversion at both ends of the node capacitor in the first pixel unit 110 and the second pixel unit 120 are +Cpar / Ccg*(U1'-U2') and -Cpar / Ccg*(U1'-U2'), respectively; as long as the above errors are compensated, the influence caused by mutual interference can be resolved.

[0070] Let Cpar / Ccg = K, then U1 = U1' + K*(U1' - U2'), U2 = U2' - K*(U1' - U2'); taking into account the actual non-ideal factors, we make appropriate adjustments to K to obtain K1 and K2, then U1 = U1' + K1*(U1' - U2'), U2 = U2' - K2*(U1' - U2'). Regarding U1 and U2, there are the formulas: Vsig1_ideal = Vrst1 - U1, Vsig2_ideal = Vrst2 - U1; regarding U1' and U2', there are the formulas: Vsig1_real = Vrst1 - U1', Vsig2_real = Vrst2 - U2'; where Vsig1_ideal is the ideal value of the first signal voltage, Vsig1_real is the ideal value of the first signal voltage, Vrst1 is the value of the first reset voltage, Vsig2_ideal is the ideal value of the second signal voltage, Vsig1_real is the actual value of the second signal voltage, and Vrst2 is the value of the second reset voltage.

[0071] Substituting the above formulas into U1=U1'+K1*(U1'-U2') and U2=U2'-K2*(U1'-U2'), and letting Vrst1=Vrst2, we have the formulas: Vsig1_ideal=Vsig1_real+K1*(Vsig1_real-Vsig2_real), Vsig2_ideal=Vsig2_real-K2*(Vsig1_real-Vsig2_real). Based on this, in the circuit implementation of this embodiment, by utilizing the differential input characteristics of the comparator and connecting a correction capacitor in parallel with the decoupling capacitor, voltage correction can be achieved during the quantization stage.

[0072] Since the actual value of the first signal voltage is larger than the ideal value, and the actual value of the second signal voltage is smaller than the ideal value, the first signal voltage Vsig1 is reduced and the second signal voltage Vsig2 is increased through correction. Therefore, when correcting the first signal voltage Vsig1, the first signal voltage Vsig1 and the second signal voltage Vsig2 are connected to the inverting input and non-inverting input of the corresponding comparator, respectively, to achieve addition. When correcting the second signal voltage Vsig2, the first signal voltage Vsig1 and the second signal voltage Vsig2 are connected to the non-inverting input and inverting input of the corresponding comparator, respectively, to achieve subtraction. In addition, the corresponding correction coefficients are designed by the size of the corresponding correction capacitors. The first correction coefficient is denoted as K1, and the second correction coefficient is denoted as K2. Both K1 and K2 are positive numbers greater than zero.

[0073] In summary, the image sensor of this invention solves the technical problem of color ratio deviation by compensating for voltage errors caused by mutual interference between two pixel units in the quantization section of the analog circuit. Furthermore, the compensation range of this invention is wide and not limited to a specific range. Therefore, this invention effectively overcomes the various shortcomings of the prior art and has high industrial application value.

[0074] The above embodiments are merely illustrative of the principles and effects of this utility model and are not intended to limit the scope of this utility model. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of this utility model. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in this utility model should still be covered by the claims of this utility model.

Claims

1. An image sensor, characterized in that, The image sensor includes: A pixel circuit includes a first pixel unit and a second pixel unit, wherein the first pixel unit generates a first signal voltage based on photoelectric conversion, and the second pixel unit generates a second signal voltage based on photoelectric conversion. The quantization circuit includes a first quantization unit and a second quantization unit, which correspond to the first pixel unit and the second pixel unit, respectively. It is used to quantize the first signal voltage and the second signal voltage through a ramp voltage, and to correct the first signal voltage and the second signal voltage based on the difference between the first signal voltage and the second signal voltage, thereby improving the deviation between the actual value and the ideal value.

2. The image sensor according to claim 1, characterized in that, The first pixel unit performs photoelectric conversion on the first color light to obtain a first charge and generates the first signal voltage, and the second pixel unit performs photoelectric conversion on the second color light to obtain a second charge and generates the second signal voltage, wherein the first color light and the second color light are different colors.

3. The image sensor according to claim 2, characterized in that, The transmittance of the first color light is greater than that of the second color light, and the first charge is greater than the second charge. Due to the influence of the parasitic capacitance between the first pixel unit and the second pixel unit, the actual value of the first signal voltage is greater than the ideal value, and the actual value of the second signal voltage is smaller than the ideal value.

4. The image sensor according to claim 3, characterized in that, The first pixel unit and the second pixel unit are two adjacent pixel units belonging to the same pixel row, wherein the first color light includes green light and the second color light includes blue light or red light.

5. The image sensor according to claim 1, characterized in that, The first quantization unit corrects the first signal voltage based on the product of the difference between the first signal voltage and the second signal voltage and a first correction coefficient; the second quantization unit corrects the second signal voltage based on the product of the difference between the first signal voltage and the second signal voltage and a second correction coefficient; wherein the first correction coefficient and the second correction coefficient are equal or unequal.

6. The image sensor according to claim 5, characterized in that, The actual value of the first signal voltage is larger than the ideal value, and the first signal voltage is corrected by addition. The first quantization unit includes a first comparator, a first decoupling capacitor, a second decoupling capacitor, a first correction capacitor, and a second correction capacitor. The non-inverting input of the first comparator receives the ramp voltage through the first decoupling capacitor, the inverting input of the first comparator receives the first signal voltage through the second decoupling capacitor, the non-inverting input of the first comparator also receives the second signal voltage through the first correction capacitor, the inverting input of the first comparator also receives the first signal voltage through the second correction capacitor, and the output of the first comparator outputs the first quantization result.

7. The image sensor according to claim 6, characterized in that, The first decoupling capacitor and the second decoupling capacitor have the same capacitance value, and the first correction capacitor and the second correction capacitor have the same capacitance value.

8. The image sensor according to claim 5, characterized in that, The actual value of the second signal voltage is smaller than the ideal value, and the second signal voltage is corrected by subtraction. The second quantization unit includes a second comparator, a third decoupling capacitor, a fourth decoupling capacitor, a third correction capacitor, and a fourth correction capacitor. The non-inverting input of the second comparator receives the ramp voltage through the third decoupling capacitor, the inverting input of the second comparator receives the second signal voltage through the fourth decoupling capacitor, the non-inverting input of the second comparator also receives the first signal voltage through the third correction capacitor, the inverting input of the second comparator also receives the second signal voltage through the fourth correction capacitor, and the output of the second comparator outputs the second quantization result.

9. The image sensor according to claim 8, characterized in that, The third decoupling capacitor and the fourth decoupling capacitor have the same capacitance value, and the third correction capacitor and the fourth correction capacitor have the same capacitance value.

10. The image sensor according to any one of claims 1 to 9, characterized in that, The image sensor also includes a ramp circuit for providing the ramp voltage.

Images