TW201110685A - Electronic image sensor, method for correlated double sampling in electronic image sensor, method for correlated double sampling, snapshop and simultaneous electronic shutter action in electronic image sensor, and imaging system with electronic image sen - Google Patents

Abstract

An electronic image sensor with a pixel array of a plurality of active pixels is provided. Each of the active pixels includes: a photo detector, providing a sensing node for producing a signal based on an amount of light incident thereon; a storing node for storing a plurality of photo-generated charges according to the signal; a first controllable potential barrier between the sensing node and the storing node; an outputting node; and a second controllable potential barrier between the storing node and the outputting node, wherein each of the sensing node, the storing node and the sampling node is not overlapped.

Description

201110685 VI. Description of the Invention: [Technical Field] The present invention relates to a complementary metal oxide semiconductor (CMOS) image sensor (hereinafter referred to as a CMOS image sensor), and particularly provides a CMOS image sensor electronic A synchronous electronic shutter action (SESA) frame storage function and a correlated double sampling function device and related methods. [Prior Art] A digital camera is an electronic product widely used today, and in a digital camera, an image sensor for converting light into electric charge; an image sensor can be classified into a charge according to a principle adopted by the image sensor. A device (charge_c〇Upie (j device) image sensor (also known as a CCD image sensor) and a CMOS image sensor, wherein the CMOS image sensor is fabricated based on complementary metal oxide semiconductor technology. For CMOS In the case of a sensor, a pixel is an element in an image sensor that produces output signals of different output intensities, and the intensity of the output of the pixel is proportional to the light of the person. Intensity, in addition, each element in the image controller is used to detect, store, and output signals. In general, image sensory stealing usually makes the aetive pixel domain passive ((4) (four) pixels come in. 201110685 • Line image sensing; in short, a pixel with a magnified n (ampHfler) or signal buffer can be called an active pixel, and only contains a photodetector and a switch. The pixel is called a passive pixel. For a CMOS image typically composed of active pixels, each line pixel contains money to sense the light diode and to maintain the measured signal. Parasitic capacitance. Please refer to Fig. 1 'Fig. 1 is a schematic diagram showing the structure of a conventional active pixel. As shown in Fig. 1, active pixel 1 (8) includes three nodes 10 and 103, and node 101 is used for The sensing node of the detection signal, the node 1〇3 is a storage point for storing the signal, and the node 105 is a sampling point for outputting the signal. In different designs, the sensing point (ie, node 1〇1) The storage points (ie, nodes 1〇3) and the sampling points (ie, nodes 105) can be overlapped or separated from each other; that is, a single node can be selected or used with each other depending on different design requirements. The three nodes that are separated from each other perform the detection, storage, and sampling operations of the image sensor. In Figure 1, the active pixel 1 further includes an amplifier 106 for the sampling point ( Lang point output The signal is amplified to generate an output 107, which is then transmitted to other circuitry of the digital camera (not shown) for subsequent signal processing. In addition, the active pixel 1 can also have a gate. Gateways 102 and 1〇4, the gateway 102 is located between the node 1〇1 and the node 103, and the gateway 1〇4 is located between the node 1〇3 and the node 1〇5. The gates are known to those skilled in the art and will not be described here. 201110685 For CMOS image sensors using active pixels, the correlation of double sampling (^related doub丨e sampling) operations is often used. Eliminate the low-frequency noise of the cm〇s image sensor, and the operation principle of the correlation double sampling is as follows: · In the operation process of correlation double sampling, at time t], a first reading can be obtained first. The voltage signal (v〇ltagesignaireadout), where V! is the noise present at the sampling point (node 1〇5). At time (4), (called = one), a signal is transmitted to the sampling point, and then obtained - the second signal is read: v_ = Vn2~. In other words g = in the phase double sampling operation, the first (four) out of operation gamma to read the miscellaneous ^ and the second - reading operation is used to read the noise (Μ and the required signal (five). The correlation double sampling operation is performed by subtracting the second second sampling value from the first reading operation, and the following signals are obtained as follows. The signal Δν, the sample value, is Δν. V〇tt, 2 ~ V〇un = (V»2 + ~ (v«.) = (vm2 - vsl) + (~-ν„·) In a small hour, the value of the town will approach zero; in other words, the broken signal of the state is that the correlation double sampling operation is visible to the high-pass data. When the length of time △, the smaller, the cutoff _= ^Low-frequency noise 円, can further suppress noise in 201110685, and therefore, usually shorten the time interval between the two read operations as much as possible. For image sensors with active pixels, usually an ideal Correlation double sampling operation time_separation~ can be fielded (mieiOse_s) or shorter. However, a master of executable correlation double sampling operation The pixel will need to have two different storage points for the domain storage operation. One of the two storage points can be used to reduce the side or the fiber. It is necessary to provide a complete correlation double. The sampling operation will require the related operation in the readout circuit of the back end. These subsequent operations will generate more information. For pre-sensing, please refer to Figure 2, and Figure 2 shows the conventional image sensing. The sound map of the device. As shown in Figure 2, the image sensing includes a pixel array (4) (4) Z2T210 ## M^t(c〇iu-^^N mrow) 4 pixel 215 'where each column is labeled 214 Each line is marked with 16 and the line processor 22〇 is respectively coupled to the line of illusion 22, which can only process one column of 214 data at a time, and the gamma is required to be gamma h (four) ~ like L (such as image Sensor 200) The mother-person performs only one reset operation or one read-out operation = executing the coffee side fiber, the lion lion === for the immediate levy == exit mechanism. However 'usually for a single column 214 For readout, you will need a thousand-column image sensor frame (f^r she) or a longer time to get an entire image object; These long-term readout times will lead to dynamic effects, and the shortcomings of these 201110685 and other photographic equipment are that today's digital cameras (which include image sensors) need to be eliminated/avoided. s. In fact, the [image/photography device] usually uses snapshots (snapsh〇t〇peration) to operate with the sub-synchronization gate (slmultane〇usekc_utteracti〇n, sesa) Figure 22 arestGmge). All the start-stop exposures in the sensor: For the image sensing 11, the control of the exposure is to turn the green speed gate _(10), which is called the === gate. (Frame storage) operation, for the sake of simplicity, the subsequent storage will operate as an electronic synchronous shutter; as described above, the pixels of the electronic ttri are image sensing at the same time. However, the current electronics offer users an image sensor that provides both true correlated double sampling and = step shutter functionality. On the other hand, as mentioned above, the main == different design requirements and different architectures, for example, by three electrons (hereinafter referred to simply as cardiac pixels), by four transistors 2, _tree_ ), for example, the optical gate active image is expected to be an active pixel composed of five electric three-day bodies (hereinafter referred to as 5 Τ active pixels). The following will pay for these non-social pixels. Figure: Figure 3 of Figure 3 shows the structure of the moving pixel. The pixel has three crystals to win. The lion has provided a non-sense system for the measurement, signal storage and signal output. Operation. Of course, 3Τ active pixels·201110685 Those who are familiar with this technique have other components. Since the structure of 3T_pixel is well known, it will not be described here. See Figure 4 and Figure 3, Figure 4 is the timing signal of the CM〇s image sensor residuals of the pixel array of the active pixel in Figure 3

Thinking. As shown in Figure 4, in this example, CM is also like the pixel array of the sensor.

The column system uses the "rolling wheel" (rolling type) mechanism, and the column wheel (or wheel) operation has been explained in the previous wheel reset/read operation. Here, the = time sequence of each column will be near the other. _ 灭, but every time - Shun riding - _ signal timing is late for a while. For the _every-column, two timing coffee (four) signals, ie, the reset signal and a list of selection signals, are required to perform a non-correlation (_-Correlated) double sampling operation. For example, in Figure 4, the operation of column 1 of the CM〇s image sensor will be described as follows, wherein the CMOS image sensor uses three active pixels 300 to form a pixel array. The operation of each column (for example, column) includes the following steps:

Step 1: Perform a reset operation. First, the reset transistor 31 is turned on (that is, as shown in FIG. 4, a reset signal 310A generates a pulse 1) to reset the node 3〇8; then the reset transistor 310 is turned off to start a deposition time length. The charge deposition operation causes ^(1)^^^+~' where the charge is centered to represent the voltage of a noise signal, and the reset signal 310A represents the voltage signal applied to the gate of the reset transistor 31〇. 201110685 Step 2: The CMOS image sensor performs a first read operation on column i to sample the signal and the noise, and thus extract a preliminary round-trip voltage V. = "And this preliminary output voltage is indicated as V!(2) in Figure 4 of Figure 4, which indicates the voltage signal obtained after the transistor 3 is selected in the conduction column (in the 4th column, the pulse 2 of the signal 306A is selected) That is, it is indicated that the gate of the column selection transistor 306 is applied to activate the transistor 3〇6). Step 3: Restart the reset transistor 31 to perform a second reset operation on the node 3〇8. That is, at this time, the reset signal 310A generates a corresponding pulse 3) and thus obtains a voltage ^(3)=v_2 + v„2, wherein the voltage %(3) represents the voltage signal obtained at this time. Step 4: Execute - the second read operation to complete each column (for example, the required operational flow of the column. At this time, 'a second reset signal level will be obtained in the second read operation' and % is taken. The voltage is expressed as V (mi2=v_2+Vn2, wherein the voltage vout2 represents the voltage signal captured after the column selection transistor 3〇6 is activated.] At this time, the column selection signal 306A has a corresponding pulse 4, that is, The column selection transistor 306 is activated. The foregoing four steps complete the flow of the read operation of a single column in a complete frame, and after completing the foregoing four steps, the output voltage of the pixel can be obtained. For: = v(10)2 =, + (vr_ _ v_) + (~_ Vn2). 201110685 Because of the %, ~ is close to zero, but the signal difference will be equal to Γ", -νϋνη, where? 姥佶r Shishan Du, θ is very miscellaneous. Since the two take # at this time, the final output voltage obtained by this non-phased double sampling operation will be vau, = Kig + 10,000. And 3T active pixel 3 (8) due to its non-correlation The double sampling operation is limited, so the active pixel lion is only by the single-section Execution signal age storage is #, sample 'read' causes the accumulated electricity generated by the pixel during the first read operation, and is destroyed in the second reset operation. See Figure 5 Figure 5 is a schematic diagram showing the structure of a conventional active pixel. Compared to the 3Τ active pixel 3〇〇 in Fig. 3, the 4Τ active pixel 5(8) shown in Fig. 5 is more-turned for signal processing and The signal is stored, and the other node 504 is used for signal sampling. However, the pixel array consisting of the active pixels of the lamp must still be read in a column wheel manner. Figure 6 and Figure 6, Figure 6 is a timing diagram of the signal of the 4T active pixel 500 shown in Figure 5. The 4T active pixel 5 can perform a true correlation double sampling operation, and its operation is based on a corresponding one One of the reset transistors 310 重置 corresponds to the 歹 select transistor 3 〇 6 column select signal by a, corresponds to a photodiode 502 - photodiode signal 5 · and corresponds to a transfer room (10) ♦ The transmission of gate 503 八 ' can also be adjusted here Instead, the photodiode is used instead of the photodiode 5〇2. As shown in the figure, when the reset point of the sampling point 5〇4 is 11 201110685 =1, the photodiode signal will be caused by the charge deposition operation. The voltage of the fiber = increase, and before the end of the charge deposition, the reset _ pass to control the reset transistor training will generate a pulse change to sample the sampling point 5 〇 4, which causes the generated noise to be captured. In the sampling point 504. v In the first reading operation, a noise can be sampled, which can be expressed as follows: Next, the voltage of the sfl number 503A is raised to start the transmitting transistor 503 and the photodiode signal 5 is still The voltage drops (as shown), so that the charge stored in the light body 502 can be transferred to the sampling point 5G4. The voltage (including signal and noise) available for the second read operation can be expressed as: ν <?,"2Ί+ν" so that you can get a final based on the two signal readings related to each other. Output voltage value: ν.丨"=ν-2-νβω丨. By constructing a CMOS image sensor with a 4-inch active pixel 500 with correlation double sampling operation function, the output signal can be effectively prevented from being interfered by noise. However, since the 4T active pixel 5〇〇 still uses a single node to perform signal detection and signal storage operation, the CMOS image sensor composed of the 4T active pixel 500 still cannot provide the function of synchronizing the electronic shutter.

Please refer to FIG. 7. FIG. 7 is a schematic structural view of a conventional active pixel 800 having an electronic synchronous shutter function. The structure of the active pixel 8〇〇 is disclosed in the paper “ASnapshotCMOS” published by Guang Yang et al. in IEEE in 1998.

Active Pixel Imager for Low Noise, High Speed Imaging,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, Shutter function. As shown in FIG. 7 - L main left-moving pixel 800 includes a reset transistor 31 一, a column of selection transistors 306, ... 5 (U, - the first transfer transistor plus and a photodiode tear). Pixel_ uses the same node 806 to perform signal storage operation and signal fetching, which means that the charge of the coffee is immediately transferred to a floating diffusion (f〇3 nodes 8〇6) 'This single node is also a storage point. For the sampling point, the operation of the electrical deduction will immediately be sampled immediately. Fine, active pixels _ still lacking _ sex double _ ship, and the quantum efficiency of the wire pixel _, especially its quantum efficiency for blue light In general, 'in order to achieve the aforementioned function of the synchronous electronic shutter, the active pixel needs to maintain (5) d) the signal it senses until it is read; however, the usual number of beakers may be up to several Ten milliseconds, but in the electronic image sensing system, there is no mechanical (four) 'make; ^ in this period of maintenance of the copper towel, the human light will continue to lead to the generation of charge' and in order to solve the pre-production problem, Use a simple charge picker inside, The high voltage source is used to take electrons (or a low voltage source to draw holes) and the switch is connected between the aforementioned voltage source and the sensing point. In addition, in the US patent disclosed by Merrill et al. (Patent No. Na Nae, it teaches Lang-Heavy 1_ and - resets the Laisaki charge before resetting the "light-pole" and is shouting (4) While maintaining the time to provide the function of the charge eliminator, Merrill's invention still does not provide the ability to perform double sampling. 13 201110685 As mentioned earlier, due to the deficiencies and shortcomings of the traditional architecture, a novel process must be provided. The system uses the CMOS bokeh green bristles to provide a digital camera with excellent correlation double sampling function and synchronous electronic shutter function. SUMMARY OF THE INVENTION According to an embodiment of the invention, an image sensor is disclosed (dectr〇) Nic image sensor), which has one pixel array including a plurality of active pixels, wherein each active pixel includes: a photodetector, which provides a measurement point to generate a light according to the intensity of an incident light. a storage point for storing a plurality of photo-generated charges according to the signal; a first controllable potential barri Er) ' is located between the sensing point and the storage point, an output point; and a second controllable potential energy barrier is located between the storage point and the output point, wherein the sensing point, The storage point and the output point are not overlapped with each other. According to another embodiment of the present invention, a method for correlated double sampling (CDS) of an image sensor is disclosed. The image sensor has a pixel array composed of a plurality of active pixels, each active pixel having a photo detector. The method comprises the steps of: integrating a plurality of photoinduced charges; Resetting a signal sampling point; performing noise sampling during a read-read operation; transmitting the plurality of photo-induced charges to the gas sampling point 201110685 - and performing a second reading operation to borrow A signal is extracted by charge sampling. According to still another embodiment of the present invention, a correlation double sampling (CDS) and a synchronous electron are performed. A method of a simulated electronic shutter action (CDS), the image sensor having a pixel array formed by a plurality of active pixels, each active pixel device having a photodetector to be incident on The intensity of the light on the pixel array generates a signal, the method comprising: accumulating a plurality of photoinduced charges according to the signal; storing the plurality of photoinduced charges until a read operation; turning on a charge extractor (charge Sink) to capture a plurality of input photo charges; reset a signal sampling point; perform noise sampling during a first read operation; pass the plurality of photoinduced charges to the signal sampling point; and turn off the charge The picker pre-resets (pre_resetting) a plurality of nodes. A further embodiment of the present invention is disclosed in the accompanying drawings - an image system having an image sensor having an array of pixels including a plurality of active pixels, wherein each active pixel includes a light detector (ph〇t〇detect〇r), which provides a sensing point to generate a signal according to the intensity of an incident light; a storage point for storing a plurality of light based on the 3 Xuan afl number (ph〇t〇_generate(J) charge; a first controllable potential barrier between the sensing point and the storage point; an output point; and a second a potential-controlled energy barrier, located between the storage point and the output point, wherein the sensing point, the storage point and the output point are not overlapped between each other 15 201110685. According to the spirit of the present invention, A pixel structure is disclosed in which a plurality of pixels are arranged in each other to perform fine measurement, signal storage, and signal sampling, and in addition, there is a further description in the pixel (four); and the pixel system of the present invention can be CMOS image sensor Efficient synchronous electronic shutter function. According to the (4) shot, its flip-fresh structure, the pixel node of the present invention can be operated by a spiUwdl structure to perform a correlation-relevant sampling operation; It has better pixel efficiency than the pixels of the previous ccd pixel or the shutter form. The spirit of the present invention provides a pixel, a pixel array, and an image sense of the front navigation structure. It is possible to provide a correlation double = function and synchronous electronic shutter function. In addition, the overflow structure of the present invention is such that: a first-pole body is used as a photodetection to enhance the quantum efficiency provided by the fourth (four), so that The invention is also capable of being implemented by a standard complementary MOS process. [Embodiment] 201110685 • A transistor is used as a switch, a shisha (lightshield) 909 and a full-deficient (ft% depleted) photodiode 914. By providing these transistors, the active pixel 900 can have sensing nodes, storage points, and output points that are separated from each other to provide synchronous power. Shutter function and correlation double sampling function. However, please note that the structure of Figure 8 is for illustrative purposes only and should not be considered as a limitation of the present invention. For example, the column selection transistor 9 in the figure 〇4 and hood 909 are optional components that can be omitted with different design requirements. In addition, for the efficiency φ quantity, here a full depletion diode is used instead of conventional light. Gate structure. However, please note that as long as the adjustment is properly adjusted, a general optical gate or photodiode can be used instead of the full-vacancy photodiode 914 of the present embodiment to provide a synchronous electronic shutter and related The pixels of the double sampling function, the aforementioned design changes are in accordance with the inventive spirit of the present invention and fall within the scope of the present invention. In the present embodiment, the active pixel 900 includes a charge extractor 91, wherein the charge extractor 910 includes a transistor 911 (ie, TX3) for use in a light sensing node 914A and a charge. A potential barrier (potentialbaiTier) is formed between the eharge drains 915. In addition, the active pixel 9 is further provided with a spill well 906' having a transistor 912 (ie, TX2) therein, and a transistor 192. The system is used to form a potential energy barrier between the sensing point 914A and the storage point 917, and the transistor 918 (ie, TX1) forms a potential energy barrier between the storage point 917 and the output point 922; and the transistor 919 is Used herein as a reset transistor for selectively coupling a reset voltage % to output point 922. In addition, transistor 9〇5 is used as a source-follower amplifier for signal buffering. As shown in Fig. 9, 17 201110685 - continuation __. like scale _ output bus bUS) ' and column selection transistor called image transmission _ selection of motion can be seen as a switch. In this embodiment, the voltage value of the Va can be a fixed voltage, or the value of the dragon can be different, and the dragon value can also be different. It can share the same transistor 9 ιι (τχ3), early-source 905 (that is, the source with the consumer), a single node (the node supplying the voltage va), and a single-voltage node (the node supplying the voltage). In order to save production costs and Wei _. If the light value of the system is allowed to be different under different steps, the active pixels in the design change can be shared by a plurality of active pixels 900 in the pixel array to share a single column selection transistor class. However, please note that the above-mentioned selection of the transistor 9〇4 is an optional component, which may also be omitted in other embodiments, and these devices are also in accordance with the spirit of the present invention and are protected by the present invention. In the bounds. In this embodiment, the voltage value of the voltage Va is set to be less than or equal to a voltage value _ Vdd ' and the voltage value of the voltage Vsink is preferably set higher than the voltage % j (ie, TX3). The voltage on it. In some embodiments, the voltage value of the voltage Vsink is at least one transistor threshold voltage higher than the voltage applied to the transistor 911 (TX3), that is, since the voltage vsink is used as a A charge picker, which can be a high voltage source (such as Vdd). In other embodiments of the present invention, the active pixel 900 may be applied in accordance with different designs; however, in the eighth figure, the dream is constructed by the ##-type transistor in the ## period. Pixel_. In addition, the hunting here consists of ^ (4) (10) on the transistor 913 as the anti-small layer of the 弁 曰 ( ( ( ( ( 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 电荷 以 电荷 电荷To perform the doubly double sampling function, the electron 911 (TM) can be kept in the off state (non-conducting state) when the electron energy is used. Please refer to Fig. 8 and Fig. 9 at the same time, and Fig. 9 is the eighth figure. The active pixel shown = 运作 does not perform the operation of the electronic lang operation. In order to clearly expose the step flow of the ninth figure, the active pixel __ structure is attached here to the top of the figure 9 As an auxiliary explanation, the month is also referred to FIG. 8 and FIG. 9 to see the first Q diagram. FIG. 1G is an embodiment of the present invention, and the timing of the plurality of signals in the judgment step is performed. Not intended; FIG. 1G includes a reset 峨 919A for application between the reset transistors 919 for application to the gate of the transfer transistor 9 8 — Tiger 918A, a signal 913A for application to the gate of the transistor for application to the transmitting transistor 912 A reset signal 9i2A of the gate, a signal for applying to the gate of the column selection transistor 9〇4, and a signal 9UA for indicating the signal 911A for application to the transmission transistor 911. Fig. 8, Fig. 9, and Fig. 1G; as shown in Fig. 8, in the subsequent step description of 201110685, the photodiode 914 is fully depleted by the pinning v〇ltage. The photodiode. At the step su〇〇1, the voltage applied to the gate of the transistor 191 is raised to turn on the transistor; after the first reset operation, the active The pixel 9 (8) starts charge deposition (step S1002). "At step S1002, the pixel 900 generates a plurality of photo-induced charges. During the operation of the charge deposition, the reset transistor 919 is turned off and applied. The voltage signal Vst〇 on the gate of the storage transistor 913 (i.e., ST0) is pulled high. In this step, a potential well (_lion read) is used to store the state produced by the full-empty photodiode state. Charge. As can be seen from Figure 9, although after resetting There are some = nl remaining in the output node 922, but via the transmission transistor 9 ΐ 8 (τ χ ι) at this time is located at the output node 922 and stored in the external image + M ^ i $ Cang point 917 potential barrier, It will ensure that the first-cut operation of the pixel_ will not be affected by the noise η. As shown in Figure 9, the operation continues in step 2. In this step, the product has been completed and the message is 922 ^ £ 〇 919. (10) Closing and performing the first-time material, when the step 4 is repeated, resetting the sampling generated by the transistor _2, it can be seen that the noise n2 stored in the step _〇3 is stored in the back-end reading 922, during the first collapse operation That is to use the operation. 3 (not shown in the figure) for correlation double sampling 201110685 In step s1 所示 5 shown in Fig. 9, the charge transfer and the second read operation are performed 'At this time, the storage transistor 913 is turned off to put The stored charge is transferred from the storage node (1) 7 to the output node 922, while the charge in the output node 922 contains the sensed signal and the noise!! After the charge transfer and sampling, the second item is operated. By taking the sampling data obtained during the two reading operations, the sensing signal can be smoothly taken out without being affected by noise. φ In the present embodiment, the gate voltage of the transfer transistor 918 (TX1) is maintained at a certain voltage value V]' and the gate voltage of the transfer transistor 912 (TX2) is maintained at another constant voltage value V2 to further Reduce switching noise (switching n〇ise). However, it should be noted that in other embodiments of the present invention (the voltage applied to the transmitting transistor 912 (TX2) in the second embodiment as shown in the figure n will be slightly lowered at step s12 〇 3 to secure it. An effective potential energy barrier is formed between the sensing node 914A and the storage point 917. All of the above related design changes are in accordance with the inventive spirit of the present invention and are also within the protection of the present invention. In the figure, the correlation phase sampling operation when the pixel 9 不 does not have the synchronous electronic shutter function is explained. In other words, since the asynchronous electronic shutter operates, the operation flow disclosed above is the operation of the column wheel type (__rcmng). Please refer to FIG. 8 and FIG. η, which are schematic diagrams showing the steps of using the active pixel 900 to perform correlation double sampling operation and synchronous electronic shutter operation in another embodiment of the present invention. In order to clearly explain the details of the operation of the active image like Xin·201110685, the active pixel 9(8) is touched for comparison in the u__. Please refer to Fig. 12, which is shown in Fig. 12. A timing diagram of a plurality of signals in the embodiment shown in the figure; in FIG. 12, a reset signal 919A for applying to the gate of the reset transistor 9191 is included, and the application of the transfer transistor 9 is called τχι). The signal 918 of the pole, the signal 9 applied to the gate of the storage transistor 913, the signal 912 对应 corresponding to the transmitting transistor 912 (ΤΧ2), the signal 904 对应 corresponding to the gate of the column selection transistor 9〇4, and a voltage In the second embodiment, steps S1201, S1202, S1204, S1205, and S1206 # are the same as steps S1001, S10O2, S1003, S1004, and S1005 in the first embodiment, respectively, so detailed operation details regarding these steps may be Referring to the foregoing description of Fig. 9. In summary, at step S1201, the reset transistor 919 is turned on to perform the first reset operation, and is switched to the off state before proceeding to step S1202. At step S1202 The storage transistor 913 is turned on, and the photodiode 914 of the present embodiment is a #full depletion photodiode, so that the charge generated during the charge deposition operation can be smoothly stored in the storage node 913. In a preferred embodiment of the present invention, the voltage applied to the signal 912A of the transmitting transistor 912 (TX2) can be slightly boosted at the end of the charge deposition operation to enhance the sensing node 914A and the storage node 917. In the present embodiment, the active pixel 9 system simultaneously provides the correlation double sampling function and the function of the synchronous electronic shutter, so when the signal is maintained (datah〇kling) operation, it can correspond to the transmission. The signal 911A of the transistor 911 (TX3) is pulled low to 22 201110685 « • The excess charge generated by the sensing node 91 at this time is discharged. That is, since the signal 912A corresponding to the second transfer transistor 912 is pulled up at step S1203, and since the transfer transistor 911 (TX3) is turned on, this causes the photodiode 914 to be subjected to a voltage Vsink. Clamping to ensure that all additional excess photoinduced charge will be drawn by the voltage, while the photoinduced charge generated by the exposure in the previous charge deposition operation is stored in the storage transistor. Here, the potential barrier formed by the transistor 912 and the hood 909 ensure that the stored data is not subject to external interference. As for the charge picker (ch which sink) composed of the power transmission body 911 (TX3) and the voltage 'toe', these frames ensure that all active pixels in the pixel array can simultaneously end the charge deposition. Operate and save its data effectively until the signal is read out. In other words, the pixel of the present invention thus provides an excellent synchronous electronic shutter function. v The reset transistor 919 in step S12G4 is turned on again and again for the first reset. In step S1205, after the reset transistor 919 is turned off again, the external transistor 904 is turned on to perform the first read operation. In step (10) (iv), the transfer transistor 918 is turned on and the transistor 913_ is stored to transfer the charge from the memory node 9Π to the output node 922'. After the charge transfer operation is completed, the transfer transistor % is turned off and the column selection transistor is turned on. 4 is turned on for the second read operation. Note that in the present embodiment, the active pixels 9 in the pixel array are still read in the column wheel mode (face_~81 birds, riding is touched 3~s, and the column is followed by a column.) The foregoing steps are performed in the pixel_sheet-column oblique operation. After the reading operation of 23 201110685 is completed to an entire frame (that is, after the reading operation of each column in the pixel array is completed), The transfer transistor 911 (TX3) is turned off and the reset transistor 919 is turned back on to prepare for the signal sensing of the subsequent frame. Since the photodiode 914 is here a full-empty photodiode, this further ensures All of the photoinduced charges that are sensed can be transferred to the storage node 917 without being trapped in the photodiode 914. Please note that the present invention is not limited to the foregoing embodiments, and may be followed by subsequent technologies. Improvements are made and these design changes are also within the scope of the invention. For example, the potential energy provided by the transfer transistor 918 (TX1), the transfer transistor (τχ2), and the transfer transistor (ΤΧ3) Can also have obstacles The same operation mode, and the change of this implementation will be explained in Fig. 13. Fig. 13 shows another embodiment of the present invention in which the social pixel 9 is simultaneously subjected to correlation double sampling. In the embodiment, the steps S1401 to S1407 are the same as the steps _ S1201 to S1207 of the 11th figure respectively. Please refer to Fig. 13 'The operation of the step sl4〇1 is similar to Step S1201 of Fig. 11. 'In step S14G2, the transmission transistor 918 (τχι)_ pole is biased at a voltage 'the voltage value is lower than the voltage Vn-w in the second diagram (eg, 13th) FIG. 11 and FIG. 11. In the present embodiment, the lower voltage Vi3jxi can completely turn off the transmitting transistor 918 (TX1) and apply it to the pole 24 of the transmitting electric day and day 918 (τχι). The voltage of 201110685 is maintained until step S1206 is executed, wherein step sl4〇6 also approximates the operation of step S1206, so it is not described here. The signal applied to the transmitting transistor 912 (τχ2) is used. 9i2A, in the _ input = material, turn the step S14G3 (its operation is similar In the step S12Q3), the gate of the transfer transistor 912 (TX2) is biased at a voltage % μ, and its voltage value is lower than the voltage in FIG. 11 (as shown in FIG. 11 and FIG. 13). The signal 9 applied to the gate of the transmitting transistor 912 (ΤΧ2) is maintained at voltage %(10) until step s14. 6 is completed. According to the spirit of the present invention, an innovative active pixel is disclosed. A plurality of nodes separated from each other are used for signal detection, signal storage, and signal output, respectively. Further, the active pixel of the present invention also has a charge extractor, and the present invention can use a complementary gold-oxygen half process. Providing an active pixel 0 having a function of synchronizing an electronic shutter - according to the spirit of the present invention, a correlation double sampling function is performed by means of an overflow, and a structure, in addition to a conventional charge-engaging device (CCD) The technology and the inter-optical form of the pixel, the _photodetection _ also has better quantum efficiency. " In accordance with the spirit of the present invention, there is provided a mine moving pixel that can simultaneously perform correlation double sampling force b and a synchronous electronic shutter function, and an image formed by the active pixel described later. In the present invention, the H-body is manufactured as a photodetector 25 201110685 a in order to achieve a better lining efficiency than the conventional technique of using a charge-for-good device and the conventional inter-optical form pixel. In addition, in a preferred embodiment of the present invention, the light-polar system of the present invention is a full-empty (5)-d-printed (4) photodiode. Furthermore, the structure of the present invention is fully compatible with the complementary MOS process. The above description is only a difficult embodiment of the present invention, and all the equivalent changes and repairs made by the scope of the present invention should be within the scope of the present invention.

[Simple Description of the Drawing] Fig. 1 is a schematic view showing the structure of a conventional pixel. Figure 2 is a schematic diagram of a conventional image sensor. Figure 3 is a schematic diagram of the structure of a conventional 3T active pixel. Figure 4 is the active active pixel shown in Figure 3 with N columns.

A schematic diagram of the timing of the image sensing n. WCM0S Figure 5 is a schematic diagram of the structure of a conventional 4T active pixel. Figure 6 is a timing diagram of the signal of the 4T active pixel shown in Figure 5. A schematic diagram of the structure of the active active pixel of the 2-step shutter function. Fig. 8 is a view showing the structure of a pixel according to an embodiment of the present invention. Fig. 9 is a view showing the operation flow of the active pixel in the eighth embodiment. When the sub-decision is in operation, FIG. 10 is a timing diagram of the plurality of signals in which the active pixels are in the present embodiment in the embodiment of the present invention. Steps for knowing 26 201110685. The 11th _ Ben (4) is another - the actual performance of the social dynamics of the operation and the synchronization of the electronic lang operation of the process of the first step of the 仃 阙 双 第 第 第 第 第 第 第 第 第 第 第 第 第 第 第 第 第 第A timing diagram of a plurality of signals. Fig. 13 is a schematic diagram showing the steps of using the active pixels to simultaneously perform the related f-sampling operation and the synchronous electronic shutter operation in still another embodiment of the present invention. [Main component symbol description] 100, 800, 900 active pixels 101 sensing node 102, 104 gateway 103 storage point 105 sampling point 106 amplifier 200 image sensor 210 pixel array 214 column 215 pixel 216 line 220 line processor 300 3T Active pixels 306, 904 column select transistor. 309, 905, 911, 912, 913, 918 27 201110685 310 reset transistor 308, 504, 806 node 500 4T active pixel 501 source with handle 502 light diode 503 transfer Transistor 802, 910 electrical extraction 906 overflow 909 hood 914 full-empty optical diode 914A light sensing node 915 charge extraction end 917 storage node 922 output node

28

Claims (1)

201110685 VII. Patent application scope: h An electronic image sensor having an array of pixels including a plurality of active pixels, wherein each active pixel comprises: a photo detector Providing a sensing point to generate a signal according to the intensity of an incident light; a storage point 'for storing a plurality of photo-induced charges according to the signal; · a first controllable potential energy barrier (controllable potential barrier) between the sensing point and the storage point; an output point; and - a second controllable potential energy barrier 'between the storage point and the output point, wherein the sensing point The storage point and the output point are not overlapped with each other. (2) The image sensor of claim i, wherein the photodetector system is fully depleted (ftjllydepleted) Light diode. The image sensor of the second aspect, wherein the first __ controllable potential energy barrier and the second controllable potential energy barrier are selectively connected to the pixel array A voltage source other than that. 4. The image sensor of claim 2, wherein each _ active image comprises: 29 201110685 a charge collector, which is selectively coupled to the photodetector, And the voltage source '5 Xuanhe He> and the picker system provide a j: and sinking node to extract a plurality of photo charges, and provide one between the photodetector and the electric source The third controllable potential energy barrier. 5. If she applies for the image sense described in item 4, the charge is not taken over and the voltage source is shared by a plurality of active pixels. 6. The image sensor of claim 2, wherein the storage point is controlled by an external signal to transfer charge upon operation. 7. The image sensor of claim 1, wherein the pixel array comprises a source-slave light-receiving device and a column-selective transistor, and the column-selective electro-crystal system is shared by a plurality of active pixels. . 8. A method for performing correlated double sampling on a shirt image sensor ((7) ting ubiesampiing, ruth CDS). The image sensor has a pixel array consisting of a plurality of active pixels. Each active pixel system has a The photodetector (the method includes: integrating a plurality of photoinduced charges; resetting a signal sampling point; performing noise sampling during a first readout; Transmitting the plurality of photoinduced charges to the signal sampling point; and 30 201110685 • performing a second read operation to extract a signal by charge sampling. 9. If the application is to be handled, please refer to the method described in item 8. The method of claim 8, wherein the first read operation is performed on a sampling point of the signal, wherein the photo-electric charge is formed after the formation of a potential energy barrier. The noise is sampled. • 11 · - Pair-image sensor for correlated double sampling (eGrrelated) and synchronous electronic shutter (simultan (10) selectr〇nicshutteraetk^ (CDS)) The detector has a pixel array formed by a plurality of active pixels, each active pixel system having a detector to generate a signal according to the intensity of light incident on the pixel array, the method comprising: The signal accumulates a plurality of photo charges; • stores the plurality of photo charges until a read operation; turns on a charge sink to capture a plurality of input photo charges; resets a signal sample point Performing noise sampling during a first read operation; transferring the plurality of photo-induced charges to the signal sampling point; and turning off the 5 meta-charges and taking the state to pre-reset (re_resetting) the plurality of nodes. 12. An image with a - image sensing n, the image system H has a pixel array comprising 31 201110685 plurality of active pixels, wherein each active pixel comprises a photo detector Providing a sensing point to generate a signal according to the intensity of an incoming light; a storage point, (4) storing the re-oil photo-induced (ph.ge_(4) charge; a first controllable potential energy barrier (control丨ablep) 〇tential barrier) between the sensing point and the storage point; an output point; and a first-controllable potential energy barrier 'between the storage point and the output point, The sensing point, the storage point and the output point are not overlapped with each other. [13] The image system of claim 12, wherein the photodetector is a full-floating type (fUllydepleted) light The image system of claim 12, wherein the first controllable potential spring barrier and the second controllable potential energy barrier are selectively coupled to the pixel array a voltage source. 15. If applying The image link described in claim 12, wherein each of the active pixels further comprises: - an electric charge H (ehargesink), which is a _ _ _ photodetector and a - voltage source 'the charge: 3⁄4 take the II system to provide _ Snap point (sinking (10) as) 32 201110685 • Take a plurality of photoinduced charges and provide a third controllable potential energy barrier between the photodetector and the voltage source. The image system of the invention, wherein the charge collector and the voltage source are shared by a plurality of active pixels. The image system of claim 12, wherein the storage point is operated by a The image system is controlled by an external signal. The image system of claim 1 is the image system of the invention, wherein the pixel array comprises a hopper and a column selection transistor, and the source _ The column selection electro-crystal system is shared by a plurality of active pixels. 33