Patent Application
20160014366
Examples relate to a charge transimpedance amplifier (CTIA) unit cell for a light sensor, capable of automatically, and effectively, accommodating relatively low and relatively high light levels.
Circuitry for a light sensor is often designed to effectively accommodate a relatively low light level or a relatively high light level, but not both. A circuit designed for a relatively low light level can saturate when used at a relatively high light level. A circuit designed for a relatively high light level can have noise that overwhelms the signal when used at a relatively low light level.
Accordingly, there exists a need for circuitry for a light sensor that can automatically, and effectively, accommodate both relatively low and high light levels.
A charge transimpedance amplifier (CTIA) input cell includes a high gain capacitor configured to integrate charge arising from photocurrent, a low gain capacitor, and a switching element that can switch the low gain capacitor to be electrically coupled in parallel to the high gain capacitor. In some examples, the switching element is a low gain switch, which can be manually activated to switch in the low gain capacitor. In these examples, the low gain switch can be electrically disposed between the low gain capacitor and a source of the photocurrent. In other examples, the switching element is a low gain transistor, which can be automatically activated to switch in the low gain capacitor when a voltage across the high gain capacitor reaches a specified threshold. In these examples, the low gain capacitor can be electrically disposed between the low gain transistor and the source of the photocurrent. The CTIA can be single-sided or can be differential.
The following description and the drawings sufficiently illustrate specific embodiments to enable those skilled in the art to practice them. Other embodiments can incorporate structural, logical, electrical, process, and other changes. Portions and features of some embodiments can be included in, or substituted for, those of other embodiments. Embodiments set forth in the claims encompass all available equivalents of those claims.
There are many types of image capturing devices such as digital cameras, video cameras, and other photographic and/or image capturing devices. These image capturing devices can use image sensors such as active pixel sensors (APS), arrays of photodiodes, or other suitable light sensing devices in order to capture an image. For example, an APS can include an array of unit cells that receives light from a lens. Each unit cell in the array generally corresponds to the smallest portion of a digital image, known as a pixel. The light causes each unit cell to accumulate an electric charge proportional to the light intensity at that location. Circuitry and/or software in the image capturing device then interprets the charge accumulated in the unit cell to produce the corresponding pixel of the final image.
Typically, each unit cell in the array includes a component to store the electric charge until it can be read and analyzed. In some unit cells, this component can be an integration capacitor. The size of the integration capacitor can vary according to the specific application of the imaging device, and is usually selected to accommodate the greatest amount of electric charge expected to be encountered for the application.
Image capturing devices are routinely exposed to both low ambient and high ambient light situations. As a result, it is desirable for an image capturing device to have a high dynamic range, e.g., the ability to perform well in both low ambient and high ambient light situations. In a low ambient light situation such as pictures taken at night, indoors, in shadows, or other situations where there is a relatively low amount of ambient light, the electric charge accumulated in the unit cell will be relatively low. As a result, a relatively small amount of capacitance is needed to store electric charge in low ambient light situations and therefore a relatively small integration capacitor can be desired. Conversely, in high ambient light situations such as a sunny day, a well-lit room, or other situations where there is a relatively large amount of ambient light, the electric charge accumulated in the unit cell will be relatively high due to the greater intensity of the light captured by the image capture device. As a result, a relatively large amount of capacitance is needed to store electric charge in high ambient light situations and therefore a relatively large integration capacitor can be needed.
As mentioned above, most integration capacitors are chosen to accommodate the greatest amount of electric charge expected to be encountered for a specific application. Because of this, integration capacitors tend to be relatively large in size so that they will not saturate and cause a loss of information. This works well for high ambient light situations which generate larger amounts of electric charge, but is less desirable in low ambient light situations where there is a relatively small amount of electric charge to store. In low ambient light situations, there will be a relatively low signal-to-noise ratio due to the lower electric charge. To combat the low signal-to-noise ratio in these situations, a relatively small integration capacitor is more desirable. This creates a dichotomy for unit cell designers: choose a small integration capacitor that will perform well in low ambient light situations but can easily saturate in high ambient light situations, or choose a larger integration capacitor that will not saturate in high ambient light situations but will perform poorer in low ambient light situations.
Additionally, in order to capture an image, most image capturing devices having an integration capacitor must reset the integration capacitor through a switch prior to capturing the image. This reset involves applying a voltage V to both sides of the integration capacitor, so that the voltage across the integration capacitor is set to zero volts. In reality, however, the voltage measured across the integration capacitor after this reset will not be exactly zero volts, but rather will be zero volts plus or minus some small amount of error. This error is known as kTC noise, or reset noise. The effects of kTC noise become significant at relatively low light levels, at which the signal is relatively small.
Accordingly, it would be desirable for a unit cell to perform optimally in both low ambient and high ambient light situations (e.g., to have a high dynamic range) while providing a low kTC reset noise.
Image sensor 102 can be an APS, an array of photodiodes, or any other suitable light sensing device that can capture images. Image sensor 102 can include, for example, a diode, a charge-coupled device (CCD), or any other photovoltaic detector or transducer. Image sensor 102 senses a scene as an array of pixels 104, where each pixel receives light from a corresponding portion of an imaged scene, and produces current in response to the received light.
A read out integrated circuit (ROIC) 106 includes a plurality of charge transimpedance amplifier (CTIA) input cells 108, with each CTIA input cell corresponding to a sensor pixel 104. Each CTIA input cell 108 receives a photocurrent generated by the corresponding sensor pixel 104, integrates the photocurrent for a particular frame duration as a stored charge, and outputs a particular voltage at the end of the frame, the voltage corresponding to the stored charge. The CTIA input cells 108 all work in parallel, with the ROIC 106 assembling and correlating the output voltages from the CTIA input cells 108. Other types of input cells can also be used, including source/follower, direct injection, buffered direct injection, and others.
An image processing unit 110 can convert the assembled and correlated information from the ROIC 106 into an electronic representation of the image incident on the image sensor 102.
Image processing unit 110 can be a combination of hardware, software, or firmware that is operable to receive signal information from the ROIC 106 and convert the signal information into an electronic image. Examples can also be implemented as instructions stored on a computer-readable storage device, which can be read and executed by at least one processor to perform the operations described herein. A computer-readable storage device can include any non-transitory mechanism for storing information in a form readable by a machine (e.g., a computer). For example, a computer-readable storage device can include read-only memory (ROM), random-access memory (RAM), magnetic disk storage media, optical storage media, flash-memory devices, and other storage devices and media. In some examples, computer systems can include one or more processors, optionally connected to a network, and can be configured with instructions stored on a computer-readable storage device.
Plot 304 shows the output voltage VOUT when the light intensity striking the sensor pixel is relatively low. When the reset switch closes, the capacitor is set to a reset voltage VREF. When the reset switch opens, the capacitor begins receiving charge from the photocurrent. The charge is said to integrate on the capacitor (206;
Plot 306 shows the output voltage VOUT when the light intensity striking the sensor pixel is relatively high. The relatively high intensity light striking the sensor pixel produces more photocurrent than the relatively low intensity. As a result, when the switch opens at time 310, the charge on the capacitor integrates more quickly, and the output voltage VOUT drops more quickly. For the relatively high light intensity, the capacitor reaches saturation at time 312, after which the output voltage VOUT remains at a minimum value VMIN. When saturation occurs, the image processing unit returns a maximum light level for the saturated pixel. In practice, saturation is undesirable because high-intensity detail is washed out in the image; all pixels having an intensity greater than a saturation intensity all take on the minimum voltage VMIN.
A sensor pixel 402 produces photocurrent in response to light incident thereon. The sensor pixel 402 output is electrically coupled to a first input to an amplifier 404, a first side of a high gain capacitor 406 having capacitance CHG, a first side of a low gain switch 410, and a first side of a reset switch 412. The amplifier 404 has a constant voltage VREF as its second input, and a variable voltage VOUT as its output. The amplifier output is electrically coupled to the second side of the high gain capacitor 406, to a first side of a low gain capacitor 408 having capacitance CLG where CLG can be greater than CHG, and to a second side of the reset switch 412. The second side of the low gain switch 410 is electrically coupled to the second side of the low gain capacitor 408. The ROIC periodically opens the reset switch 412 to start each video frame, and closes the reset switch 412 briefly to end each video frame.
The configuration of
In most cases, the ROIC switches between high gain and low gain for all pixels, together, and does so on a video frame-by-frame basis. For a particular frame, the ROIC sets all the pixels to high gain, or all the pixels to low gain. The ROIC typically does not switch gains during a frame, and typically only switches gain between frames.
The sensor pixel 502, amplifier 504, high gain capacitor 506, and reset switch 512 are similar in structure and function to similarly numbered elements 4xx in
Low gain transistor 510 functions as an open circuit for output voltages VOUT greater than a threshold voltage below VLG. Low gain transistor 510 functions as a conductor for output voltages VOUT less than the threshold voltage below VLG. During the initial portion of a frame, the output voltage is relatively high, the low gain transistor 510 remains open, the low gain capacitor 508 is removed from the circuit, and the charge integrates on the high gain capacitor 506. If the output voltages VOUT decreases to the threshold voltage below VLG, the low gain transistor 510 inserts the low gain capacitor 508 into the circuit, and for the remainder of the frame, any further charge integrates on both the high gain capacitor 506 and the low gain capacitor 508.
Potential advantages to the automatic switching in of the low gain transistor 510 include allowing for per pixel dual gain, and keeping dual gain always active (as opposed to selecting either a high gain or a low gain at the beginning of a frame).
The circuits shown in
At 1102, method 1100 resets all the integrating capacitors in the CTIA input cell. Examples of such capacitors can include high gain capacitor 506 (
At 1202, method 1200 produces photocurrent from a sensor pixel having light incident thereon. At 1204, method 1200 resets a high gain capacitor and a low gain capacitor to respective specified reset voltages at a beginning of a video frame. In some examples, the specified reset voltages are the same; in other examples, they can differ. At 1206, method 1200 integrates charge arising from the photocurrent on the high gain capacitor. The method 1200 senses a voltage across the high gain capacitor. If the sensed voltage has dropped to a specified threshold voltage, then at 1208 method 1200 automatically activates the low gain capacitor to be electrically coupled in parallel with the high gain capacitor. Method 1200 switches in the low gain capacitor. Switching in the low gain capacitor allows the voltage on the sum of the capacitors to be read. Until the low gain capacitor is switched in, the voltage at the output of the CTIA is just that due to the high gain capacitor. At 1210, method 1200 integrates the charge arising from the photocurrent on both the low gain capacitor and the high gain capacitor. Method 1200 samples a second voltage across both the low gain capacitor and the high gain capacitor. At 1212, method 1200 returns the first and second voltages at an end of the video frame. The first and second voltages correspond to a light intensity incident on the sensor pixel integrated over the video frame.
The method 1200 of
In an alternate configuration, the CTIA input cell can include three capacitors, rather than two. When a first of the three capacitors reaches saturation, a first transistor switches in a second capacitor in parallel to the first capacitor. When the second of the three capacitors reaches saturation, a second transistor switches in a third capacitor in parallel to the first and second capacitors. Such a configuration can automatically switch among three gain levels, with the gain level for each pixel being automatically switched independent of the other pixels.
In further alternate configurations, the CTIA input cell can include four, five, six, or more than six capacitors. These alternate configurations can also include three, four, five, or more than five transistors to switch in the respective capacitors as needed.
It will be understood that the absolute sign of the voltages presented herein is arbitrary. For instance, the plus and minus inputs on the amplifiers 204, 404, 504 may be switched, thereby switching positive voltages to negative voltages, and vice versa. If the absolute signs of the voltages are switched from those discussed above, all references of “greater than” and “less than”, “above” and “below, and the like, should be switched as well.
The Abstract is provided to comply with 37 C.F.R. Section 1.72(b) requiring an abstract that will allow the reader to ascertain the nature and gist of the technical disclosure. It is submitted with the understanding that it will not be used to limit or interpret the scope or meaning of the claims. The following claims are hereby incorporated into the detailed description, with each claim standing on its own as a separate embodiment.
