Background of the Invention CMOS image sensors are integrated circuits designed specifically to capture and process incident light. The core of the sensor is generally made up of an array of pixels or picture elements. The pixels comprise photodiodes to sense the light, and CMOS transistors that take care of the amplification and transfer of the signal sensed by the photodiode.

As light strikes the array of pixels, it is converted into photo-current, which in turn gets converted into digitized data by an Analog-to-Digital Converter (ADC). The pixel arrays are based on either active or passive pixel sensors. Passive pixel sensors do not amplify the photodiode's signal within them, whereas, active pixel sensors include amplification circuitry within the pixel.

Figure 1 illustrates a standard prior art three-transistor active pixel cell 100. It includes a photodiode 104 and three transistors 101, 102 and 103. The photodiode 104 converts the light impinged upon it into an electrical signal. NMOS transistor 101 has its source tied to the cathode of the photodiode 104, its drain tied to a voltage potential Vres and its gate tied to a reset control signal, RESET. When the RESET signal is in a high state, transistor 101 becomes active and pre-charges the photodiode's 104 capacitance (not shown) to the voltage Vres. A second NMOS transistor 102 is used to separate the photodiode 104 from the external circuitry at the output, i. e. image processing hardware. The NMOS transistor102 drain is tied to the voltage supply rail, VDD, its gate is tied to the cathode of the photodiode 104 and its source tied to one drain/source terminal of transistor 103. NMOS transistor 102 is configured as a source follower to separate the potential on the photodiode 104 from a read-out circuit (not shown) connected to the output, i. e. it is acting as a buffer.

Transistor 103 is a third NMOS transistor that acts as a read access transistor. One of

its source/drain terminals is tied to the drain of transistor 102 its other drain/source terminal tied to the pixel cell output and its gate tied to a control signal, SELECT.

The diode 104 is built by doping an n+ region onto a p-type substrate. For the diode 104 to work as a photodiode, it is reverse biased. A photodiode104 capacitance (not shown in the diagram) is formed due to the depletion region of the p-n junction.

During the reset time of the pixel 100, the photodiode capacitance is pre-charged to a potential close to Vres. Once reset is complete the integration time begins, and incoming photons create a photocurrent in the depletion region that discharges the photodiode's capacitance. This discharge establishes a voltage drop in the photodiode's capacitance, which is detected after the completion of the integration time. The higher the voltage drop in the photodiode capacitance during the integration time, the higher the picture resolution is possible.

For large pixel arrays, the issue of time lag between the read-out of the first row of pixels 100 and the subsequent row becomes a problem. This becomes an even bigger issue when trying to sense moving objects, since different images are caught by consecutive rows of the sensor array, creating linear distortion of the image.

One solution is to activate all the pixels 100 at the same time and after the integration time is complete, to freeze the state of all the pixels 100 until they have all been processed. This requires the same integration time for all active pixel cells 10Q Without the use of a mechanical shutter, this is not a possibility with the classic three transistor APC 100.

Therefore there is a need for an APC capable of maintaining its integrated charge for the time it takes for the array of pixels to be processed.

Summary of the Invention The invention is directed to an active pixel cell comprising a photodiode, an element for resetting the voltage on the photodiode and a capacitance for receiving the charge from across the photodiode. The pixel further includes a switch coupled between the photodiode and the capacitance having a closed position for connecting

the photodiode to the capacitance and an open position for electrically isolating the capacitance from the photodiode. With the switch in the open position the charge on the capacitance is maintained substantially constant. An amplifier provides an output signal representing the charge on the capacitance.

In accordance with an aspect of this invention, the amplifier comprises an amplifying transistor and the capacitance comprises the input capacitance of the amplifying transistor. Further, the isolating switch and the charge resetting element both comprise transistors.

In accordance with a further aspect of this invention, the transistors are CMOS transistors that may be integrated on a common substrate. In the preferred embodiment, the amplifying transistor and the resetting transistor are NMOS transistors and the isolating transistor is a PMOS transistor.

In accordance with another aspect, the invention is directed to an active CMOS pixel cell comprising a photodiode, first transistor for resetting the photodiode and a second transistor that has an input capacitance for receiving a charge from across the photodiode and that is connected to a common node with the first transistor. A third transistor, which has an open position and a closed position, is coupled between the photodiode and the common node, whereby the charge across the photodiode is applied to the input capacitance of the second transistor when the third transistor is in the closed position and the photodiode is electrically isolated from the input capacitance when the third transistor is in the open position.

In accordance with an aspect of the invention, the charge on the input capacitance is maintained substantially constant when the third transistor is in the open position. This is achieved by having a photo-leakage current from the first transistor flowing in one direction to the common node and a photo-leakage current from the third transistor flowing in an opposite direction to the common node, therby cancelling one another out. The photo-leakage currents are preferably substantially equal.

In accordance with a specific aspect of the invention, a fourth transistor is connected to the second transistor for selectively connecting the pixel to an output bus. In the preferred embodiemtn, the first, second and fourth transistors are NMOS transistors and the third transistor is a PMOS transistor.

The invention is further directed to a CMOS image sensor comprising an array of rows and columns of active pixel cells of the type defined above that are integrated on a chip.

Other aspects and advantages of the invention, as well as the structure and operation of various embodiments of the invention, will become apparent to those ordinarily skilled in the art upon review of the following description of the invention in conjunction with the accompanying drawings.

Brief Description of the Drawings The invention will be described with reference to the accompanying drawings, wherein: Figure 1 illustrates a standard 3-transistor active pixel cell; and Figure 2 illustrates a 4-transistor active pixel cell in accordance with the present invention.

Detailed Description of the Invention The present invention provides a charge storage capability for the standard three-transistor pixel cell 100 illustrated in figure 1. This is achieved with the aid of an additional transistor that is complementary to the reset transistor. In this case, a PMOS transistor is used to complement the NMOS reset transistor. This additional transistor acts both as a typical switch to connect the photodiode capacitance with the gate capacitance of the source-follower transistor, and a device that reduces photo- current discharging of the source-follower transistor's gate capacitance. The switch is kept on during the integration time to ensure equal potentials on source-follower gate and photodiode capacitances. The pixel cell is capable of storing charge for some storage time, which is limited because of analog nature of the storage mechanism, but long enough to read out the entire pixel array. This storage time just depends on the

physical dimensions of the gate terminal of the source follower transistor. This pixel cell is also capable of maintaining the stored charge in the presence of light thus making the need for external means of cutting out the light, such as a mechanical shutter, unnecessary.

Figure 2 represents an embodiment of the present invention. The active pixel cell 200 includes a photodiode 204 and three transistors 201, 202 and 203 found in the standard three transistor APC. Photodiode 204 capacitance 205 and the gate capacitance 206 of transistor 202 are not discrete components but are the capacitances due to the depletion region of the p-n junction of photodiode 204 and the input capacitance of transistor 202 respectively. In accordance with the present invention, a transistor 207 complementary to the reset transistor 201 i. e. PMOS transistor, is introduced as the fourth transistor to modify the standard three-transistor active pixel cell 100 of figure 1. In addition, the reset transistor201, the complementary transistor 207 and gate capacitance 206 all share a common node.

A transistor complementary to the reset transistor203 is used as the switch 207, instead of a matching transistor in order to obtain a circuit where the photo- currents that are generated in the transistor drain/source-body junctions and which flow into the common node, compensate each other.

If an NMOS transistor is used as the switch 207, its first p-n junction is connected in parallel with the main photodiode 204 junction and the second p-n junction is connected in parallel with the source-follower transistor 202 gate capacitance 206. This second p-n junction's photo-leakage current discharges the source-follower's gate capacitance 206 when the switch 207 is off, therefore, defeating the purpose of introducing a charge storage element.

The PMOS transistor 207 is serially connected between the cathode of the photodiode 204 and the gate of the source-follower transistor 202. The gate of transistor 207 is connected to control signal STORE. The NMOS reset transistor201 is connected at the common node on the line between the PMOS transistor207 and the source-follower transistor 202. The drain of transistor 202 is connected to the

power supply VDD and its source is connected to a drain/source terminal of selection transistor 203.

The functionality of the new 4T-pixel cell is as follows: During reset, both the PMOS transistor 207 and the NMOS reset transistor 201 are switched on, so a path from the reset voltage, Vres to the cathode of the photodiode 204 is established.

The photodiode capacitance 205 and the source-follower's gate capacitance 206 are pre-charged to the same potential. These capacitances are not physically part of the architecture, but rather formed due to the source follower transistor's 202 input capacitance and a capacitance of the depletion region in the p-n junction of the photodiode 204. The source follower transistor 202 is constructed in such a manner that its gate region is insensitive to the light.

Next, the reset transistor 201 switches off and the integration time begins. The PMOS transistor 207 is still in activation and consequently, the voltage on both the photodiode and source-follower gate capacitances 205, 206 are very close during the entire integration time. After the integration time is complete, the PMOS transistor 207 switches off. As a result, the link between both capacitances disconnects, yet the voltage on both capacitances remains the same. At this point, photo-leakage currents are responsible for the changes in voltage on the gate capacitance 206. The photo-leakage current from the PMOS transistor's 207 p+-body junction is charging the gate capacitance 206 and the current of the n+-body junction of the NMOS reset transistor 201 flows in the opposite direction and discharges the gate capacitance 206. However, when the dimensions for both p+ and n+ are selected adequately, in terms of their doping profiles and transistor sizes, both photo-leakage currents compensate each other, and they do not affect the potential on the gate capacitance 206. Therefore, the gate capacitance 206 of the source-follower transistor 202 stores, for a particular time, the proper value of the pixel 200 voltage.

-The storage time can be obtained by selecting the appropriate dimensions of the gate of the source-follower transistor 202. Once the voltage stored on the gate capacitance 206 exceeds that of the threshold voltage of source-follower 202, this transistor will turn on and generate a current proportional to the gate capacitance 206. When the pixel cell200 selection signal, SELECT is in a high state, the selection transistor203 becomes active and passes the information originally stored on both the photodiode capacitance 205 and the source follower's 202 gate capacitance 206 to the output, which is the main data bus.

While the invention has been described according to what is presently considered to be the most practical and preferred embodiments, it must be understood that the invention is not limited to the disclosed embodiments. Those ordinarily skilled in the art will understand that various modifications and equivalent structures and functions may be made without departing from the spirit and scope of the invention as defined in the claims. Therefore, the invention as defined in the claims must be accorded the broadest possible interpretation so as to encompass all such modifications and equivalent structures and functions.