Amorphous silicon active pixel sensor with rectangular...

Details Amorphous silicon active pixel sensor with rectangular... Amorphous silicon active pixel sensor with rectangular...

1999-02-02

2001-06-26

Lee, John R. (Department: 2878)

Radiant energy

Photocells; circuits and apparatus

Photocell controlled circuit

C257S292000, C348S294000

Reexamination Certificate

active

06252218

ABSTRACT:

BACKGROUND OF THE INVENTION
The present invention relates generally to an active image sensor and particularly to an active image sensor with an amorphous silicon sensing layer and a semiconductor readout layer.
Many modern imaging systems such as digital cameras, camcorders, scanners, copiers, and the like require an image sensor capable of capturing an image with high resolution and color fidelity. A majority of the imaging systems use a solid-state image sensor due to constraints on system size, weight, power consumption, heat dissipation, and durability. The ability of the imaging system to capture images with high resolution and color fidelity depends to a large extent on the image sensor used. Image sensor resolution is typically measured in number of pixels contained in the image sensor. The resolution of a solid-state sensor can be increased by shrinking the size of the pixels so that more pixels fit within a predetermined area. An image incident on an image sensor having more pixels per unit of area can be captured with greater resolution than an image sensor having fewer pixels per unit of area. Pixel geometry and symmetry between pixels can also affect the number of pixels per unit of area. Additionally, the symmetry between pixels can affect color aliasing. Other factors, such as electrical noise generated by the pixels can affect image quality. Additionally, for color images, the color fidelity of the image can be affected by a predetermined color pattern assigned to the pixels in the image sensor.
Desirable attributes of an image sensor include low electrical noise, flexibility in color pattern, optimized pixel density and geometry for increased resolution, symmetry between adjacent pixels, and maximized pixel fill-factor. Low electrical noise prevents an image signal generated by the pixel from being corrupted by noise; furthermore, noise and leakage current from an individual pixel can affect the image signal from adjacent pixels. Additionally, pixel electrical noise can be affected by pixel geometry and symmetry between pixels. Circuitry or algorithms to filter out the electrical noise can be minimized or eliminated by reducing the noise generated by each pixel. A flexible color pattern allows the image sensor to be tailored to specific applications such as a RGB color pattern for photography or video, or a CMY color pattern for printing or scanning, for example.
Pixels having a rectangular geometry can generate additional noise due to abrupt rectangle edges which create leakage current in an active device such as a photodiode, for example. A pixel geometry that eliminates noise due to abrupt edges is desirable. Further, it is desirable to maximize the active area of the pixel devoted to capturing the image so that the ratio of the active area to the pixel area approaches a fill-factor of 100%. By removing components, such as transistors and signal routing lines, from the pixel, the area that was occupied by the components can be used to maximize the active area of the pixel thereby increasing the fill-factor.
Previous image sensor designs include photo-gate active pixel sensors, bulk silicon photodiode pixel sensors with three transistors, pulsed bipolar CMOS active pixel sensors, and rectangular CMOS pixel sensors.
The photo-gate active pixel sensor utilizes four transistors inside of a CMOS photo-gate pixel for low noise operation and for electronic shuttering. However, this number of transistors results in a pixel area that is not competitive with state-of-the-art CCD image sensors that have pixel areas approaching 5×5 &mgr;m
2
. The number of transistors in the photo-gate sensor results in a lower fill-factor when compared to the state-of-the-art CCD image sensor.
The pulsed bipolar CMOS active pixel sensor employs a vertical bipolar device in a CMOS process. The use of bipolar devices results in a smaller pixel area due to fewer transistors in the pixel; however, disadvantages of this design include image lag and the need to amplify an output signal from the pixel with a high &bgr; gain bipolar transistor. The high &bgr; gain bipolar transistor exhibits signal degradation under low current conditions and limits scalability of the pixel due to an integrating amplification stage in a readout path of the pixel. Further, an extra emitter terminal is required to prevent an overflow condition. The extra emitter terminal creates an unsymmetrical pixel layout in the CMOS process. In the pulsed bipolar CMOS active pixel sensor, pixel area is not minimized because CMOS design rules require adjacent pixels to be spaced apart to comply with design rules for N-well spacing.
The bulk silicon photodiode pixel sensor is not competitive with the state-of-the-art CCD image sensor because the photodiode and the transistors are integral to the same silicon layer and the area occupied by the transistors reduces the active area available for the photodiode resulting in a fill-factor of about 30% using 1.2 &mgr;m CMOS process technology.
Rectangular CMOS active pixel sensor designs utilize a rectangular pixel geometry with the pixels laid out in a rectangular grid. It is clearly understood that the rectangular grid does not result in an optimized pixel density. Bayer's color pattern was developed as the color filter array (CFA) pattern for the rectangular grid. In the Bayer CFA pattern, color density for Red, Green, and Blue sensors is in the ratio of 1:2:1. Therefore, there are two Green pixels for each Red pixel and Blue pixel in the array. In general, the color density for the Bayer CFA pattern is not the best choice because it undersamples two of the sensor colors (Red and Blue) and the color pattern kernel is not symmetric. In the Bayer CFA pattern, linear horizontal, linear vertical, and linear diagonal image features never pass through more than two different color sensor locations because the Red and the Blue sensor locations are never adjacent to sensors of the same color. The rectangular pixel geometry is not amenable to flexible color patterning.
In
FIG. 1
, there is provided for purposes of illustration a prior art rectangular CMOS active pixel sensor designated generally as
100
. The sensor
100
has a plurality of subpixels
101
. Four of the subpixels
101
define a pixel
103
, shown in heavy outline. The pixel
103
has two Green subpixels
105
, one Red subpixel
107
, and one Blue subpixel
109
. This arrangement of the color pattern for the subpixels
101
in the pixel
103
corresponds to the Bayer CFA pattern wherein the color density of the Red subpixel
107
, the Green subpixels
105
, and the Blue subpixel
109
is in the ratio of 1:2:1, wherein the pixel
103
has two Green subpixels
105
for one Red subpixel
107
and one Blue subpixel
109
.
FIG. 2
illustrates lineal image feature traversals across prior art sensor
100
. In a first horizontal traverse as shown by an arrow
133
, the image feature traverses the Green subpixels
105
and the Red subpixels
107
. In a second horizontal traverse as shown by an arrow
122
, the image feature traverses the Blue subpixels
109
and the Green subpixels
105
. The first horizontal traverse
133
never crosses the Blue subpixels
109
. Similarly, the second horizontal traverse
122
never crosses the Red subpixels
107
. Because the Bayer CFA color pattern kernel is not symmetric the first horizontal traverse
133
and the second horizontal traverse
122
never make a sequential crossing of the Red subpixels
107
, the Green subpixels
105
, and the Blue subpixels
109
in a single traverse of the sensor
100
. The same is true for a first vertical traverse as shown by an arrow
135
where only the Green subpixels
105
and the Red subpixels
107
are traversed. In a second vertical traverse as shown by an arrow
124
, only the Green subpixels
105
and the Blue subpixels
109
are traversed. Additionally, in a first diagonal traverse as shown by an arrow
137
, only all the Green subpixels
105
are traversed. In a second diagonal traverse as shown by an arrow
126
only the Red subpixels
1

Affiliated with

Also associated with

Rating

Amorphous silicon active pixel sensor with rectangular... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Amorphous silicon active pixel sensor with rectangular..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Amorphous silicon active pixel sensor with rectangular... will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2458046