World's Smallest 2.0µm-Pixel MOS Image Sensor

February 9, 2005

Panasonic, the leading brand by which Matsushita Electric Industrial Co., Ltd. is known, today announced the development of the world's smallest image sensor. The revolutionary MOS image sensor has only 2.0 × 2.0 μm pixels which is now officially the world's smallest sensor, where the previous standard was 3.1 μm. Matsushita's technological breakthrough allows two million pixels on a 1/4-inch MOS sensor instead of one million. This size reduction was also achieved without sacrificing sensitivity by minimizing the number of transistors required for each pixel. Details of the sensor will be presented at the International Solid-State Circuits Conference (ISSCC) held in San Francisco during February 6 - 10, 2005.

One problem industry has experienced with reducing pixel size in the past has been a corresponding reduction in sensitivity. Panasonic has solved the problem by developing three new technologies: the 0.15μm design rule, sharing the detection amplifier circuit, and pulse driving the electric source line.

In conventional CMOS sensors, the wiring on a chip is usually 0.25μm wide. This limits the possible miniaturization of the photodiode area. In the new sensor, the 0.15μm design rule cuts 40 per cent from the amount of space taken up by wiring, allowing more space for photodiodes.

Conventional CMOS image sensors require four transistors for each pixel: a detection amplifier circuit with three transistors, and a charge read-out transistor. In the new sensor, a detection amplifier is shared by four pixels, one each from two rows and two columns. This means that four pixels now require only one detection amplifier circuit and four charge read-out transistors (one for each pixel).

Furthermore, each detection amplifier circuit has been redesigned to need only two transistors instead of three. The detection amplifier circuit in conventional CMOS image sensors usually requires an output transistor, a resetting transistor, and a row-selecting transistor. But in the new sensor, by replacing the conventional direct electric source for pulsing, the column-selecting transistor can be eliminated. Thus the 16 transistors required by four pixels in a conventional design are reduced to just six in the new chip.

Explore further: What neuroscience can learn from computer science

Related Stories

What neuroscience can learn from computer science

August 10, 2015

What do computers and brains have in common? Computers are made to solve the same problems that brains solve. Computers, however, rely on a drastically different hardware, which makes them good at different kinds of problem ...

New conductive ink for electronic apparel

June 25, 2015

University of Tokyo researchers have developed a new ink that can be printed on textiles in a single step to form highly conductive and stretchable connections. This new functional ink will enable electronic apparel such ...

Future biosensors could be woven into clothes

June 23, 2015

Commonly used health tests, such as pregnancy and blood sugar tests, involve putting a drop of fluid on a test strip, which is infused with a substance designed to detect a specific molecule.

Amplifying small motions in large motions

June 17, 2015

For several years now, the research groups of MIT professors of computer science and engineering William Freeman and Frédo Durand have been investigating techniques for amplifying movements captured by video but indiscernible ...

Recommended for you

Interactive tool lifts veil on the cost of nuclear energy

August 24, 2015

Despite the ever-changing landscape of energy economics, subject to the influence of new technologies and geopolitics, a new tool promises to root discussions about the cost of nuclear energy in hard evidence rather than ...

0 comments

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.