Thursday, December 30, 2010

Samsung Proposes SEFET Sensor, Pixart Flips Color Filter and Microlens

Samsung patent application US20100320515 proposes a single-electron JFET-like structure (SEFET) to make small and sensitive pixel:

The photoelectrons are collected in N-doped area and drift close to the p-channel area due to the doping gradient. Then the collected electrons modulate the channel conduction by changing the N-gate potential:


Few different readout schemes are proposed, for example, this one (Vin is N-doped area voltage):

The SEFET pixel is said to be able to achieve a conversion gain of 1-9mV/e, as measured at the N-doped gate. Less than 1um pixel size is claimed to be possible.

One thing I was unable to find in the Samsung application is how the N-doped gate is reset before the integration starts.

Pixart application US20100320552 proposes to place color filter on top of microlens to improve the light passage through the metal stake:

"Compared with the prior art, such arrangement is advantageous in that the micro lens 18 is located much closer to the photo diode 12. As a result, the micro lens 18 has a wider chief ray angle, and the photo diode 12 has a better light collection efficiency."

Tuesday, December 28, 2010

Sony Aims to Capture 30% of Mobile Sensor Market

Forbes quotes Sony saying that it intends to win 30% of the market for image sensors used in mobile phones, up from an expected 10% for the fiscal year ended March 2011.

Monday, December 27, 2010

Sony to Invest $1.2B to Double Image Sensor Production

Reuters: Sony will invest $1.2B in the next financial year to double its output of image sensors. The sum includes a deal announced last week to buy back a semiconductor production line from Toshiba, which has been estimated by an industry source at 50 billion yen ($600 million).

Sony will take advantage of a Japanese government subsidy for environmentally friendly businesses to help with the investment, it said in a statement but declined to say how much that would be.

It will convert part of the plant in Nagasaki, southern Japan, for the production of CMOS sensors and invest in wafer processing equipment for CMOS image sensors. The investment will bring its total production of image sensors, including CCD and CMOS types, to 50,000 300mm wafers a month by March 2012.

Market Watch and Business Week too present their versions of the story.

Update: Sony published an official PR here.

"Sony Corporation ("Sony") today announced that Sony plans to invest approximately 100 billion yen in Sony Semiconductor Kyushu Corporation's Nagasaki Technology Center ("Nagasaki TEC") in the fiscal year ending March 31, 2012, to increase the production capacity for CMOS image sensors.

This investment plan includes (i) the transfer of the semiconductor fabrication facilities from Toshiba Corporation ("Toshiba") contemplated under a non-binding memorandum of understanding between Sony and Toshiba jointly announced on December 24, 2010, (ii) refurbishment of a part of the above semiconductor fabrication facilities into new wafer lines capable of manufacturing CMOS image sensors, and (iii) refurbishment and equipment of a part of production facilities at Nagasaki TEC Building 3 for wafer processing to differentiate Sony's CMOS image sensors with Sony's independently developed unique technologies. Through the investment plan, Sony will utilize a governmental subsidy to be provided by the Ministry of Economy, Trade and Industry in Japan - the "subsidy for programs to promote siting low-carbon job-creating industries" - mainly in connection with the investment mentioned in (iii) above.
"

Sunday, December 26, 2010

Shoji Kawahito Received Takayanagi Memorial Award

Shizuoka University Professor Dr. Shoji Kawahito has received the 24th Takayanagi Memorial Award for his contribution to development of high-performance and high-functionality imaging devices. (Kenjiro Takayanagi is the father of Japanese television.)

Thursday, December 23, 2010

Chipworks Analyzes Samsung 1.4um BSI Sensor

Chipworks got a hold of Samsung S5K4E5 - a 1.4um pixel-based 5MP 1/4-inch mobile sensor. This is said to be Samsung's first BSI sensor with TSV technology. Samsung has indicated that it uses a lower-cost bulk silicon manufacturing process which they estimated to be 4x to 5x lower wafer cost compared with SOI.

The reverse engineering is ongoing. Meanwhile Chipworks released a photo of the sensor's backside:

Wednesday, December 22, 2010

Optrima to Mass Produce Consumer ToF-Based 3D/RGB/Audio Camera

Business Wire: Optrima announced it has reached mass-production quality for a consumer electronics 3D, video (HD) and audio camera. The camera, the result of collaboration with its partner Namuga, is said to be the world’s first embedded 3D camera with high definition RGB.

"The availability of a high-performance, low-cost 3D camera with built-in audio and video will dramatically contribute to the adoption of 3D gesture-based interfaces and applications by consumers," said Andre Miodezky, CEO of Optrima. "We’re very excited to be the first company to bring such a product to the market and look forward to working with our customers to bring the product to the market in 2011."

Consumers will be able to utilize the camera for gesture-based applications such as video games, interactive, touch- and controller-less user interfaces, as well as video conferencing and other exciting new user experiences, whether on desktop computer systems or connected TVs.

Sony to Buy Back Image Sensor Fab from Toshiba

Reuters: Sony plans to buy back a Nagasaki semiconductor plant from Toshiba Corp for about 50 billion yen ($597.2 million) to double output capacity for CMOS image sensors used in smartphones and other devices, the Nikkei business daily reported.

Sony sold it to Toshiba in 2008. Acquiring the plant back will double Sony's image sensor output capacity to the equivalent of about 40,000 wafers a month, the daily said.

By boosting output and lowering production costs, Sony aims to catch up with such firms Samsung and U.S. players in the CMOS sensor market, Nikkei said.

Update: EETimes publishes an article with details about the March 2008 complex deal between Toshiba and Sony.

Update #2: Reuters quotes Sony Executive Deputy President Hiroshi Yoshioka saying in an interview this week that it was in the process of deciding how to cover a shortage in CMOS sensor output capacity.

Update #3: Sony published an official PR here.

Nagasaki Semiconductor Manufacturing Corporation (NSM), which was established in March 2008 and is located in the Nagasaki Technology Center of Sony Semiconductor Kyushu Corporation ("SCK"), has been manufacturing the high-performance "Cell Broadband Engine" processor, the graphics engine "RSX" and other high-performance semiconductors and leading-edge SoCs for applications in digital consumer products of Toshiba and Sony. The facilities to be transferred would be the fabrication facilities and equipment for the 300 mm wafer line located within the Nagasaki Technology Center purchased by Toshiba from Sony and SCK and leased to NSM in 2008 and other facilities that Toshiba and Sony will agree to transfer among those in which Toshiba invested in connection with the operation by NSM after the purchase.

Current NSM ownership: Toshiba 60%, Sony 20% and SCEI 20%
Number of employees: Approximately 460 as of December 1, 2010

Teledyne to Acquire DALSA

Business Wire: Teledyne and DALSA jointly announce that they have entered into a definitive agreement that provides for the acquisition of DALSA by a wholly-owned subsidiary of Teledyne. The aggregate value for the transaction is approximately CAD $341 million, taking into account DALSA’s stock options and net cash as of September 30, 2010. That represents a premium of 27.7% over the twenty-day volume weighted average trading price.

Headquartered in Waterloo, Ontario, DALSA has approximately 1,000 employees worldwide. For the twelve months ended September 30, 2010, DALSA had sales of approximately CAD $201 million.

"I view this transaction as a natural evolution for DALSA and a positive development for all DALSA stakeholders including shareholders, employees, executives, local communities, customers and vendors," commented Dr. Savvas Chamberlain, Chairman of the Board and Founder of DALSA. "The decision to be a part of a larger organization recognizes that in order for DALSA to become a billion dollar company, we need to team up with an industry leader with complementary technologies. Finally, as the founder of the company, I am pleased to see DALSA’s name live on, in its new incarnation as Teledyne DALSA."

"Teledyne and DALSA are each acknowledged leaders in digital imaging technology but our product lines and customer bases are almost entirely complementary. For example, DALSA produces among the world’s most advanced visible light imaging sensors and cameras for commercial applications, while Teledyne produces extreme resolution infrared sensors and subsystems primarily for government applications," said Dr. Robert Mehrabian, Chairman, President and CEO of Teledyne. "The combined strengths of Teledyne’s and DALSA’s leading imaging technologies will allow us to develop new infrared and visible light products that serve our respective markets and customers."

The definitive agreement contains a termination fee in the amount of approximately CAD $10.2 million, which is payable by DALSA to Teledyne in certain circumstances if the transaction is not completed. DALSA and Teledyne expect that the transaction will close in February 2011.

Update: EETimes too published an article with some more background about DALSA and Teledyne.

Ambarella iOne Supports 32MP Imaging on Android Platform

Business Wire: Ambarella announced the iOne SoC HD camera for Android OS platform. The iOne is based on a triple ARM CPU architecture. Dual-core 1-GHz ARM Cortex A9 CPUs provide the horsepower to run Android applications while a third 533-MHz ARM-11 handles real-time camera tasks and enables camera boot time of under one second.

The camera part supports simultaneous still photography and HD video recording - up to 32MP still image processing with high speed capture of 5MP images at 30 fps, and Electronic Image Stabilization (EIS) with CMOS sensor rolling shutter correction.

The iOne SoC is sampling now to early access customers.


Venture Beat adds that cameras with the iOne chips will likely appear in the second half of 2011. Ambarella has 350 employees. Current Ambarella chips are used in devices such as the Sony Bloggie video camera and a host of others.

Tuesday, December 21, 2010

Varioptic Aims to Raise EUR 10 million in 2011

ReportLinker: Varioptic aims to raise EUR 10 million (USD 13.1M) during the first half of 2011 to finance a production capacity increase, French daily Les Echos wrote, quoting CEO Hamid Farzaneh.

Microsoft to Improve Kinect Resolution

Eurogamer: Microsoft is working to improve the accuracy of Xbox Kinect so that it could detect finger movement and hand rotation. Kinect depth sensor is currently running at 320x240 resolution and 30fps speed, while its hardware could support 640x480 resolution. The issue relates to the USB controller interface, Eurogamer said. It is capable of around 35MB/s, but it only uses around 15/16MB/s. This artificial limit is in place because multiple USB devices can be used at once on Xbox 360.

Now Microsoft is working on a technology to allow greater throughput in this regard. At a 640x480 depth sensor resolution Kinect could begin to detect fingers and hand rotation – an effective quadrupling of its accuracy.

EU Key Enabling Technologies in Imaging

EU Expert Group on Key Enabling Technologies has had a talk on imaging technologies in Europe. Leti-Minatec presented a proposal on the important areas to develop:

Monday, December 20, 2010

Cypress Designs Image Sensors for ARRI

Business Wire: Cypress announced that its CMOS image sensor has been designed into 35 mm ALEXA digital motion picture camera from ARRI (Arnold & Richter Cine Technik GmbH), headquartered in Germany. A Cypress custom CMOS imager sensor also was designed into ARRI’s ARRIFLEX D20/D21 digital motion picture camera, which preceded the ALEXA.

TSMC 0.9um Pixel/Process Presentation

Somebody put all IEDM 2010 papers into open access. Image sensor session papers start with S14P. One of the most interesting papers is TSMC invited presentation S14P01.PDF:

A Leading-Edge 0.9μm Pixel CMOS Image Sensor Technology with Backside Illumination: Future Challenges for Pixel Scaling (Invited)
S.G. Wuu, C.C. Wang, B.C. Hseih, Y.L. Tu, C.H. Tseng, T.H. Hsu, R.S. Hsiao, S. Takahashi, R.J. Lin, C.S. Tsai, Y.P. Chao, K.Y. Chou, P.S. Chou, H.Y. Tu, F. L. Hsueh, L. Tran,
Taiwan Semiconductor Manufacturing Company, Hsin-Chu, Taiwan, R.O.C.

Some interesting quotes from the paper talking about the new N65 300mm BSI process performance:

TSMC uses "Backside antireflection (BARC) layers to further enhance optical sensitivity."

"wafer-to-wafer ... average thickness variation (are) within 3% of the mean value. ...within-wafer TTV (total thickness variation) for 300mm wafers ... improved to be less than 15%, which is comparable to existing 200mm BSI process as shown in Fig. 3."


One of the problems TSMC observed in small pixels is an increased PRNU level. TSMC team has implemented 4 techniques to improve it:

  1. tightened design rule with advanced lithography (N45/N28…)
  2. better pixel-to-pixel isolation
  3. antireflection film optimization
  4. novel structure for light guidance

The PRNU-treated pixel shows 7X PRNU reduction for 1.1um pixel and a 4X reduction on the 0.9um pixel:


Crosstalk is another major challenge in small pixel design. TSMC team has identified CFA spectral crosstalk being the dominant component, followed by optical diffraction, and by electrical crosstalk. TSMC calls for novel ideas in color patterning other than traditional Bayer pattern and new system-level color processing algorithm to reduce the CFA spectral crosstalk problem:


Being with Advasense with our high full well pixel claim, I immediately noticed the full well vs pixel size graph, with 0.9um pixel having just ~30% of 1.4um pixel full well:


And finally, there are color sample pictures from the new 1.1um and 0.9um pixels. 0.9um colors looks a bit pale, but this is somewhat expected at this early design stage:


Congratulations to S.G. Wuu and the whole TSMC team! I'm very impressed by your cutting-edge work!

Sunday, December 19, 2010

Chipworks: Kinect is Powered by Two Aptina Sensors

Chipworks reverse engineering report reveals that the primary depth sensing IR sensor in Microsoft Kinect is Aptina's MT9M001C12STM. This is 1.3MP sensor based on 5.2um pixels.

Another Chipworks report says that the secondary RGB camera in Kinect is powered by Aptina MT9M112214STM sensor. I believe Chipworks meant 1/4-inch 1.3MP MT9M112 SoC.

Range-Imaging Sensors and Applications 2011 Workshop

The TOF Range-Imaging Sensors and Applications (RISA) international workshop on Jan 27-28, 2011 in Trento, Italy will bring together for the first time experts of Range-Imaging (RIM) cameras from the metrology as well as sensor side and let the two communities communicate much more on the RIM topic in order to exchange ideas, activate collaborations and provide possible improvements both in the sensor and application sides. The workshop agenda has a lot of interesting stuff:

TOF Range-Imaging sensors, characteristics and performances
Dario Piatti - Dept. DITAG, Politecnico of Torino, Italy

Integrated calibration of range cameras and their application for deflection measurement
Derek Lichti - Department of Geomatics Engineering, The University of Calgary, Canada

SR-4000 and CamCube3.0 ToF cameras: tests and comparison
Fulvio Rinaudo – Dept. DITAG, Politecnico of Torino, Italy

ToF cameras for architectural survey
Filiberto Chiabrando – Dept. DITAG, Politecnico of Torino, Italy

Indoor Navigation using Range Imaging
Tobias Kohoutek - Institute of Geodesy and Photogrammetry, ETH Zurich, Switzerland

Range Imaging - from calibration to modelling
Wilfried Karel and Norbert Pfeifer - Institute of Geodesy and Photogrammetry, TU Vienna, Austria

Photogrammetric Calibration and Accuracy Test for a Range Camera
Jan Boehm - Dept. of Civil, Environmental and Geomatic Engineering, UCL London, UK

A Probabilistic Approach to TOF and Stereo Data Fusion
Carlo Dal Mutto, Pietro Zanuttigh, Guido M. Cortelazzo - Dept. of Information Eng., University of Padova, Italy

Time-of-Flight cameras in robotics - state of the art and selected applications
David Droeschel - Autonomous Intelligent System Group, University of Bonn, Germany

Beyond range: innovating fluorescence microscopy
Alessandro Esposito - Medical Research Council, Cambridge, UK

A 128x96 pixel 3D camera SOC with internal depth map generation capability
Richard Walker and Robert K. Henderson, Institute for Integrated Micro and Nano Systems, The University of Edinburgh, Edinburgh, U.K.

Range Imaging in Ambient Assisted Living Applications
Alessandro Leone - CNR, Institute for Microelectronics and Microsystems, Lecce, Italy

Electronics-based pixels for 3D range imaging
Matteo Perenzoni and David Stoppa - FBK, Trento, Italy

3D Vision based on CMOS SPAD Imagers
Edoardo Charbon and Shingo Mandai - TU Delft, Delft, Belgium

Design and characterization of demodulating pixels in 0.18um CMOS technology
Lucio Pancheri and David Stoppa - FBK, Trento, Italy

MESA Imaging: Product strategy and commercial applications for TOF cameras
Gerald Dahlmann - MESA Imaging AG, Zurich, Switzerland

Optrima Technology Overview about DepthSense and OptriCam
Daniel Van Nieuwenhove - Optrima NV, Brussels, Belgium

Are 40k pixels enough for everyone? Challenges and opportunities for PMD depth sensing technology in the consumer market
Jochen Penne - PMD Technologies, Germany

Technology Demo: Demonstration by MESA Imaging, Optrima NV and PMD Technologies

Thanks to A.T. for the link!

Saturday, December 18, 2010

Fujitsu Launched Surround View Processor for Automotive

EETimes: Fujitsu 360-degree Wrap-Around Video Imaging Technology synthesizes images from four cameras to recreate an omnidirectional view of a vehicle's surroundings. The system designers can select the optimal perspective to give the driver the best view for any given driving situation. It is even possible to present a virtual "third person" perspective of the vehicle from multiple angles. The technology eliminates blind spots and provides visual assistance when the driver is backing up, turning corners or merging. The solution is based on Fujitsu MB86R11 "Emerald-L" 2D/3D Graphics SoC.

Friday, December 17, 2010

3D News: Primesense, GestureTek, TYZX, Guinness

In a little more than one week passed since Primesense released its OpenNI drivers the open source community produced an amazing quantity of 3D-based games. Youtube has tens of videos demos of these games, simple applications and trials. Here is one of them:



Imaging Executive Podcast published an interview with GestureTek co-founder and president Vincent John talking about different projects with 3D cameras.

TYZX is in the Guinness World Records Museum, not as record-holder, but as part of a new interactive display created by design firm Electroland. When pedestrians pass the Hollywood Boulevard, LA - located museum, the three TYZX cameras detect their presence and triggers the display 3D avatars on a series of large flat-panel displays. The avatars perform various world-record feats, such as juggling chainsaws.

Update: Business Wire: Primesense announces the launch of its OpenNI (Open Natural Interaction) consortium, an industry-led, nonprofit organization formed to accelerate the introduction of Natural Interaction applications into the marketplace. The role of OpenNI is to certify and promote the compatibility and interoperability of Natural Interaction devices, applications and middleware, while encouraging developers to build applications and create experiences that will shape the future of Natural Interaction.

Thursday, December 16, 2010

Hamamatsu Enhances IR Sensitivity by MEMS Structures

Recently Hamamatsu has announced a number of silicon-based sensors with enhanced NIR sensitivity. These sensors are said to utilize MEMS structures fabricated by using "unique laser processing technology" to achieve a significant enhancement of NIR sensitivity. Another explanation of MEMS action is on Electro Optics site: "Using unique laser processing technology, MEMS structures can be fabricated on the silicon surface, which act to reduce reflections and increase the surface area of the active element. This process drastically increases the sensitivity in wavelengths longer than 800nm."

PIN diode NIR extension is quite big:


In case of BSI FT-CCD the MEMS structure is formed at the backside and NIR extension is somewhat less than in the PIN diode case:


MDT: The S11510 BSI CCD features QE "of 40% at 1,000 nm, without the need for a deep depletion structure, with its corresponding drawback of higher dark signal. The S11510 series is available with 1,024 or 2,048 pixels, with each pixel measuring 14 by 14 µm." The new NIR CCD products have been announced about a year ago, but the PR had no reference to the MEMS technology at that time.

Wednesday, December 15, 2010

JVC Developed 8.3MP/60fps Camera Processor

JVC has developed high-speed camera processors able to work with 8.3MP 60fps video stream. The LSI enables shooting and recording Full 2D/3D HD at 60p fps speed, and also ultra-high-resolution 4K2K (3840 x 2160/60p) images of approximately four times the resolution of Full HD. High-speed photography with high-speed processing also is possible, delivering JPEG processing at 8.3MP/60fps speed.

TowerJazz Presentation at ISE 2010

I was given TowerJazz presentation at Image Sensor Europe in March 2010 titled:

Foundry business for CMOS Image Sensors
Dr. Avi Strum, VP & GM Specialty BU

Below are few slides from the presentation talking about the company's approach and technology achievements:


A slide showing stitching quality for large sensors:


BSI becomes an option for TowerJazz customers:


Lightpipe is another available option:


All in all, it seems that access to a lot of advanced technologies is now available to the smaller players on the image sensor market.

Tuesday, December 14, 2010

Next Thing after PTC Series

Albert Theuwissen looks for the themes for his excellent blog, as PTC series nears its end. Nobody seems to propose the theme in two days passed since the request was posted, so I decided to duplicate it here.

Sunday, December 12, 2010

Small Cameras for Endoscopic Applications

BioPhotonics published an article on small cameras and their endoscopic applications. The overall endoscope market is growing at a modest 3% a year, said Teri Minor, a senior analyst at the technology market research firm Frost & Sullivan. But there are segments within it, such as the gastrointestinal one, that are growing much faster. “They have new technologies, like capsule endoscopy, which have growth rates through 2016 that could be as much as 18 or 19 percent a year,” she said.

Given Imaging second-generation pill-camera measures 11 x 26 mm and weighs <4 g. The camera snaps 35 pictures a second, and its wide field of view allows doctors to see twice the surface area of that seen with other capsules, said Given Imaging CEO Homi Shamir.

In September 2009 Medigus Ltd. of Omer, Israel announced a 1.2-mm video camera, claiming to have produced the smallest such device in the world. Medigus currently is working with medical device companies and academic teams around the globe to incorporate it into various devices. As the company's product development manager Ariel Smoliar noted, Medigus' sensor fits approximately 50,000 (220H X 224V) of 2.2um pixels into an area that measures 0.7 mm on a side. This is five times the number of fibers in a fiber bundle solution, making the small-area sensor effectively high resolution compared with that alternative.

Measuring only 1.2 mm in diameter and 5 mm in length, this CMOS camera for medical applications is as small as the point of a pen. Inside the camera cylinder are a sensor, four lenses and a miniature printed circuit board connected to a four-wire cable. The camera is designed to be used in disposable instruments.
Courtesy of Medigus Ltd.

The sensors for the Medigus camera are manufactured by TowerJazz. TowerJazz runs 8-in. wafers, which means there can be close to 50,000 half-square-millimeter sensors on each one. That multitude is part of what enables the sensors to be produced cheaply enough to render the camera disposable. The advantage of this approach is that it eliminates the need for sterilization between procedures, which can be expensive and a source of infection if not done properly.

The small size of these sensors – as compared with those in other applications – is not necessarily the result of going with a smaller pixel, said Avi Strum, TowerJazz VP and GM. The need to capture clear images in low light makes it difficult – if not impossible – to shrink the pixels much from the size of those used in devices such as cell phones.

Thanks to A.L. for sending me the link!

Friday, December 10, 2010

Omnivision Bus Tour Presentation

Omnivision published its Wedbush Bus Tour Presentation dated by Dec. 2, 2010. Most of the slides are taken from the Annual Shareholder Presentation from Sept. 2010 with some updates.

When talking about BSI-2 advantage, the company compares it with FSI pixels:



Omnivision quotes large market shares on few non-cellular markets and compares itself to Aptina:


The last year total market share is quite nice too:


Omnivision targets fast growing markets:


The resolution mix shifts toward 2MP+, but VGA is still more than a half of the company's unit sales:

Omnivision's VGA Sensor Awarded by Electronic Design Magazine

Electronic Design Magazine announced 2010 Best Electronic Design Award Winners. Omnivision OV7735 VGA sensor won the award in the sensor category. OV7735 is based on 3um FSI pixels, delivers sensitivity of 3300 mV/lux-sec and is said to enable less than 3mm camera module hight.

The article does not explain why OV7735 has been selected among many image sensors announced this year. “We rely on the expertise of our staff and contributing editors to ferret out the best of the many new technologies, products and standards that we have seen and wrote about over the past 12 months,” said Editor-in-Chief Joe Desposito. “These guys are in the trenches every day covering this industry and they know about all the great new innovations that have been introduced in the last year.

Thursday, December 09, 2010

Initial Speaker Line-Up at Image Sensors Europe 2011

2011 Image Sensors Europe Conference to be held in London, UK at March 22-24, 2011 has quite impressive preliminary list of speakers:

Ir. Jan Bosiers, R&D Director, DALSA Corporation, Netherlands
Dr Howard Rhodes, Senior VP Process Engineering, OMNIVISION TECHNOLOGIES, US
Tsutomu Haruta, Senior Manager, SONY CORPORATION, Japan
Lindsay Grant, Imaging Division Process Manager, ST MICROELECTRONICS, UK
Richard Crisp, Principal Technologist and Senior Director, TESSERA, US
Prof Karen Egiazarian, Professor - Department of Signal Processing, TAMPERE UNIVERSITY OF TECHNOLOGY, Finland
Dr Sami Khawam, CTO, RICAtek, UK
Dr Thomas Baechler, Section Head Image Sensing, CSEM, Switzerland
Dr Edoardo Charbon, Professor of Microelectronics, DELFT UNIVERSITY OF TECHNOLOGY, The Netherlands
Dr Renato Turchetta, Science & Technology Facility Council, RUTHERFORD APPLETON LABORATORY, UK
Dr Randy Bockrath, CEO, IMATEST LLC, US
Dr York Haemisch, Senior Director Corporate Technologies, PHILIPS RESEARCH LABORATORIES, Germany
Dr Nils Friedrich, Process Technology, PMD TECHNOLOGIES GmbH, Germany
Dr Sandro Tedde, Research Scientist, SIEMENS AG, Germany
Dr Guy Menants, VP R&D, CMOSIS, Belgium
Speaker from POLITECNICO DI MILANO, Italy

The final program is yet to be finalized. Thanks to R.S. for letting me know!

Wednesday, December 08, 2010

3D Sensing News: TYZX, Optrima, Primesense

PR Newswire: TYZX announced it has received a substantial investment from Takata, a global manufacturer of automotive safety systems with more than 33,000 employees worldwide. The Takata investment follows an earlier investment by Microsoft co-founder Paul Allen. TYZX is rumored to be profitable since the first investment.

Industry experts believe that vehicles with 3D vision safety features can help better protect drivers and passengers in collisions, help cars avoid accidents, keep vehicles in appropriate road lanes, and prevent many accidental backup injuries and fatalities, among other benefits.

Meanwhile, Softkinetic-Optrima demoed their system in TI booth at Embedded Show in Yokohama, Japan, Dec. 1-3, 2010. A Youtube video shows the demo:



Update: Primesense has released the drivers to its platform as open source and founded OpenNI.org to promote the further defelopment. In the meantime the open community efforts to reverse engineer Primesense algorithms go on. ROS.org published its take on how the Primesense engine works. Here is the explanation from the site:

"Depth is calculated by triangulation against a known pattern from the projector. The pattern is memorized at a known depth. For a new image, we want to calculate the depth at each pixel. For each pixel in the IR image, a small correlation window (9x9 or 9x7, see below) is used to compare the local pattern at that pixel with the memorized pattern at that pixel and 64 neighboring pixels in a horizontal window (see below for how we estimate the 64-pixel search). The best match gives an offset from the known depth, in terms of pixels: this is called disparity. The Kinect device performs a further interpolation of the best match to get sub-pixel accuracy of 1/8 pixel... Given the known depth of the memorized plane, and the disparity, an estimated depth for each pixel can be calculated by triangulation."

Tuesday, December 07, 2010

US Cars May Be Required to Have Cameras

CNN: The National Highway Traffic Safety Administration (NHTSA) proposes a new auto safety regulation that would require rearview back-up cameras in all new cars, pickups and SUVs by 2014. Based on the proposal, drivers must be able to see directly behind the vehicle whenever the vehicle is shifted into reverse. The rule would be phased in over the next four years, starting with 10% of new cars sold expected to comply with the mandate by Sept. 2012; 40% by Sept. 2013 and 100% by Sept. 2014.

The agency estimates that the addition of rear-view camera equipment would cost between $159 to $203 per car, or $88 to $158 on vehicles already equipped with some sort of display screen -- like one used for navigation. NHTSA says the total approximate cost to equip their estimate of 16.6 million vehicles sold in 2014, would be between $1.9B and $2.7B.

Via electronsholes.

Pixart, TPV, Cheng Uei to Form Optical Touch JV

Taiwan Economic News: Cheng Uei Precision Industry, a supplier of connectors for electronic devices, TPV Technology Ltd., the world`s largest TV maker, and Pixart Imaging plan to set up a joint venture on production of optical touch modules, Guo Tai-chiang, chairman of Cheng Uei said.

Industry insiders noted that the joint venture will source touch sensors and ICs from Pixart and optical touch components from Cheng Uei, while relying on TPV`s enormous influence in the monitor industry to cash in on rapidly growing demand for Internet TVs and all-in-one PCs. TPV is expected to ship 15 million LCD TVs and 55 million units of LCD monitors and all-in-one PCs to sustain its No. 1 position in the industry this year.

Sunday, December 05, 2010

Hands-On Image Sensor Evaluation Course

Albert Theuwissen published his experience of teaching the first "Hands-On Evaluation of Image Sensors" course. The 2-day course took place two weeks ago in Barcelona under CEI-Europe umbrella. Next time the improved version of this unique course will be delivered in May 2011 in Copenhagen.

Saturday, December 04, 2010

Panasonic 3D D-Imager is now Off-the-Shelf Product Available from Distributors

PR Web: Panasonic 3D ToF D-IMager is now available for purchase through Arrow and Digi-Key.

The proprietary CCD-based D-IMager's has an average current consumption of 0.4A out of 24V DC supply, allowing it to be lightweight and fan-less. Other spec parameters are quoted below:

Friday, December 03, 2010

Hynix Targets Notebooks with its 1MP 720p Video Sensor

Hynix announced 1/6-inch 1MP 720p/30fps HD video sensor aimed to notebook market. The YACC611CBDFH (Hi-162) sensor outputs raw RGB and needs an external ISP. The sensor is based on 2.0um pixels and packaged in ShellUT CSP package.

Thursday, December 02, 2010

CMOSIS Announces 600fps VGA Sensor

CMOSIS previewed 1/3" VGA (648 x 488) format sensor. The new CMV300 will use the same 8T pixel structure as the high-resolution series CMV2000/CMV4000 / CMV12000. With 7.4 µm square pixels, a noise level of 9 e- and a full well charge of 20,000 electrons, the new CMV300 realizes a dynamic range of more than 66 dB, which can be extended through several HDR operational modes.

The CMV300 pixel architecture features its low parasitic light sensitivity and CDS in global shutter mode. The sensor's 12-bit ADC delivers 600 full VGA frames per second over four LVDS outputs of 600 Mbit/sec each.

CMV300 comes in a monochrome version fitted with micro lenses, and in a color version featuring RGB Bayer patterns as well as micro lenses. Prototypes will be out by mid-2011.

Sony CineAlta Moves Beyond Bayer

Hugo Gaggion, Sony Broadcast & Professional Solutions CTO, presented 2K/4K roadmap at NYC 2010 HD World show in mid-October. Sony published a video of this presentation discussing color filter pattern innovations, among many other things.

The previous generation of Sony CineAlta CCD-based Super 35mm cameras used linear RGB pattern:


To maintain the high resolution in green in all directions, including diagonal one, Sony proposes a new Q67 pattern for new 4K camera systems:


Bayer Pattern can produce a lot of artifacts:


Sony Q67 pattern produces less color artifacts:


Eventually Sony plans to switch over to full-color 4K sensors:


The competitive landscape on Super 35mm camera market, as Sony sees it:


Thanks to A.A. for sending me the link!

Wednesday, December 01, 2010

Andor Launches Cooled sCMOS Camera

Andor announced the launch of the Neo camera platform based on scientific CMOS (sCMOS) technology. The company calls Neo sCMOS the most significant camera launch in the recent times. The sensor spec has been improved by TE cooling down to -40C. Specifically, the read noise went down to 1e- at 30fps full resolution speed and 1.4e- at 100fps speed (was 2e in not cooled sensor).

Thanks to R.C. for sending me the info!

EVG Improves Wafer Level Lens Molding Process

PR Newswire: EV Group (EVG) announced that it has developed a new micro-lens molding process that can enable volume production of very-high-resolution (up to 8MP and higher) wafer-level optics for use in smart phones, pico projectors and other applications. The new Monolithic Lens Molding (MLM) capability, which was developed in-house by EVG's process development team, is available as an option on the company's IQ Aligner UV nanoimprint lithography (UV-NIL) system or can be upgraded to existing equipment. EVG expects to ship its first IQ Aligner with the MLM option in the first half of 2011.

In wafer-level camera production today, glass substrates are typically used as carrier and spacer wafers for the lenses, which are composed of an optical polymer material. The different material characteristics of these components limit resolution and picture quality. EVG's MLM process overcomes this limitation by eliminating the need for glass substrates. Instead, the polymer is molded between two stamps and then cured with UV exposure by the EVG's IQ Aligner system. By omitting the glass substrates, wafer-level optics manufacturers face fewer constraints on the optic and lens stack design—enabling the production of thinner lens wafers and significantly shorter optical stacks.

In addition, since the IQ Aligner molds the micro-lenses using a room-temperature UV-NIL process versus thermal imprinting, a high degree of precision alignment is achieved between the various elements in the optical lens stack—maximizing device performance.

Omnivision Pleased with OmniBSI-2 Progress

Seeking Alpha published Omnivision quarterly earnings call transcript. Some interesting quotes:

Shaw Hong, CEO:

...we shipped record volumes of image sensors exceeding 180 million units. This is the highest level of units ever shipped by OmniVision in the first quarter.

...Now pursuing design wins for BSI-2 and we believe we are more than a generation ahead of most competitors. We anticipate shipping BSI-2 based products in the next calendar year.

...our CameraCube products are gaining traction and we are shipping those products in steady volumes.


Ray Cisneros, VP WW Sales:

In our second fiscal quarter we shipped a record high of nearly of 185 million units as compared to 134 million units in our prior quarter. The average selling price in our fiscal second quarter was $1.29 as compared to $1.44 in the prior quarter.

The sequential drop in the average selling price was driven by the shift in our products mix mainly due to a surge in shipments of our new BSI HD sensor OV9726 and our new specialized VGA product of the OV3738. These chips are ... smaller in size and therefore carry a lower ASP.

...OmniBSI has been adopted ... by all Tier 1 customers in all segments.

...we're going to refrain from breaking out too much detail in regards to our specific plans around BSI-2. ... we are extremely, extremely pleased with the engineering progress that has transpired with this platform. We are extremely pleased with the product developments that are directly taking place today and we're very, very pleased with our engagements with Tier 1 leading customers to adopt Omni BSI-2. All that being said we're lining up for some time within the next calendar year to start ramping up production and ramping up initial projects with customers. I would say sometime in the early, the latter half of the Q2, Q3 timeframe calendar year next year would be the ramp of some of these activities.

Anteryon Shows Profits but its IPO Delayed

BNR quotes Reuters saying that Anterion initially aimed at an IPO in the third quarter of this year. Early next year it must be clear how soon that step can be realized. Partly to the poor stock market climate, the company has decided to attract a strategic investor - Qualcomm.

Meanwhile, this year for the first time since its separation from Philips in 2006 Anteryon has shown profits on a turnover of around 20M euros. The company expects that sales will grow next year. Anteryon supplies its wafer level lenses to all major mobile phone brands including Nokia, Samsung and Motorola.

CAT PUMA Pursues Image Sensor Applications

Chip Estimate: Cambridge Analog Technologies (CAT) has developed Precision Ultra Micro-power Amplification (PUMA) technology-based ADC IP applicable to image sensor applications. The company implements MIT-developed low-power switched capacitor techniques. While most switched cap circuits rely on op-amps, CAT uses zero-cross detectors (ZCD) to switch on/off capacitor charging sources:


The ZCD-based circuits are said to consume one to two orders of magnitude less power than op-amp for the same settling time and accuracy. Now power numbers are given in the article, while white papers published on the company's site suggest much smaller savings.

Tuesday, November 30, 2010

Omnivision Reports Quarterly Results

PR Newswire: Omnivision reports financial results for the fiscal second quarter 2011 ended October 31, 2010. Revenues for the quarter were $239.5M, as compared to $193.1M in the previous quarter, and $183.3M a year ago. GAAP net income for the last quarter was $28.9M, as compared to $16.9M in the previous quarter, and $8.1M a year ago.

Gross margin for the last quarter was 28.2% as compared to 26.9% for the previous quarter and 24.0% a year ago. The sequential increase in the gross margin reflected a continuing yield improvement in the company's BSI products. Omnivision ended the period with cash, cash equivalents and short-term investments totaling $396.7M, an increase of $47.7M from the previous quarter.

Based on current trends, the company expects next quarter revenues will be in the range of $230M to $250M.

Monday, November 29, 2010

Broadcom Buys ISP Vendor SightIC

Globes: Broadcom acquired Israeli image processing company SightIC Vista. The acquisition was reportedly for $10-20 million - below the threshold requiring notification.

SightIC president and CEO Noam Sorek, VP product management Ron Fridental, and VP R&D Dr. Ilia Vitsnudel founded the company in 2001. The three specialists in video and imaging processing had worked together for a decade before co-founding the company.

SightIC products include stand-alone video ISP and ISP IP blocks for licensing. Other than generic ISP pipeline, the company developed improved digital zooming and stills and video stabilization IP. SightIC has customers from among the world's top electronics manufacturers, including Samsung and LG. Relatively little was invested in the company by private investors, with no involvement by venture capital funds.

Sunday, November 28, 2010

Caltech Sues STMicro, SETi, Siliconfile, Toshiba, Nokia, LG, Pantech Over Camera Patents

Bloomberg: Nokia, LG and "chipmakers including STMicroelectronics" were sued by the California Institute of Technology, which said the companies’ mobile phones and components infringe its patents.

Caltech lists nine patents on camera technology that Nokia, LG and Pantech phones allegedly infringe. Also named in the Nov. 24 complaint were SETi, Siliconfile and Toshiba.

The case is California Institute of Technology vs. STMicroelectronics, 10-9099, U.S. District Court, Central District of California (Los Angeles).

Two years ago Caltech sued few camera companies on CMOS sensor patents. Some of them settled the issue with Caltech, some others denied the infringement.

Saturday, November 27, 2010

Sony Announces New CCD, CMOS Sensors, GVIF Automotive Interface

Cx-News vol 62: Sony announced 2/3-inch 2.83MP ICX674ALG/AQG CCDs capable to deliver 1080p HD video in B&W and color versions.

The main differentiation of the new 4.54um-pixel CCDs is "Output Channel Count Switching":
The CCDs have vertical transfer registers that can transfer charge in both directions as well as horizontal transfer registers. By changing the vertical and horizontal transfer register drive timing and by switching the charge transfer direction according to the required frame rate, these devices support single-channel output, horizontal division 2-channel output, vertical division 2-channel output, and horizontal and vertical division 4-channel output:


1/4-inch 1.4MP IMX076LQZ/CQZ CMOS sensors are targeting industrial applications and feature new high sensitivity 2.8um pixel (somehow Sony calls it "2.8um diagonal"). As a result of the improvements, Sony was able to acquire in 1/4-inch sensor sensitivity and SNR equivalent to the Sony's existing 1/3-inch 1.39MP CMOS sensors. The new sensors deliver 30 fps in 12-bit A/D conversion mode and 30 and 60 fps in 10-bit full resolution mode. IMX076LQZ is optimized for the long exit pupil distance while IMX076CQZ is developed for the short exit pupil distance.



Sony also introduces 1.95Gbps Gigabit Video Interface (GVIF) for automotive applications. The interface transmits 24-bit color data with EGA, VGA, WVGA, SVGA, Dual VGA, or XGA resolution to the distance of up to 10m over a shielded twisted pair. Sony seems to use no pre-emphasis on the transmitting side, so the receiver needs quite a heavy equalization to open the eye:


Sony shows a prototype of camera module with GVIF interface:

Friday, November 26, 2010

Yole and Chipworks Analyse Omnivision BSI Packaging in iPhone 4

November issue of Yole Développement's Packaging Magazine has an article on iPhone BSI sensor packaging. The 5 Mp iPhone camera module is said to be assembled by LG Innotek. The module dimensions are 9.2 mm x 9.2 mm x 6.2 mm thick:


One interesting discovery was that gold studs are used to connect the die bond pads instead of TSV. This type of packaging for a CIS application has typically only been seen in some DSLR camera sensors. While OmniVision/TSMC do have a TSV process for BSI parts, the back bonding scheme has provided what is likely a higher yielding alternative that satisfied Apple’s specification:

Wednesday, November 24, 2010

IEEE Sensors 2010 Conference Papers and Awards

IEEE Sensors 2010 conference held on Nov. 1-4 in Hawaii had few image sensor papers:

IN-PIXEL BURIED-CHANNEL SOURCE FOLLOWER IN CMOS IMAGE SENSORS EXPOSED TO X-RAY RADIATION
Y. Chen, J. Tan, X. Wang, A.J. Mierop, and A.J.P. Theuwissen
Delft University of Technology, THE NETHERLANDS
CMOSIS nv, BELGIUM
DALSA B.V., THE NETHERLANDS
Harvest Imaging, BELGIUM

LOW-POWER AND HIGH-SPEED CURRENT-MODE CMOS IMAGER WITH 1T BIASING SCHEME
F. Tang and A. Bermak
Hong Kong University of Science and Technology, HONG KONG

MISMATCH REDUCTION FOR DARK CURRENT SUPPRESSION
D. Sander and P. Abshire
University of Maryland, USA

ARCHITECTURE OF THREE DIMENSIONAL COMPRESSIVE ACQUISITION CMOS IMAGE SENSOR
M. Zhang, A. Bermak, and P. Xu
Hong Kong University of Science and Technology, HONG KONG

A 64 x 64 PIXELS 30μW VISION SENSOR WITH BINARY DATA COMPRESSION
N. Massari, M. De Nicola, N. Cottini, and M. Gottardi
Fondazione Bruno Kessler, ITALY

A VARIABLE TOPOLOGY PARTITIONED PIXEL AMPLIFIER FOR LOW AND HIGH LIGHT LEVEL DETECTION IN A CMOS IMAGER
Y. Dattner, and O. Yadid-Pecht
University of Calgary, CANADA

The best poster award went to:

FLUORESCENT IMAGING AND LOCALIZATION WITH ANGLE SENSITIVE PIXEL ARRAYS IN STANDARD CMOS
A. Wang, P.R. Gill, and A. Molnar
Cornell University, USA

The 2nd best student paper award went to Yang Xu for her master thesis:

A CMOS IMAGE SENSOR WITH CHARGE DOMAIN INTERLACE SCAN
Y. Xu, A. Mierop, and A.J.P. Theuwissen
Delft University of Technology, THE NETHERLANDS
DALSA Professional Imaging, THE NETHERLANDS
Harvest Imaging, BELGIUM

Interlace scanning in CCDs was/is always done in the charge domain, in CMOS this was done in the voltage or digital domain with extra loss of SNR. In the work of Xu, charge domain interlace scanning in CMOS imagers is realized for the first time, with the same SNR characteristics as with CCDs.


Thanks to A.T. for letting me know!

CMOSIS Develops Dual Gain Readout HDR Sensor

New Electronics: CMOSIS has demonstrated a 1MP image sensor featuring dual gain readout to reach DR of more than 89.5dB. The DR is increased by two approaches. First, the dark noise is said to be reduced. Second, the integration of the image charge is said to take place simultaneously at low and high amplification settings. This dual gain readout results in two different images which are read out sequentially and combined externally into one final image. The sensor has twice the number of column amplifiers than columns and the gain of each amplifier is programmable up to 16.

This approach is to be used in a number of future products with different pixel size, frame rate, etc.

Update: Electronics Weekly too published an article about the new CMOSIS 1MP imager dubbed CLN1000 and featuring 10um pixels. The sensor consumes 300 to 350mW, and is housed in a JLCC-84 package. The article also states low dark noise of less than 3.4e- at a full well capacity of 100 ke-. EETimes quotes the same numbers too.

Update #2: Photonics Online published an article about the CMOSIS prototype too.

Update #3: Here is the original CMOSIS PR.

Monday, November 22, 2010

ISSCC 2011 Abstracts Published

The new version of ISSCC 2011 Advance Program includes abstracts. Some of them are quoted below:

A 1.32pw/frame∙pixel 1.2v CMOS Energy-Harvesting and Imaging (EHI) APS Imager
S. U. Ay
University of Idaho, Moscow, ID

A CMOS energy-harvesting and imaging (EHI) APS imager capable of 7.4fps video capture and 3.5μW power generation is designed, fabricated, and tested in 0.5μm CMOS. It has a 54×50 array of 21μm2 EHI pixels, 10b supply-boosted SAR-ADC and charge-pump circuits consuming 14.25μW from 1.2V resulting in a lowest power imager with 1.32pW/frame∙pixel.

An 80μvrms-Temporal-Noise 82dB-Dynamic-Range CMOS Image Sensor with a 13-to-19b Variable-Resolution Column-Parallel Folding-Integration/Cyclic ADC
M-W. Seo, S. Suh, T. Iida, H. Watanabe, T. Takasawa, T. Akahori, K. Isobe,
T. Watanabe, S. Itoh, S. Kawahito
Shizuoka University, Hamamatsu, Japan
Brookman Technology, Hamamatsu, Japan
Sanei Hytechs, Hamamatsu, Japan

A 1Mpixel, 7.5μm pixel pitch, 0.18μm CMOS image sensor with column-parallel folding-integration and cyclic ADCs has 80μVrms (1.2e-) temporal noise, 82dB dynamic range using 64 samplings in the folding-integration ADC mode. Variable gray-scale resolution of 13b through 19b is obtained by changing the number of samplings for pixel outputs.

A Sub-Electron Readout Noise CMOS Image Sensor with Pixel-Level Open-Loop Voltage Amplification
C. Lotto, P. Seitz, T. Baechler
Heliotis, Root Längenbold, Switzerland
CSEM, Zurich, Switzerland
CSEM, Landquart, Switzerland
EPFL, Neuchâtel, Switzerland

A 256×256 pixel, 11μm pixel pitch, 0.18μm CMOS image sensor featuring pixel-level open-loop voltage amplification reaches a readout noise of 0.86e- and a dynamic range of 90dB in a single readout using a low-complexity readout circuit with 60fps. A reset method based on negative feedback allows the use of open-loop amplification while achieving PRNU of 2.5% and a peak linearity error of 1.7%.

A 300mm Wafer-Size CMOS Image Sensor with In-Pixel Voltage-Gain Amplifier and Column-Level Differential Readout Circuitry
Y. Yamashita, H. Takahashi, S. Kikuchi, K. Ota, M. Fujita, S. Hirayama, T. Kanou,
S. Hashimoto, G. Momma, S. Inoue
Canon, Kawasaki, Japan

A 1.6Mpixel, 202×205mm2 CMOS image sensor on 300mm wafer consists of pixels of 160μm pitch with a 0-to-24dB variable gain in-pixel voltage amplifier. Reset and integrated signals are simultaneously read out from the pixel through a pair of column lines. It achieves a sensitivity of 25Me-/lux∙s, random noise of 13e-rms and operates at 100fps with global synchronous shutter.

An Angle-Sensitive CMOS Imager for Single-Sensor 3D Photography
A. Wang, P. R. Gill, A. Molnar
Cornell University, Ithaca, NY

A 0.18μm 3D CMOS image sensor composed of angle-sensitive pixels captures both local incident angle and intensity. The 400×384 pixel array has a 7.5μm pitch and local diffraction gratings over each pixel. One such chip, using one lens and ambient light, enables post-capture refocus and range finding accurate to ±1.3cm at 50cm.

A 1/13-inch 30fps VGA SOC CMOS mage Sensor with Shared Reset and Transfer-Gate Pixel Control
R. Johansson, A. Storm, C. Stephansen, S. Eikedal, T. Willassen, S. Skaug, T. Martinussen, T. Whittlesea, G. Ali, J. Ladd, X. Li, S. Johnson, V. Rajasekaran, Y. Lee, J. Bai, M. Flores, G. Davies, H. Samiy, A. Hanvey, D. Perks
Aptina Imaging, Oslo, Norway
Aptina Imaging, Bracknell, United Kingdom
Aptina Imaging, San Jose, CA
Aptina Imaging, Corvallis, OR

This paper describes a 1/13-inch VGA SoC CMOS image sensor, with a 1.75μm pixel pitch capable of outputting 30fps at full resolution. The paper focuses on the sensor core, with a size of 1.77mm2 and a power consumption of 17mW. Thel pixel architecture uses a pixel-sharing scheme that improves low-light performance.

A 1/2.33-inch 14.6M 1.4μm-pixel Backside-Illuminated CMOS Image Sensor with Floating Diffusion Boosting
S. Lee, K. Lee, J. Park, H. Han, Y. Park, T. Jung, Y. Jang, B. Kim, Y. Kim, S. Hamami, U. Hizi, M. Bahar, C. Moon, J. Ahn, D. Lee, H. Goto, Y-T. Lee
Samsung Electronics, Yong-In, Korea
Samsung Semiconductor, Ramat-Gan, Israel

A 1/2.33-inch 14.6Mpixel CIS is developed by employing a 1.4μm BSI pixel with a floating diffusion boosting scheme driven by coupling with additional row-wise metal-line, achieving 30% higher QE than that of an FSI sensor, 87lux for SNR=10, and no image lag, for high-sensitivity and high-speed applications.

An APS-C Format 14b Digital CMOS Image Sensor with a Dynamic Response Pixel
D. Pates, J-H. Lyu, S. Osawa, I. Takayanagi, T. Sato, T. Bales, K. Kawamura, E. Pages, S. Matsuo, T. Kawaguchi, T. Sugiki, N. Yoshimura, J. Nakamura, J. Ladd, Z. Yin, R. Iimura, X. Fan, S. Johnson, A. Rayankula, R. Mauritzson, G. Agranov
Aptina Imaging, San Jose, CA
Aptina Imaging, Tokyo, Japan
Aptina Imaging, Bracknell, United Kingdom

A 16M APS-C format CMOS image sensor with 14b SAR-ADC and 8-lane LVDS output is fabricated and characterized. A 4.78μm dynamic response pixel with ring gate transistors and no STI provides 62% QE, responsivity of 49.5ke-/lux∙s, and dark current of 17e-/s @ 60°C. The readout noise floor is 2.2e- with column FPN of 0.11e- in HCG mode.

A 17.7Mpixel 120fps CMOS Image Sensor with 34.8Gb/s Readout
T. Toyama, K. Mishina, H. Tsuchiya, T. Ichikawa, H. Iwaki, Y. Gendai, H. Murakami, K. Takamiya, H. Shiroshita, Y. Muramatsu, T. Furusawa
Sony, Kanagawa, Japan
Sony LSI Design, Kanagawa, Japan

A 17.7Mpixel CMOS image sensor with a 27.5mm optical format realizes 120fps at 12b using 90nm CMOS. This sensor achieves 2.75e-rms random noise at 12b, 120fps with a maximum data rate of 34.8Gb/s. The 16 channels of scalable low-voltage signaling I/F with embedded clock operate at 2.376Gb/s each and the single-slope ADC ramp generator runs at 2.376GHz.

A 160×128 Single-Photon Image Sensor with On-Pixel 55ps 10b Time-to-Digital Converter
C. Veerappan, J. Richardson, R. Walker, D-U. Li, M. W. Fishburn, Y. Maruyama, D. Stoppa, F. Borghetti, M. Gersbach, R. K. Henderson, E. Charbon1
Delft University of Technology, Delft, The Netherlands
STMicroelectronics, Edinburgh, United Kingdom
University of Edinburgh, Edinburgh, United Kingdom
Fondazione Bruno Kessler - IRST, Trento, Italy
EPFL, Lausanne, Switzerland

A 160×128 pixel array is presented detecting photons with 55ps time resolution. Each pixel comprises a counter, a time-to-digital converter, and a 10b memory, while a frame is read out every 20μs. The sensor is well suited for applications such as fast fluorescence lifetime imaging, optical rangefinding, and time-correlated single-photon counting.

Ambarella Introduces 3D HD Processors

EON: Ambarella announces a complete 3D camera solution that includes the new S3D, a 3D video pre-processor. When combined with Ambarella’s A7 HD camera SoC, the S3D enables development of cameras up to 1080p60 resolution, while combining it with Ambarella’s A5s HD camera SoC enables resolutions of 1080p30, 1080i60 or 720p60. Both the A7 and A5s feature dual stream H.264 encoding, image stabilization with rolling shutter correction for CMOS sensors and wireless connectivity options.

The S3D can accept input from CMOS sensors with MIPI, parallel, or SLVDS interfaces and supports runtime parallax adjustment and HDMI 1.4 3D information.

The S3D and full 3D camera reference design kit with A7 or A5s SDK are available immediately.

Saturday, November 20, 2010

2010 Image Sensor Market by Region

Newswire Today: Electronics.ca announces a new report titled "Image Sensors Market Report". According to the report, Asia Pacific region is the largest single region by sales with $3.3B projected for 2010 and average growth of 10% annually through 2015. Korean and Taiwanese suppliers are gaining in market share, especially for camera phones.

Japan has traditionally dominated image sensor production, and still manufactures over 90% of the world's CCDs. Japan will remain the second largest region for production with $2.2B in revenue expected for 2010.

The Americas' share of the image sensor market has jumped from 3% in 2007 to over 12% in 2010. This jump is likely the result of some large image sensors supplier recording revenue for sales to a different region in 2007, and now recording it in the Americas region, either due to a relocation of manufacturing, or a reclassification in the region originally credited for the sale to better reflect where the products were actually shipped.

Europe is currently the smallest region by sales with $416M in revenue and 6% of the total market projected for this year.

Looking at the bigger picture, the whole optoelectronics market is $22.5B this year and is expected to reach $37B by 2015. Segments within this category include Image Sensors, Lamps, Displays, Couplers, Infrared Devices, Laser Pickups, Laser Transmitters, and Other Optoelectronics. Lamps (LEDs) constitute the majority share of the world's optoelectronics revenue with 33% share, followed by image sensors with 30% share.

Friday, November 19, 2010

Pixelplus Shows Nice Quarterly Profit (for the company size)

PR Newswire: Pixelplus announced financial results for the Q3 2010, which ended on September 30, 2010.

The quarterly net income was $1.4M, compared to a net income of $600K in Q2 2010, and a net income of $150K in Q3 2009. The gross margin was 40.5%, compared to 33.6% in Q2 2010. The company's increase in gross margin was mainly due to its sales of high margin-driven products.

The revenue for the quarter was $6.1M, compared to $6.0M in Q2 2010, and $3.1M in Q3 2009.

"The key to our consistent growth and success will be our ability to transform our organization to achieve long-term profitability. For this aim, I believe our strategic and ongoing cost-control measures should help us decrease our cash requirements, concentrate on our core business, maintain our continuity, and improve the efficiency of our operations," said Dr. S.K. Lee, CEO and Founder of Pixelplus, "We continue to dedicate all of our efforts and energies on achieving our long-term growth targets and business strategies through developing new products, entering new markets, and securing new design wins."

4MP/800fps 35mm-Format Camera Announced

Vision Research unveils Phantom v341 - a high-speed camera tailored to military, scientific, research, and range applications.

The v341 incorporates a 4MP/800fps 35mm format sensor with 3 gigapixels/second throughput. The custom-designed 2560 x 1600 pixel CMOS sensor featurs peak QE of 60%, global shutter speed down to 1us and Extreme Dynamic Range (EDR) mode with two different exposures within a single frame. Color and monochrome versions are available.

Thursday, November 18, 2010

DALSA Announces 300fps X-Ray Sensors

Web Newswire: DALSA announced Xineos-1313-EO CMOS X-Ray detector panel designed to operate at full resolution frame-based panoramic dental imaging at frame rates exceeding 300 fps, as well Medium Field of View (MFOV) Cone Beam CT applications at 30 fps. The panel resolution is not clear from the announcement.

3D News Roundup

Panasonic seems to be busy promoting its 3D ToF D-Imager in the 6 Youtube videos. Most of the videos appeared earlier at Panasonic site. One of them talks about security applications with D-Imager providing the data to enable accurate person and anti-tailgating detection:



Meanwhile, the reverse engineering of Microsoft Kinect goes on. Future Picture site explores a projection pattern that allows Kinect to sense the depth. The red cycles show repeating elements in the pattern:


The full patten is arranged as 3x3 matrix of dots. Each part has different brightness and centering dot:


Via electronsholes.

Update: The open access Sensors journal published the Stanford University paper titled "Photo-Detectors for Time of Flight Positron Emission Tomography (ToF-PET)". The paper reviews recent advances in time of flight positron emission tomography (ToF-PET) based on two types of photo-detectors: photomultiplier tubes (PMTs) and silicon photo-multipliers (SiPMs) with a special focus on SiPMs.

Silicon Photodiode QE Extended to 80% due to Surface Plasmon Resonance

OptoIQ: In a paper to be presented on the upcoming IEDM a team led by the National Nano Device Laboratories (Hsinchu, Taiwan) will report an external QE of >80% for photodiodes in the visible range. The QE extension was achieved by harnessing a local surface plasmon resonance (LSPR) effect "due to self-accumulated local electric field in these corners" of a metal pattern on top of the photodiode. Each pattern has a period of 2.4µm and unit area of 1.44µm2.

Wednesday, November 17, 2010

Aptina Presents HDR Demo Video

Aptina uploaded a nice Youtube presentation video showing the need in HDR imaging and performance of its MT9V022 VGA sensor:

Omnivision Announces 1/10-inch Sensor for Endoscopy

PR Newswire: Omnivision announced a start of mass production of the ultra-compact 1/10-inch OV6930 sensor with a packaged footprint of only 1.8 x 1.8 mm. It is targeted to serve a broad range of medical endoscopy applications, including bronchoscopy, colonoscopy, gastroscopy, OB/GYN and urology, and can facilitate development of medical devices that require an outside diameter under 2.6 mm.

The OV6930 relies on OmniPixel3-HS pixel with low-light performance of 3300 mV/lux-sec and is capable of operating up to 30fps in 400 x 400 HVGA or 60 fps in 400 x 200 resolution. The sensor allows cabling up to 14 ft.

OmniVision also announced OV6930 design wins for two new medical device modules from Hong Kong based COMedia, Ltd. The OV6930 is now shipping in volume to multiple customers.

Tuesday, November 16, 2010

Advasense Offers Low-Noise Sensor IP

Advasense is now offering low-noise image sensor IP blocks coming from its CIS technology.

Two basic families of IP are offered:

1. FCP IP for small pixels of 1.1u and with high FWC that is said to pave the way to small pixels

2. CDS (4T) Readout and ADC with very low read noise for 4T pixel designs:

The ADVA004 CDS based Readout Pipeline is said to have the lowest temporal noise in the industry of 40uV. The design is compact and low power. The Readout (RO) Pipeline can be connected to any 4T pixel array.

ADVA004 Key Parameters:
  • Readout noise level is 40uV
  • Column level ADC- 2nd order sigma delta,
    12bit resolution (scalable via decimation filter length),
    Conversion time is less than 2us.
  • Easily connected to any 4T pixel array
  • Soft IP Block or Hard IP block
The ADC and CDS Readout are also offered as separate blocks ADVA002 and ADVA003 respectively. Advasense is already working with several companies and probably license them its IP.

Thanks to A.G. for sending me the link!

Monday, November 15, 2010

Image Sensors at ISSCC 2011

ISSCC 2011 advance program has been published and, as always, has a lot of interesting image sensor content:

Session 23 - Image Sensors

23.1 An 80μVrms-Temporal-Noise 82dB-Dynamic-Range CMOS Image Sensor with a 13-to-19b Variable-Resolution Column-Parallel Folding-Integration/Cyclic ADC,
Shizuoka University; Brookman Technology; Sanei Hytechs


80uV noise is an excellent number. However, 19db ADC seems to be an overkill for the stated DR of 82dB.

23.2 A Sub-Electron Readout Noise CMOS Image Sensor with Pixel-Level Open-Loop Voltage Amplification,
Heliotis; CSEM; EPFL


23.3 A 320×256 90dB SNR and 25μm-Pixel-Pitch Infrared Image Sensor,
CEA-LETI-MINATEC; Sofradir


23.4 A 16 Mfps 165kpixel Backside-Illuminated CCD,
Kinki University; DALSA PI; University of Arizona; NHK Science and Technical Laboratories


23.5 A 300mm Wafer-Size CMOS Image Sensor with In-Pixel Voltage-Gain Amplifier and Column-Level Differential Readout Circuitry,
Canon


23.6 A 128×96 Pixel Event-Driven Phase-Domain ΔΣ-Based Fully Digital 3D Camera in 0.13μm CMOS Imaging Technology,
University of Edinburgh; STMicroelectronics


23.7 An Angle-Sensitive CMOS Imager for Single-Sensor 3D Photography,
Cornell University


Seems that many groups work on single-lens 3D cameras.

23.8 A 1/13-inch 30fps VGA SoC CMOS Image Sensor with Shared Reset and Transfer-Gate Pixel Control,
Aptina Imaging


23.9 A 1/2.33-inch 14.6M 1.4μm-Pixel Backside-Illuminated CMOS Image Sensor with Floating Diffusion Boosting,
Samsung Electronics; Samsung Semiconductor


23.10 An APS-C Format 14b Digital CMOS Image Sensor with a Dynamic Response Pixel,
Aptina Imaging


23.11 A 17.7Mpixel 120fps CMOS Image Sensor with 34.8Gb/s Readout,
Sony; Sony LSI Design


- Sony steadily increases speed of its sensors. Just at this year ISSCC Sony presented 10MP/50fps sensor and now there is a new advance.

Session 6 - Sensors & Energy Harvesting

6.6 Indirect X-ray Photon-Counting Image Sensor with 27T Pixel and 15erms Accurate Threshold,
Caeleste; Vrije Universiteit Brussel


6.7 A 1.32pW/frame∙pixel 1.2V CMOS Energy-Harvesting and Imaging (EHI) APS Imager,
University of Idaho


There is also a forum on "CMOS Image Sensors for 3D Capture" and Jed Hurwitz tutorial on "Layout — The Other Half of Nanometer Analog Design".

Sunday, November 14, 2010

Caeleste and SensL Present SPAD Ranging Array

Cealeste published its recent SPIE presentation on SPAD ranging array with nice explanations of its operation and measurement results. The presentation itself is a continuation of one from a year ago. The summary page shows impressive performance level:

Saturday, November 13, 2010

CCD Romantic Story

Electronics Weekly: David Manners wrote a sentimental story about Kazuo Iwama, Sony's president in 1976-1982.

In 1973 as a deputy president he ordered Sony to start developing CCD for a consumer video camera.

The development was much more difficult than anyone had imagined. The project dragged on, and on. At every board meeting Iwama was asked: "When are you going to get a return on the investment in CCD?" His invariable response was: "Maybe not while I'm alive."

In 1985 Sony produced the first camcorder using a CCD and it was a huge success.

Iwama had died in 1982. In Iwama's granite tombstone there is embedded a CCD.

Sony CCD history is digested here.

Friday, November 12, 2010

Tessera Imaging & Optics Head Replaced

Tessera appointed John Keating as interim EVP of Imaging & Optics, reporting directly to Tessera's Chairman and CEO Henry Nothhaft. Keating has been with Tessera since 2003, and is the company's SVP of Corporate Development. He replaces Michael Bereziuk, who is leaving the company for personal reasons.

Thursday, November 11, 2010

Small Camera Modules Review

OptoIQ published a popular review of camera module technologies: wafer-scale optics integration, multi-layer TSV integration inside module and traditional camera modules.

AWAIBA 250x250-pixel NanEye for medical applications