Archives for December 2015

Image Sensors World        Go to the original article...

Brigates appears to abandon all its markets, but surveillance and security, as its recent product list suggests:

Go to the original article...

Image Sensors World        Go to the original article...

ZTE Mobile World Magazine features Axon dual camera smartphone, starting page 16:

AXON has a 13-megapixel camera on the back and a secondary 2-megapixel camera that sits above it to help capture 3D depth for a refocusable photo. To get a better idea of the story behind the challenging work, Mobile World talked with two engineers from ZTE’s camera lab.

Interviewee: Zhu Yufei, responsible for the North American camera team and camera solution planning

Why did you decide to introduce a dual lens camera into AXON? How will the smartphone camera develop in the future?

Dual-lens camera technology is prospective. Photos taken by a single-lens camera are some kind of object simulation, while photos taken by a dual lens camera are more vivid because dual lens cameras capture more information. ...In terms of intelligence, we improve image stabilization and low-light photography through better algorithms and relevant technologies on AXON.

Are there any differences between ZTE’s dual-lens camera and other similar products?

The distance between the two lenses of AXON is wider. AXON outperforms other dual-lens smartphones in the large aperture effect, depth calculation accuracy, and distance of objects from the camera.

What are the trends for future smartphone cameras?

Both single-lens cameras and dual-lens cameras will coexist. Single-lens cameras will dominate in the near future, and dual lens cameras will continue to improve.

Interviewee: Xiao Longan, responsible for the camera software development

What are the tough nuts to crack in introducing the dual-lens camera to AXON? How did you overcome these difficulties?

...No smartphone project at ZTE has required so many human resources on the R&D of the camera app, and no camera app has received so much attention.

Which camera feature do you like most?

Background blur and refocus...

Can you explain more about how the refocus feature works?

The dual-lens camera captures and adds depth data to each object in the image. These data allows you to refocus a picture and add a blur effect to select parts of the image.

Go to the original article...

Image Sensors World        Go to the original article...

Korea-based KB Securities analysts come up with a report on camera module market. Few quotes, in a somewhat broken Google translation:

"The camera module shipment growth to 2019 is expected to reach four times the mobile device shipment growth...

The premium smartphone dual (Dual) and 3D, pixels in the low-end phones and increased competition expected with OIS...

Dual Camera module adopted is because it is expected to accelerate in 2016 from this...

Front high pixel camera module, dual camera module, OIS rate mounted on the side as expected the speedy pace..."

Mobile devices and camera module shipment forecast

Pixel mobile camera modules share outlook

Go to the original article...

Image Sensors World        Go to the original article...

Changchun, China-based Gpixel has expanded its image sensor portfolio, now offering 9 standard products for machine vision and industrial applications:

Go to the original article...

Image Sensors World        Go to the original article...

Nikkei reports that Sony announces a structural reform and personnel changes that will take effect on Jan 1, 2015 (probably a typo, should be 2016). Sony Device Solution Business Group will be split in 3 organizations: "Automotive Division," "Module Division" and "Product Development Division."

Nikkei sees several reasons for the re-org:

• The reorg will strengthen Sony image sensor business.
• Sony can better address the fast growing automotive image sensor market where On Semi currently has the largest share.
• By offering modules rather than image sensors, it becomes possible to better support customers that do not have capabilities to use tune and adapt image quality.

The other reasons for shifting focus from image sensors to camera modules might be an attempt to replicate Sharp camera module success, and an intention to capture a larger chunk of the imaging foodchain.

Go to the original article...

Image Sensors World        Go to the original article...

ST publishes a promotional video on its automotive image sensors and image processors solutions:



Update: Upon a closer look, VG6640 image sensor and STV0991 image processor appear to be the new, previously unannounced devices.

Go to the original article...

Image Sensors World        Go to the original article...

SystemPlus publishes reverse engineering of Fujitsu Iris Authentication Module IR Camera Module & IR LED. The module has been extracted from Fujitsu Arrows NX F-04G smartphone that uses iris scan as the next biometric login technology. Compared to fingerprint sensors, Fujitsu claims that the solution features a faster, safer and more secure authentication. It is also a cost effective solution due to the reuse of standard CIS and LED components. OSRAM is said to be the IR LED manufacturer and has designed this 810nm LED exclusively for this iris scan application (this sounds somewhat contradictory to the re-use statement).

Go to the original article...

Image Sensors World        Go to the original article...

GlobeNewsWire: Grand View Research estimates the global LiDAR market size at over $260M in 2014. The market is expected to grow to $944.3M by 2022 driven by improved automated processing ability in data processing and image resolution capabilities.

Go to the original article...

Image Sensors World        Go to the original article...

Nikkei publishes an article on Olympus IEDM 2015 presentation on stacked visible and IR sensor. While it mostly re-iterated the previously published data, some info is new:

"The laminated image sensor is made by combining (1) an image sensor equipped with an RGB color filter for visible light (top layer) and (2) a near-infrared image sensor (bottom layer). Each layer functions as an independent sensor and independently outputs video signals.

The visible-light image sensor is a backside-illuminated type, and its light-receiving layer (made of semiconductor) is as thin as 3μm. Each pixel measures 3.8 x 3.8μm, and the number of pixels is 4,224 x 240."

Go to the original article...

Image Sensors World        Go to the original article...

CBS publishes an interview with Apple management team, that includes a part about Apple camera team and its head Graham Townsend:

One of the most complex engineering challenges at Apple involves the iPhone camera, the most used feature of any Apple product. That's the entire camera you're looking at in my hand.

Graham Townsend: There's over 200 separate individual parts in this-- in that one module there.

Graham Townsend is in charge of a team of 800 engineers and other specialists dedicated solely to the camera. He showed us a micro suspension system that steadies the camera when your hand shakes.

Graham Townsend: This whole sus-- autofocus motor here is suspended on four wires. And you'll see them coming in. And here we are. Four-- These are 40-micron wires, less than half a human hair's width. And that holds that whole suspension and moves it in X and Y. So that allows us to stabilize for the hand shake.

In the camera lab, engineers calibrate the camera to perform in any type of lighting.

Graham Townsend: Go to bright bright noon. And there you go. Sunset now. There you go. So, there's very different types of quality of lighting, from a morning, bright sunshine, for instance, the noonday light. And then finally maybe--

CBS: Sunset, dinner--

Graham Townsend: We can simulate all those here. Believe it or not, to capture one image, 24 billion operations go on.

CBS: Twenty-four billion operations going on--

Graham Townsend: Just for one picture--

Go to the original article...

Image Sensors World        Go to the original article...

Filmmaker IQ publishes a short video lecture on image sensors. While not perfectly accurate in some parts, it's amazing how much info one can squeeze in a 13 min video:

Go to the original article...

Image Sensors World        Go to the original article...

Yole publishes an interview with a Heimann Sensor CEO Joerg Schieferdecker. Heimann Sensor is a 14-year old German IR image sensor manufacturer. Schieferdecker talks about a number of emerging thermal IR imaging applications, such a contactless thermometer embedded into a mobile phone:

"We are working to develop such a spot thermometer within Heimann Sensor. Many players in the mobile phone business are looking at these devices but we think that two more years of development are needed in order to get them working. First, single pixel sensors will be used and then arrays will be introduced into the market. The remaining issues are significant. For example, getting a compact device, within a thin ‘z budget’, is very challenging and needs further development at device and packaging level. In the same way, heat shock resistance is a challenge at the moment.

The applications of remote temperature sensing are just huge. Fever measurement, checking outdoor temperatures from your mobile phone, water temperature measurement for a baby’s bath or bottle - the list is endless. Most require 1 or 2°C accuracy, which is already available, while body temperature needs further development to reach 0.2°C accuracy.

We expect to see the first mobile phone with this feature around Christmas 2017."

Go to the original article...

Image Sensors World        Go to the original article...

BDTI publishes an article on the next generation Cadence Tensilica Vision P5 Processor IP featuring lower power and higher speed:


The base P5 core is said to be less than 2 mm2 in size in a 16nm process, with minor additional area increments for the optional FPU and cache and instruction memories. Lead customers have been evaluating and designing in Vision P5 for several months, and the core is now available for general licensing.

Go to the original article...

Image Sensors World        Go to the original article...

Belgian local DSP Valley Newsletter publishes an article about Caeleste (see page 6). Few quotes:

"Caeleste was created in December 2006, by a few people around the former CTO of Fillfactory/Cypress. After an organic growth start, Caeleste is now 22 persons large and fast growing with a CAGR of 30-50%. It is specialized in the design and supply of custom specific CMOS Image Sensors (CIS). Caeleste is still 100% owned by its founders.

...In fact, Caeleste has almost the largest CIS design team in Europe, exclusively devoted to customer specific CMOS image sensors."

Some of Caeleste R&D projects:

• Caeleste has the world record in low noise imaging: We have proven a noise limit of 0.34 e-rms; this implies that under low flux conditions (e.g. in astronomy) each and every photon can be counted and that the background is completely black, provided that the dark current is low enough.
• Caeleste has also designed a dual color X-ray imager, which allows a much better discrimination and diagnosis of cancerous tissue than the conventional grey scale images.
• Caeleste has also its own patents for 3D imaging, based on Time of Flight (ToF) operation. This structure allows the almost noise free accumulation of multiple laser pulses to enable accurate distance measurements at long distance or with weak laser sources.

Go to the original article...

Image Sensors World        Go to the original article...

Oppenheimer publishes December 14, 2015 report on TowerJazz business talking about the CIS part of it, among many other things:

"CMOS Image Sensors (CIS). Approximately 15% of sales, ex-Panasonic. Tower offers advanced CMOS image sensor technology for use in the automotive, industrial, medical, consumer, and high-end photography markets. Tower estimates the silicon portion of the CMOS image sensor market to be a ~$10B TAM, with ~$3B potentially being served via foundry offerings. Today, roughly two-thirds of this market serves the cellular/smartphone camera market; however, the migration to image-based communication across the automotive, industrial, security and IoT markets is expanding the applications for Tower's CIS offerings. Tower is tracking toward 35% Y/Y growth in this segment in 2015, outpacing the industry's 9% compounded growth rate, as the company has gained traction in areas previously not served by specialty foundries.

Tower has IP related to highly customized pixels, which lend the technology to a wide variety of applications. We see opportunities in automotive, security/surveillance, medical imaging, and 3D gesture control driving sustainable growth in this segment long term."

Oppenheimer: "We expect image sensor growth to outpace the overall optoelectronics segment, and forecast growth at a 6% CAGR from 2014 to 2019 to approximately $13.3B, from $10.4B in 2014"

Go to the original article...

Image Sensors World        Go to the original article...

Bloomberg publishes an article about famous 24-year old hacker George Hotz making a self-driving car in his garage based on 6 cameras and a LIDAR. Hotz says that his technology is much better than one of Mobileye used in Tesla Model S: “It’s absurd,” Hotz says of Mobileye. “They’re a company that’s behind the times, and they have not caught up.”

Mobileye spokesman Yonah Lloyd denies that the company’s technology is outdated, “Our code is based on the latest and modern AI techniques using end-to-end deep network algorithms for sensing and control.”

SeekingAlpha reports that Tesla comes to Mobileye's rescue. "We think it is extremely unlikely that a single person or even a small company that lacks extensive engineering validation capability will be able to produce an autonomous driving system that can be deployed to production vehicles," says Tesla. "[Such a system] may work as a limited demo on a known stretch of road -- Tesla had such a system two years ago -- but then requires enormous resources to debug over millions of miles of widely differing roads. This is the true problem of autonomy: getting a machine learning system to be 99% correct is relatively easy, but getting it to be 99.9999% correct, which is where it ultimately needs to be, is vastly more difficult. Going forward, we will continue to use the most advanced component technologies, such as Mobileye’s vision chip, in our vehicles. Their part is the best in the world at what it does and that is why we use it."

Go to the original article...

Image Sensors World        Go to the original article...

Infineon and pmd unveil their latest 3D image sensor chip of the REAL3 family. Compared to the previous version, the optical sensitivity as well as the power consumption of the new 3D sensors has been improved and the built-in electronics take up little space. The chips make it possible for cell phones to operate mini-camera systems that can measure 3D depth data.

The optical pixel sensitivity of the new 3D image sensors is now double that of the previous version. This means that their measurement quality is just as good, while at the same time working with only half of the emitted light output and the camera’s system power consumption is almost halved. The improved optical pixel sensitivity is the result of applying one microlens to each of the pixels of the 3D image sensor chip.

The new 3D image sensor chips were specifically designed for mobile devices, where most applications only need a resolution of 38,000 pixels. The previous 100,000-pixel matrix was accordingly scaled down, and other functional blocks, such as the analog/digital converter for the chip area and performance range were optimized. Thus, the system cost is lower: the sensor chip area is almost halved, and, because of the lower resolution, smaller and less expensive optical lenses can be used.

The three new REAL3 3D image sensors differ in their resolutions: the IRS1125C works with 352 x 288 pixels, the IRS1645C with 224 x 172 pixels and the IRS1615C with 160 x 120 pixels. In this respect, the IRS1645C and IRS1615C are produced on half the chip area of the IRS1125C.

Google’s “Project Tango” using Infineon's IRS1645C 3D image sensor chip. The complete 3D camera for Google “Tango” – consisting of IRS1645C and an active infrared laser illumination – is housed in an area of approximately 10 mm x 20 mm. With a range up to 4 meters (13 feet), a measuring accuracy of 1% of the distance and a frame rate of 5 fps, the 3D camera subsystem consumes less than 300 mW in active mode.

The IRS1125C will be available in volume in Q1 2016. The start of production for the smaller IRS1645C and IRS1615C is planned for Q2 2016. All three types are exclusively delivered as a bare die to allow maximum design flexibility while minimizing system costs.

Go to the original article...

Image Sensors World        Go to the original article...

Adimec publishes a nice demo of smear in CCDs and virtual absence of it in CMOS sensors. The smear manifests itself in:

• Contrast reduction of the image
• Questionable patterns depending on image contents
• Higher noise level in the dark areas of the image
• The three effects above combined: degradation of DRI (detection, recognition, identification)
• For color: a shift of white balance, depending on the smear level

Go to the original article...

Image Sensors World        Go to the original article...

IS&T Electronic Imaging (EI) Symposium publishes a number of interesting presentations, such as "A novel optical design for light field acquisition using camera array:"



"3D UHDTV contents production with 2/3 inch sensor cameras:"

Go to the original article...

Image Sensors World        Go to the original article...

TechInsights announces a reverse engineering report of 12MP Sony sensor from Apple iPhone 6s and 6s Plus, to be published on Dec 22, 2015:

"The Sony 12 megapixel CMOS image sensor is fabricated using a 4 metal (Cu), 90 nm CMOS image sensor process. The backside illuminated die is wafer bonded to an underling control ASIC using through silicon vias (TSVs) to provide die-to-die interconnections. The image sensor features a 1.22 µm large pixel with RGB color filters and a single layer of microlenses. Four pixels share a common readout circuit comprising four transfer gates, a reset gate, select gate and source follower in a 1.75T architecture."

Go to the original article...

Image Sensors World        Go to the original article...

Brightway publishes a promo video of its gated imaging ADAS camera:

Go to the original article...

Image Sensors World        Go to the original article...

2017 International Image Sensor Workshop (IISW) announces its time and location. The workshop is to be held on May 30th – June 2nd, 2017 in Hiroshima, Japan. The Grand Prince Hotel Hiroshima is located on a small island in south Hiroshima with a nice archipelago view from the hotel. Miyajima and the downtown Hiroshima, which are UNESCO world heritages, are within the reach and are good excursion candidates.

Go to the original article...

Image Sensors World        Go to the original article...

TI releases its first SoC for 3D ToF imaging integrating ToF Sensor, AFE, Timing Generator, and Depth Engine on a single chip. While the OPT8320 chip is based on 30um sized Softkinetic pixel technology, it's made by TI and most of the IP is developed in-house in TI. TI publishes a detailed datasheet with example applications, and compares it with a older non-integrated ToF imager:

Go to the original article...

Image Sensors World        Go to the original article...

JBHNews quotes a rumor that Microsoft Lumia 1050 smartphone, to be released in January or February 2016, would feature 50MP camera having a variable aperture, a zoom lens, an OIS and a laser AF. "Overall, Lumia 1050 would be a defining phone in the history of Microsoft."

Go to the original article...

Image Sensors World        Go to the original article...

Optics.org: Researchers from MIT media lab (Prof. Ramesh Raskar) claim that exploiting the polarization of light allows them increasing the resolution of conventional 3-D imaging devices by as much as 1,000 times. The new 3D imaging system, which they call Polarized 3D, in a paper they are presenting at the International Conference on Computer Vision in Santiago, Chile, on December 11-18, 2015:


"Today, 3-D cameras can be miniaturized to fit on cellphones,” said Achuta Kadambi, a PhD student in the MIT Media Lab and one of the system’s developers. “But this involves making compromises to the 3-D sensing, leading to very coarse recovery of observed geometry. That is a natural application for polarization, because you can still use a low-quality sensor, but by adding a polarizing filter gives a result quality that is better than from many machine-shop laser scanners.”

Polarization affects the way in which light bounces off of physical objects. Light within a certain range of polarizations is more likely to be reflected. So the polarization of reflected light carries information about the geometry of the objects it has struck. Microsoft Kinect is used for a coarse depth estimation, while polarization helps to interpolate and improve its depth accuracy.

“The work fuses two 3-D sensing principles, each having pros and cons,” commented Yoav Schechner, an associate professor of electrical engineering at Technion — Israel Institute of Technology. “One principle provides the range for each scene pixel: This is the state of the art of most 3-D imaging systems. The second principle does not provide range. On the other hand, it derives the object slope, locally. In other words, per scene pixel, it tells how flat or oblique the object is.”

Thanks to AG for the link!

Go to the original article...

Image Sensors World        Go to the original article...

Yole Developpement research "Sensors and Data Management for Autonomous Vehicles report 2015" shows the roadmap from manned driving to "feet off" to "hands off" to "mind off" and the sensors needed for that:


"Currently, the most advanced commercial car with autonomous features embeds in the order of 17 sensors. Yole anticipates 29+ sensors by 2030. Today, two sensor businesses dominate: ultrasonic sensors and cameras for surround. Together they are worth a $2.4B market value, but ultrasonic sensors comprise 85% of current market volume. By 2030, cameras for surround will lead the market at $12B, followed by ultrasound sensors and long-range radar ($8.7B and $7.9B respectively), and then short-range radar ($5B). Overall 2030 market value is estimated at $36B.

All of these evolutions will likely lead to mass-market adoption of semi-autonomous vehicles by 2030. We expect more than 22M units (18% of all vehicles sold) of level-2 vehicles and 10M units of level-3 vehicles by 2030, along with 38M units of level-2 vehicles and 1M units of Level-4 vehicles."

Go to the original article...

Image Sensors World        Go to the original article...

Odos Imaging CEO Chris Yates publishes an article in Vision Systems Design showing its high speed camera capabilities in shooting a bullet:

Camera measures the bullet deceleration due to the air drag

Go to the original article...