Wednesday, October 31, 2018

Nano-Antenna Based IR Detectors Review

MDPI publishes Universiti Sains Malaysia, King Saud University, and Qassim University paper "Nano-Antenna Coupled Infrared Detector Design" by Mohamed H. Mubarak, Othman Sidek, Mohamed R. Abdel-Rahman, Mohd Tafir Mustaffa, Ahmad Shukri Mustapa Kamal, and Saad M. Mukras.

"Since the 1940s, infrared (IR) detection and imaging at wavelengths in the two atmospheric windows of 3 to 5 and 8 to 14 μm has been extensively researched. Through several generations, these detectors have undergone considerable developments and have found use in various applications in different fields including military, space science, medicine and engineering. For the most recently proposed generation, these detectors are required to achieve high-speed detection with spectral and polarization selectivity while operating at room temperature. Antenna coupled IR detectors appear to be the most promising candidate to achieve these requirements and has received substantial attention from research in recent years. This paper sets out to present a review of the antenna coupled IR detector family, to explore the main concepts behind the detectors as well as outline their critical and challenging design considerations. In this context, the design of both elements, the antenna and the sensor, will be presented individually followed by the challenging techniques in the impedance matching between both elements. Some hands-on fabrication techniques will then be explored. Finally, a discussion on the coupled IR detector is presented with the aim of providing some useful insights into promising future work."

Arm Develops Always-On Face Unlock

Arm reports that its machine learning-based always-on mobile face unlock achieves over 98% accuracy. It relies quite heavily on a low power low resolution image sensor and a RGB-depth sensor:

e2v Unveils 5MP 2.8um-Pixel GS Sensor

Teledyne e2v announces the expansion of its Snappy family of sensors with a new 5MP device. The Snappy 5M is designed for barcode reading, 2D scanning and similar applications. Available in both monochrome and color, the Snappy 5M has a small 1/1.8-inch format, containing a 2.8μm, low-noise, global shutter pixel. The device can stream video at ~50fps at 10 bits over a 4 wire, MIPI CSI-2 interface.

Snappy 5M is designed to enable fast, extended range scanning and includes unique features and region of interest modes:

  • A Fast Self Exposure (FSE) mode automatically calculates the optimum integration time that is applied to the first image from the device. The mode is user programmable and provides continuous fast decoding, tolerating any kind of lighting or dynamic lighting environment. This is advantageous compared with conventional auto exposure methods, improving convergence speed and robustness.
  • A Smart ROI feature searches for barcodes in the image frame, and reports their locations as metadata in the image footer. The regions of the image containing barcodes are discerned from the background image to considerably reduce downstream image processing (FPGA/CPU/DSP) power, time and cost. Up to 16 different regions can be detected simultaneously. Other forms of repetitive signatures such as printed character strings can also be detected for document scanning and OCR applications.

Omnivision Introduces 16MP Sensor for Mainstream Smartphones

PRNewswire: OmniVision announces OV16E10 — the new generation in its 16MP sensor family built on 1.12um PureCel Plus stacked die technology.

With 16-MP cameras moving into a greater variety of smartphones, consumers at all price points expect their phones to deliver high-quality images and video,” said Xinjing Wang, product marketing manager at OmniVision. “Following the success of the previous-generation OV16B10 in the high-end mobile segment, this new image sensor brings high resolution with the same excellent image quality and features to the mainstream smartphone market.

Integrated into this sensor is a 2x1 Microlens PDAF. The OV16E10 is adapted for dual-camera and tri-camera designs. Its built-in frame rate synchronization supports image fusion in dual- and triple-cameras while simplifying camera system architectures. The OV16E10 also features a gyro interface that reads and synchronizes the motion data from an external gyroscope to enable image stabilization for video and still capture.

The sensor also supports multiple resolution configurations, including 16MP output at 30 fps, 4K2K video at 60fps, 1080p video at 120fps and 720p video at 180fps.

Imec High Speed Imager

MDPI piblishes Imec paper "Analysis and Design of a CMOS Ultra-High-Speed Burst Mode Imager with In-Situ Storage Topology Featuring In-Pixel CDS Amplification" by Linkun Wu, David San Segundo Bello, Philippe Coppejans, Jan Craninckx, Andreas Süss, Maarten Rosmeulen, Piet Wambacq, and Jonathan Borremans.

"This paper presents an in-situ storage topology for ultra-high-speed burst mode imagers, enabling low noise operation while keeping a high frame depth. The proposed pixel architecture contains a 4T pinned photodiode, a correlated double sampling (CDS) amplification stage, and an in-situ memory bank. Focusing on the sampling noise, the system level trade-off of the proposed pixel architecture is discussed, showing its advantages on the noise, power, and scaling capability. Integrated with an AC coupling CDS stage, the amplification is obtained by exploiting the strong capacitance to the voltage relation of a single NMOS transistor. A comprehensive noise model is developed for optimizing the trade-off between the area and noise. As a proof-of-concept, a prototype imager with a 30 µm pixel pitch was fabricated in a CMOS 130 nm technology. A 108-cell memory bank is implemented allowing dense layout and parallel readout. Two types of CDS amplification stages were investigated. Despite the limited memory capacitance of 10 fF/cell, the photon transfer curves of both pixel types were measured over different operation speeds up to 20 Mfps showing a noise performance of 8.4 e−."

Tuesday, October 30, 2018

Nikkei on Sony Image Sensor Production Expansion

Nikkei reports that Sony will spend 600 billion yen ($5.34 billion) over the three years through fiscal 2020. The investment, representing a 30% increase over 460 billion yen spent over the previous three fiscal years, aims to lift production capacity by 20% to 30%.

Sony manufactures image sensors at factories located in Nagasaki, Kumamoto and Yamagata prefectures. The current monthly production capacity for 300 mm wafers is around 100,000.

Sony held a 52% share of the image sensor market in 2017, according to IHS Markit. Second-place Samsung Electronics controlled 19%, with the gap between them widening 8 percentage points from 2016.

Sony Reports Image Sensor Sales Growth, Increases Forecast

Sony presents improved image sensor sales and outlook in its latest quarterly report:

SeekingAlpha: The company's earnings call reveals few more details on the image sensor business:

"The back-illuminated ToF sensor that we started to mass-produce from the second half of this fiscal year has started up smoothly. We expect that demand for image sensors will continue to expand for the next several years as multiple sensor cameras and larger-sized image sensors are adopted in smartphones. In order to proactively adapt to this increased demand, we have brought forward a capital expenditure plan for image sensors in our third mid-range plan and begun to consider increasing the amount.

As a result, we might increase our image sensor capital expenditures by about 20% above our previous plan, which will bring our production capacity to nearly a maximum that can fill in to our existing facilities by fiscal 2020.

...the Semiconductor business is very capital-intensive. So we remain very careful and cautious about proceeding with this. This capacity increase this time around is, as I mentioned, to meet the demand leading to 2020. And I mentioned the production capacity per month of 130k [wafers]. In the existing buildings, we can increase the production up to 136k [wafers]. We are looking into such a expansion opportunity. So using the existing buildings as much possible, why the increase in the production capacity?

...On our side, the imager for smartphones, our margins is about 20% plus.

SPAD Imaging Lecture

Carnegie Mellon University's Matthew O'Toole gives a presentation on SPAD time-resolved imaging:

TowerJazz Talks about Tier 1 DSLR Customer, More

SeekingAlpha: TowerJazz Q3 earnings call has few updates on the company's image sensor business:

"In the image sensor area, I personally met with three very exciting customers in the past quarter. Capitalized by our 300 millimeter Japanese factory, we won the leading sole position as a foundry supplier with a multi-generation roadmap for a Tier 1 DSLR customer, the high level and working level meeting, reconfirmed both the appreciation for the joint activity milestone achievements, and the timeline for release of our first and second generation products.

Another meeting was with an extremely exciting company in the area of augmented reality resulting in a reconfirmation of our wind for the next generation time of flight product for our next-generation or per time-of-flight product for the next-generation product, expecting high volume release in 2020 and continued further detailed operational and technical program alignments for subsequent generations.

Lastly, I met with one of the top revenue leaders in the image sensor market with whom he presently have no business with strong discussions and activities, including wafers that will shortly be shipped from a 300 millimeter line with our most advanced processes. We are encouraged and confident to gain market share with this customer in this area.

Our image sensor management, the General Manager, Dr. Avi Strum with additional leading technical experts, held a prestigious seminar in San Francisco this past month. This unique CIS technical seminar showcased our advanced platform and cutting edge technical solutions and advanced manufacturing capabilities for a wide range of imaging growing markets.

28 handpicked sensor companies attended with TowerJazz providing detailed technical reviews, analysis and notes on the customer and industry benefits for our customized pixel solutions for global shutter, near infrared enhanced pixels, Backside Illumination and stacked wafer, 3D sensing and LiDAR among others. On November 5, we will hold additional seminar addressing the European sensor market.

...we'll be having continued ramp within large area image sensors and that has already taken off. That's where we're serving studio and medical applications right now coming into the end of 2019, 2020 timeframe, where we'll be wrapping the DSLR first generation products of a very, very big customer.

We have many projects for high-end image sensor running right now, in Uozu. It is a well known that especially the high-end image sensor project takes time to ramp. But once they are ramping, the longevity of such projects is about between eight to 12 years. We have started this activity, for example, we said DSLR Tier 1 customer a year-ago, but it takes time for that product to get into production. Once it's in production, it would keep for many, many years.

Samsung Announces 48MP 0.8um Pixel Sensor

BusinessWire: Samsung introduces two 0.8 μm pixel sensors – the 48MP ISOCELL Bright GM1 and the 32MP ISOCELL Bright GD1.

Demand for ultra-small, high-resolution image sensors are growing as smartphones evolve to deliver new and more exciting camera experiences for users,” said Ben K. Hur, VP of System LSI marketing at Samsung Electronics. “With the introduction of our cutting-edge 0.8μm-pixel Samsung ISOCELL Bright GM1 and GD1 image sensors, we are committed to continue driving innovation in image sensor technologies.

The GM1 and the GD1 sensors are based on the company’s latest pixel isolation technology ISOCELL Plus. In addition, Tetracell technology, where four pixels are merged to work as one to increase light sensitivity, the GM1 and GD1 can deliver light sensitivity equivalent to that of a 1.6μm-pixel image sensor at 12MP and 8MP resolution, respectively. The sensors also support Gyro-based electronic EIS for fast and accurate image capture. A real-time HDR feature is added to the GD1.

The Samsung ISOCELL Bright GM1 and GD1 are expected to be in mass production in Q4 this year.

Monday, October 29, 2018

Panasonic to Start Production of Camera with Organic Image Sensor

Panasonic announces a 8K Multipurpose Camera AK-SHB 810 based on the world's first organic film image sensor presented at the number of conferences and in papers. The mass production of the new camera is expected to start at the Fall of 2019. "8K / 4K / HD video output is possible from the camera unit and the image processing unit AK-SHU 810 connected with optical fiber cable."

GPixel Expands its Global Shutter Sensor Range

Gpixel expands of its GMAX product family with the addition of two new 2.5 µm GS sensors for c-mount camera. The GMAX2505 is a 5MP, 2600 x 2160 resolution 1/2” sensor, while the GMAX2509 is a 9MP, 4200 x 2160 resolution 2/3” sensor. The new devices follow the 25MP, 1.1” GMAX0505 introduced earlier this year. Gpixel’s GMAX c-mount line up are all pin compatible and use the same pixel based on the world smallest 2.5 µm charge domain global shutter technology.

The sensors delivers a read noise of less than 1.5 e- RMS and DR of more than 70 dB. They also include advanced in-pixel light pipe technology for a shutter efficiency of 1/10000 and improved angular response. Offering full resolution frame rates of up to 292 fps at 10-bit output and 121 fps at 12-bit, both the GMAX2505 and GMAX2509 can support the 12-bit frame rate with half the LVDS data output pairs providing added flexibility in the choice of system FPGA. Housed in a compact 22 x 17.5 mm LGA package where the optical center matches the mechanical package center, makes them a perfect fit for 29 mm mini cube cameras.

We are thrilled with the significant expansion of our advanced global shutter c-mount sensor line up. It is a result of our ambitious plans to be a leading provider of high-quality industrial imaging solutions,” commented Wim Wuyts, Chief Commercial Officer of Gpixel. “Designed with leading camera manufacture’s input, the range matches our customers’ needs for flexible operational strategies that enables quick adjustments for different business and applications as well as savings on overall R&D implementation costs.

Mono sensor samples and an evaluation board will be commercially available from March 2019, with volume production planned for October 2019.

ON Semi Unveils 50MP CCD Targeting Inspection of Smartphone Displays

BusinessWire: ON Semi is aiming to inspection of smartphone displays with the introduction of a new 50 MP CCD. Said to be the highest resolution Interline Transfer CCD image sensor commercially available, the KAI-50140 provides the detail and high uniformity for inspection of smartphone displays, circuit board and mechanical assembly inspection as well as aerial surveillance.

The KAI-50140 is designed in a 2.18 to 1 aspect ratio to match the format of modern smartphones, reducing the number of image captures required to inspect a full display. The 4.5 µm Interline Transfer CCD pixel used in the KAI-50140 provides high resolution with a true electronic global shutter while preserving critical imaging performance. The new device supports frame rates up to 4 fps through the use of flexible 1, 2, or 4 output readout architecture, and shares the same pin definitions as both the popular 29 MP KAI-29050 and KAI-29052 as well as the KAI-43140 image sensors. This allows the KAI-50140 to be incorporated into existing camera designs with only minor electrical changes – speeding time-to-market for adoption of the new device.

As the specifications and aspect ratios of smartphone displays continue to expand in both physical size as well as pixel density, we were compelled to develop an image sensor designed specifically for inspection of these devices,” said Herb Erhardt, VP and GM, Industrial Solutions Division, Intelligent Sensing Group at ON Semiconductor. “The unique combination of high resolution, unmatched image uniformity, and 2.18:1 pixel arrangement provided by the KAI-50140 enables higher quality, more efficient inspection of modern smartphone displays.

The KAI-50140 is sampling today with production planned by the end of 2018.

Omnivision Announces 24MP Sensors with 2nd Generation 0.9um Pixels

PRNewswire: OmniVision announced the second generation of its 0.9um pixels used in 1/2.83-inch 24MP sensors, built on the PureCel Plus stacked die technology. The OV24B sensor family is aimed to front- and rear-facing smartphone cameras and available in four versions:

  • OV24B1Q: 4-cell Bayer imaging; in-pixel binning for optimal sensitivity and low-light performance; and pad locations on the top and bottom of the chip to reduce module size in the x-direction, for front-facing cameras in thin-bezel infinity display phones
  • OV24B2Q: 4-cell Bayer imaging; in-pixel binning; type-2, 2x2 microlens PDAF, for main cameras due to video and low-light performance
  • OV24B1B: monochrome imaging; improved low-light sensitivity; for rear-facing multi cameras for Bokeh and zoom
  • OV24B10: standard Bayer imaging; for main cameras due to high 24MP resolution, as a single camera or as part of a multi-camera arrangement for Bokeh and zoom

As sub-micron pixel technology continues to mature in tandem with evolving performance expectations, it is enabling extremely high-resolution imaging and multi-camera designs, which continue to be growing market trends. These trends apply to both front-facing and rear-facing smartphone cameras,” said James Liu, product marketing manager at OmniVision. “The new OV24B sensor family, with its on-chip re-mosaic, 2x2 ML PDAF, and high-speed video capability, can address the demands of these growing market trends, enabling mobile cameras to provide the very best user experience for autofocus, low-light performance, and high-resolution video and still image capture.

All four versions can output high-resolution streaming video in a variety of formats, including 24MP at 30 fps, 4-cell binning 6MP at 60fps, 4K2K video at 60fps, 1080p video at 120fps, or 720p video at 240fps. The OV24B is in production now.

Sunday, October 28, 2018

MIPI CSI-2 Past, Present, and Near Future

Synopsys presentation at MIP DEvcon in Seoul has a couple of slides summarizing the past progress and the future plans, including recently announced A-PHY (automotive PHY):

It appears that Sony multi-camera D-PHY solution is being considered for future standardization:

Saturday, October 27, 2018

Panasonic Flash LiDAR Paper

MDPI Special Issue on The International SPAD Sensor Workshop publishes Panasonic paper "A 250 m Direct Time-of-Flight Ranging System Based on a Synthesis of Sub-Ranging Images and a Vertical Avalanche Photo-Diodes (VAPD) CMOS Image Sensor" by Yutaka Hirose, Shinzo Koyama, Motonori Ishii, Shigeru Saitou, Masato Takemoto, Yugo Nose, Akito Inoue, Yusuke Sakata, Yuki Sugiura, Tatsuya Kabe, Manabu Usuda, Shigetaka Kasuga, Mitsuyoshi Mori, Akihiro Odagawa, and Tsuyoshi Tanaka.

"We have developed a direct time-of-flight (TOF) 250 m ranging Complementary Metal Oxide Semiconductor (CMOS) image sensor (CIS) based on a 688 × 384 pixels array of vertical avalanche photodiodes (VAPD). Each pixel of the CIS comprises VAPD with a standard four transistor pixel circuit equipped with an analogue capacitor to accumulate or count avalanche pulses. High power near infrared (NIR) short (less than 50 ns) and repetitive (6 kHz) laser pulses are illuminated through a diffusing optics. By globally gating the VAPD, each pulse is counted in the in-pixel counter enabling extraction of sub-photon level signal. Depth map imaging with a 10 cm lateral resolution is realized from 1 m to 250 m range by synthesizing subranges images of photon counts. Advantages and limitation of an in-pixel circuit are described. The developed CIS is expected to supersede insufficient resolution of the conventional light detection and ranging (LiDAR) systems and the short range of indirect CIS TOF."

Friday, October 26, 2018

Automotive News

FLIR talks about "educating" the automotive makers on LWIR night vision importance:

Vioneer, a Atolive, Sweden spin-off, talks about night vision and LiDAR in cars:

BusinessWire: LiDAR developer Innovusion has raised $30m in Series A funding. On the heels of the funding, the company’s LiDAR system, the Innovusion Cheetah, is now available for ordering by customers worldwide. Innovusion Cheetah is said to be the world’s first image-grade LiDAR system to support level 4+ (fully autonomous under certain operating conditions) autonomous driving solutions.

The Cheetah LiDAR is said to be able to detect objects more than 200 meters away and generates an dense point cloud (300 lines), more than double of that of other providers in the space, while maintaining a high frame rate and achieving low-cost in high-volume production. Innovusion is ramping production to meet worldwide demand starting in Q4 2018. Innovusion was founded in November 2016 and began shipping samples of Innovusion Cheetah LiDAR in the second quarter of 2018.

Socionext shows its Omniview system that can change a virtual view point based on the car multiple cameras:

Nikkei: New Lexus ES is said to become the first production car with digital side mirrors:

Venturebeat: Eyesight raises $15m for global expansion of its in-cabin sensing technologies:

Thursday, October 25, 2018

iniVation Ultra-Low-Power 65 nm DVS with Built-in Noise Filter

iniVation, a spin-off of the University of Zurich and the ETH Zurich, announces the successful development of next-generation Dynamic Vision Sensor technology. Highlights of the design include:
  • Event-based architecture enables ultra-high speed (equivalent frame rate beyond 10 kfps)
  • Pixel-parallel noise and spatial redundancy suppression
  • Ultra-low power, 18 nW per pixel, 26 pJ per event @ 1.2 V
  • Bandwidth up to 180 M events per second
  • Dynamic range beyond 100 dB
  • Compact 10um pixel design, fabricated with 65 nm process

Lucid Vision Labs Demos Sony BSI ToF Sensor-based Camera

Lucid Vision Labs announces Helios ToF 3D camera. The camera has four 850nm VCSEL laser diodes and integrates Sony’s new DepthSense IMX556PLR BSI ToF sensor with 10μm pixel and high modulation contrast ratio. The camera can produce depth data at 60 fps with 640×480 resolution.

Our new Helios ToF 3D camera shows the potential for Time of Flight technology in a variety of industrial applications, including robotic navigation, 3D inspection and logistic automation such as bin-picking and package dimensioning,” says Rod Barman, Founder and President at LUCID Vision Labs. “Improved depth accuracy and precision at an attractive price point will enable many 3D applications to run more efficiently while reducing overall system cost.” The first models of the Helios camera will be available by middle of 2019."

Imaging Becomes ST Major Growth Engine

ST Q3 2018 earnings call depicts a dramatic growth in imaging business:

"Third quarter net revenues increased 18% year-over-year, on strong growth for Imaging, Power Discrete and Automotive products. On a sequential basis, ST's revenues increased 11.2%, 120 basis points above the midpoint of our guidance, on higher than expected sales of Imaging product.

During Q3, AMS revenues reached $899 million, an increase year-over-year of 36.7% on triple-digit growth in Imaging and growth in Analog and MEMS.

Now, let's move to our Q4 outlook. The key drivers of our sales growth sequentially will continue to come from Imaging, Automotive and Power Discrete... What we see moving from Q3 to Q4, the drivers in the revenue will be definitely our image sensor together with the Automotive and Power Discrete... Definitely, Imaging is driving our revenue.

Wednesday, October 24, 2018

PCO sCMOS Cameras e-Book

PCO publishes a nice e-book "Everything You Always Wanted to Konow about sCMOS Cameras, but Were Afraid to Ask." Few interesting tables from the book:

Tobii Next Gen Eye Tracker Relies on Custom Imager

Tobii next generation eye-tracking platform IS5 relies on custom-designed image sensor:

"The brand new Tobii EyeSensor, a custom CMOS image sensor specifically designed and optimized for eye tracking. The EyeSensor combines carefully tuned state of the art pixel technology with advanced logic, enabling the IS5 platform to provide significantly enhanced eye tracking performance compared to off-the-shelf image sensors."

Mediatek Helio P70 Promises Lower "mA per second" Metrics

PRNewswire: MediaTek Helio P70 application processor promises lower power in many camera functions:

"For photography, its new high-resolution depth engine gives a 3X boost in depth-mapping performance, which enables smoother 24fps visual bokeh previews while also saving 43mA per second versus competitor alternatives.

With HDR-capable displays and photography a key trend for the latest smartphones, the P70 is unique among peers in its focus on real-time HDR capture, recording and processing and its hardware-driven ability to process RAW HDR images. Other HDR-centric enhancements includes RAW-domain multi-frame HDR capture and Zig-Zag HDR.

MFNR(MFLL) performance has been improved by 20%, meaning faster shot-to-JPEG processing so you’ll never experience a stutter as you capture a burst of photos. Further hardware enhancements include a hardware warping engine used for electronic image stabilization (EIS) that saves 23mA per second vs GPU; an anti-blooming engine prevents white-out; and accurate AI facial detection with intelligent scene detection that generates better 3A (AE, AF and AWB).

Helio P70 supports a variety of AI-infused photo and video experiences including real-time beautification, scene detection and artificial reality (AR) capabilities. The chipset features improved deep-learning facial detection with up to 90 percent accuracy. With support for a super-sized 32MP single camera, or 24+16MP dual cameras, the chipset gives device makers more design flexibility. Its three ISPs are tuned to provide significant power savings, reducing consumption by 18 percent compared to previous Helio generation dual-camera setups.