Autosens publishes a presentation "LIDAR systems for automotive: Benefits and the challenges for OEMs" by Jaguar Landrover's Andy Lewin:
Saturday, March 03, 2018
Friday, March 02, 2018
Noise in Charge Domain Sampling Readouts
MDPI Special Issue on the 2017 International Image Sensor Workshop publishes Delft University paper "Temporal Noise Analysis of Charge-Domain Sampling Readout Circuits for CMOS Image Sensors" by Xiaoliang Ge and Albert J. P. Theuwissen.
"In order to address the trade-off between the low input-referred noise and high dynamic range, a Gm-cell-based pixel together with a charge-domain correlated-double sampling (CDS) technique has been proposed to provide a way to efficiently embed a tunable conversion gain along the read-out path. Such readout topology, however, operates in a non-stationery large-signal behavior, and the statistical properties of its temporal noise are a function of time. Conventional noise analysis methods for CMOS image sensors are based on steady-state signal models, and therefore cannot be readily applied for Gm-cell-based pixels. In this paper, we develop analysis models for both thermal noise and flicker noise in Gm-cell-based pixels by employing the time-domain linear analysis approach and the non-stationary noise analysis theory, which help to quantitatively evaluate the temporal noise characteristic of Gm-cell-based pixels. Both models were numerically computed in MATLAB using design parameters of a prototype chip, and compared with both simulation and experimental results. The good agreement between the theoretical and measurement results verifies the effectiveness of the proposed noise analysis models."
"In order to address the trade-off between the low input-referred noise and high dynamic range, a Gm-cell-based pixel together with a charge-domain correlated-double sampling (CDS) technique has been proposed to provide a way to efficiently embed a tunable conversion gain along the read-out path. Such readout topology, however, operates in a non-stationery large-signal behavior, and the statistical properties of its temporal noise are a function of time. Conventional noise analysis methods for CMOS image sensors are based on steady-state signal models, and therefore cannot be readily applied for Gm-cell-based pixels. In this paper, we develop analysis models for both thermal noise and flicker noise in Gm-cell-based pixels by employing the time-domain linear analysis approach and the non-stationary noise analysis theory, which help to quantitatively evaluate the temporal noise characteristic of Gm-cell-based pixels. Both models were numerically computed in MATLAB using design parameters of a prototype chip, and compared with both simulation and experimental results. The good agreement between the theoretical and measurement results verifies the effectiveness of the proposed noise analysis models."
Espros Expands BSI Manufacturing Capacity
Espros February newsletter (not on the web site yet) explains the company plans to expand its BSI production line:
"One of the most important factors is time to market. The semiconductor industry is in a very difficult position in this respect because the supply chain and thus the lead time is quite long.
Typically, more than 20 weeks from the start of wafer processing to final chips is not abnormal. Thus, if a shortage in the supply chain occurs, e.g. due to increasing demand, allocation for 30-50 weeks is not really special. In the case of our products where we do rather complex backside processing, we had to add another 11 weeks to the supply lead time one year ago. This situation was not satisfying at all and we decided to invest heavily into additional in-house capabilities.
The result is now that we can do the backside and the back-end processing of the wafers within days, if needed. We showed this performance in a recent project where lead time from wafer production start until the chips were in the lab was only nine weeks. Not just wafer processing and backside processing, but also assembly including AR coating and band-pass filter bonding."
"One of the most important factors is time to market. The semiconductor industry is in a very difficult position in this respect because the supply chain and thus the lead time is quite long.
Typically, more than 20 weeks from the start of wafer processing to final chips is not abnormal. Thus, if a shortage in the supply chain occurs, e.g. due to increasing demand, allocation for 30-50 weeks is not really special. In the case of our products where we do rather complex backside processing, we had to add another 11 weeks to the supply lead time one year ago. This situation was not satisfying at all and we decided to invest heavily into additional in-house capabilities.
The result is now that we can do the backside and the back-end processing of the wafers within days, if needed. We showed this performance in a recent project where lead time from wafer production start until the chips were in the lab was only nine weeks. Not just wafer processing and backside processing, but also assembly including AR coating and band-pass filter bonding."
CIS Saves Fabs from Closure
IC Insights publishes a report on worlds fab closures. It turns out that CIS market expansion has saved two 300mm fabs from closure:
"Renesas sold its 300mm logic fab to Sony in 2014. Sony repurposed that fab to make image sensors. In 2017, Samsung closed its 300mm Line 11 memory fab in Yongin, South Korea, also repurposing it to manufacture image sensors."
According to other sources, the re-purporsed Samsung fab is located in Hwaseong in Gyeonggi Province and should start CIS manufacturing in 1H 2018.
"Renesas sold its 300mm logic fab to Sony in 2014. Sony repurposed that fab to make image sensors. In 2017, Samsung closed its 300mm Line 11 memory fab in Yongin, South Korea, also repurposing it to manufacture image sensors."
According to other sources, the re-purporsed Samsung fab is located in Hwaseong in Gyeonggi Province and should start CIS manufacturing in 1H 2018.
Thursday, March 01, 2018
CSEM Flash LiDARs
AutoSens publishes CSEM Christophe Pache presentation on flash LiDARs with a Q&A session starting at 20:20 time:
Renesas Announces Automotive Stereo Image Processor
BusinessWire: Renesas announced the R-Car V3H SoC for computer vision and AI processing at industry-leading low power levels, targeting automotive front cameras for use in mass-produced Level 3 (conditional automation) and Level 4 (high automation) autonomous vehicles. The new SoC is optimized for use in stereo front cameras and achieves five times the computer vision performance of its predecessor, the R-Car V3M SoC.
The R-Car V3H algorithms include Dense Optical Flow, Dense Stereo Disparity, and Object Classification. The integrated IP for CNN accelerates deep learning at industry-leading low power levels of only 0.3 watts, achieving more than two times of the deep neural network performance of the R-Car V3M.
Samples of the R-Car V3H SoC will be available from Q4. Mass production is scheduled to begin in Q3, 2019.
The R-Car V3H algorithms include Dense Optical Flow, Dense Stereo Disparity, and Object Classification. The integrated IP for CNN accelerates deep learning at industry-leading low power levels of only 0.3 watts, achieving more than two times of the deep neural network performance of the R-Car V3M.
Samples of the R-Car V3H SoC will be available from Q4. Mass production is scheduled to begin in Q3, 2019.
Dalsa Expands CMOS X-Ray Sensor Manufacturing in Holland
BusinessWire: Teledyne DALSA is expanding its manufacturing capacity given increased demand for the company’s CMOS-based digital X-ray detectors. Teledyne DALSA is a leader in the design and assembly of specialized CMOS X-ray detectors which offer superior lag-free, real-time imaging at both higher resolution and reduced X-ray dose levels.
Teledyne detectors feature proprietary active pixel architecture which offers higher image quality, QE and SNR compared with image intensified charge coupled devices (IICCDs), amorphous silicon (a-Si) or amorphous selenium (a-Se) detectors and even other CMOS-based competitive products.
In order to service its global customer base, Teledyne currently operates X-ray detector manufacturing and assembly locations in the Netherlands, Canada and the US.
“The expansion of the cleanroom facilities in the Netherlands will help us satisfy greater anticipated demand from dental and medical imaging OEMs,” said Robert Mehrabian, Chairman, President and CEO of Teledyne. “In fact, we currently expect that demand for our detectors will more than double by 2020 compared with 2017.”
Teledyne detectors feature proprietary active pixel architecture which offers higher image quality, QE and SNR compared with image intensified charge coupled devices (IICCDs), amorphous silicon (a-Si) or amorphous selenium (a-Se) detectors and even other CMOS-based competitive products.
In order to service its global customer base, Teledyne currently operates X-ray detector manufacturing and assembly locations in the Netherlands, Canada and the US.
“The expansion of the cleanroom facilities in the Netherlands will help us satisfy greater anticipated demand from dental and medical imaging OEMs,” said Robert Mehrabian, Chairman, President and CEO of Teledyne. “In fact, we currently expect that demand for our detectors will more than double by 2020 compared with 2017.”
Pixelplus Launches "Moving Image HDR" Sensors
South China Morning Post publishes an article about Pixelplus launching "moving image HDR sensors" for the automotive applications.
“HDR technology was developed several years ago on the image signal processing (ISP) level, but the issue with digital artefact persisted. Our solution is an image sensor technology that eliminates artefact because functionality is built inside the image sensor itself,” says SK Lee, president and CEO of Pixelplus. “Our plan is to be among the global top five image sensor providers, especially for the automotive industry based on new camera platforms.”
“HDR technology was developed several years ago on the image signal processing (ISP) level, but the issue with digital artefact persisted. Our solution is an image sensor technology that eliminates artefact because functionality is built inside the image sensor itself,” says SK Lee, president and CEO of Pixelplus. “Our plan is to be among the global top five image sensor providers, especially for the automotive industry based on new camera platforms.”
Lensvector Done with Imaging
PRWeb: LensVector seems to shift its liquid crystal lens efforts away from imaging to light fixtures, possibly a result of appointing a new CEO:
A Clever Idea on Paper Falls Short in Tests
Imaging Resource publishes a review of Light.co L16 computational camera and the conclusion is quite bad:
"After years of hype and teasers, we finally got our hands on one, and suffice it to say, the image quality and performance leave a lot to desired.
...shooting out in the real world, the L16 is pretty much underwhelming on all fronts."
"After years of hype and teasers, we finally got our hands on one, and suffice it to say, the image quality and performance leave a lot to desired.
...shooting out in the real world, the L16 is pretty much underwhelming on all fronts."
 |
| Fine "detail" crop |
 |
| Light L16 camera |
Subscribe to:
Posts (Atom)