wilzy123
Founding Member
Indeed. Noting also that market cap is not the only criterion for inclusion.
BRN Discussion Ongoing
- Thread starter TechGirl
- Start date
Indeed. Noting also that market cap is not the only criterion for inclusion.
Fullmoonfever
Top 20
If wanting to track & dig around FDA medical devices.... here's a start point.
www.fda.gov
October 5, 2022 update: 178 Artificial Intelligence and Machine Learning (AI/ML)-Enabled Medical Devices were added to the list below. With this update, the FDA has also added the ability to download the list as an Excel file.
And yes, Siemens pop up in the list as well for e.g.
Artificial Intelligence and Machine Learning (AI/ML)-Enabled Medical D
The FDA has updated the list of AI/ML-enabled medical devices marketed in the United States as a resource to the public.
October 5, 2022 update: 178 Artificial Intelligence and Machine Learning (AI/ML)-Enabled Medical Devices were added to the list below. With this update, the FDA has also added the ability to download the list as an Excel file.
And yes, Siemens pop up in the list as well for e.g.
Fullmoonfever
Top 20
As Sailesh said in that previous interview.
“At the very low end, we have added an ARM M33 MCU and spiking neural network with BrainChip core licensed for selected applications – we have licensed what we need to license from BrainChip including the software to get the ball rolling.
Not a bad little site Avnet Silica for info as a reseller / supplier.
DNA, making these products feature- and pin-compatible. This allows easy scalability and code reuse from one device to another.
The RA41 if based on an 100 MHz Arm® Cortex®-M33 core, offers 512 kB code flash memory and 128 kB SRAM, and provides USB 2.0 Full-Speed, SDHI, Quad SPI, safety features, and advanced analog with one ADC unit.
The RA4E1 group is built on a highly efficient 40nm process and is supported by an open and flexible ecosystem concept, called Flexible Software Package (FSP), including FreeRTOS and Azure RTOS, but can be replaced and expanded by any other RTOS or middleware user’s need.
RA4E1 group is suitable for IoT applications requiring many connectivity options, large embedded RAM, 100 MHz performance and low power consumption.
READ MORE
The Renesas RA4M2 is based on the latest Arm® Cortex®-M33 core with TrustZone, and offers leading performance with an operating frequency of 100 MHz. With its additional memory options of of 512KB / 348KB / 256KB Flash as RA4M3’s 1MB / 768KB Flash, it enables a cost-effective design. The RA4M2 group also offers additional low pin count package options (LQFP and QFN 48-pin).
The RA4M2 is built on a highly efficient 40nm process and is supported by an open and flexible ecosystem concept, called Flexible Software Package (FSP), using FreeRTOS as base, but can be replaced and expanded by any other RTOS or middleware user’s need. RA4M2 is suitable for IoT application requiring strong Security, large embedded RAM with parity/ECC and low power consumption.
READ MORE
With the RA4M3 group, Renesas expands their Renesas RA4 series. It is based on the latest Arm Cortex-M33 Core with TrustZone with increased 1 MB on-chip Flash and 128KB on-chip RAM. The integrated Standby RAM along with the low power consumption of the Renesas 40nm process is the optimum for low power applications. The memory block swap feature, new on the RA4 series, in conjunction with the Security Crypto Engine is making the RA4M3 the best choice for applications where in field firmware updates are required.
The enhanced integrated Security Crypto Engine with several cryptography accelerators, key management support, tamper detection and power analysis resistance in concert with the newly implemented Arm TrustZone is making the devices the best choice in the market for connected applications. The memory can be extended with Quad-SPI, Octa-SPI and SD-Card interface for more memory hungry applications.
READ MORE
The Renesas RA4M1 group uses the high-performance Arm® Cortex®-M4F core and offers a Segment LCD Controller and a Capacitive Touch Sensing Unit input for intensive HMI designs. The RA4M1 is built on a highly efficient low power process and is supported by an open and flexible ecosystem concept—the Flexible Software Package (FSP), built on FreeRTOS—and is expandable to use other RTOSes and middleware. The RA4M1 is suitable for applications where a large amount of Capacitive Touch channels and a Segment LCD controller are required.
DOWNLOAD DATASHEET
https://www.avnet.com/wps/portal/silica/products/new-products/npi/2020/renesas-ra6m4/
READ MORE
https://www.avnet.com/wps/portal/silica/resources/technical-videos/renesas-ra-techtalks/
WATCH VIDEOS NOW
“At the very low end, we have added an ARM M33 MCU and spiking neural network with BrainChip core licensed for selected applications – we have licensed what we need to license from BrainChip including the software to get the ball rolling.
Not a bad little site Avnet Silica for info as a reseller / supplier.
Avnet: Quality Electronic Components & Services
Avnet is a global leader of electronic components and services, guiding makers and manufacturers from design to delivery. Let Avnet help you reach further.
www.avnet.com
Renesas RA - 32-bit MCU family
With ARM® Cortex®-M Core, the new MCU series delivers the ultimate promise of IoT with software flexibility
The Renesas RA Family is a new 32-bit MCU family built on the Arm® Cortex®-M core architecture. Offering a wide range of performance and features, the Renesas RA Family meet the scalability, power consumption and performance needs of nearly any embedded systems end-product:- Strong Security
- Secure Crypto Engine (SCE) IP
- An extra layer of embedded hardware security providing tamper detection and resistance to side-channel attacks
- All built on top of Arm’s v8-M TrustZone® in combination with Cortex M33 processor cores
- Arm Core
- Based on Arm’s next-generation Cortex-M23/M33 processor cores, and Cortex-M4 core
- Flexible Software Solution
- Supported by an open and flexible ecosystem concept, the Flexible Software Package (FSP) uses FreeRTOS as a base
- Can be replaced and expanded by any other RTOS or middleware
- Best-in-Class Peripheral IP
- Excellent HMI capacitive touch technology
- The industry’s highest code flash memory capacity
- Wide range of connectivity solutions
Renesas RA product series
The four Renesas RA Family MCU series are based on 32-bit Arm® Cortex®-M cores. All four Renesas RA Series have been designed on commonDNA, making these products feature- and pin-compatible. This allows easy scalability and code reuse from one device to another.
- RA2 Series
- RA4 Series
- RA6 Series
- RA8 Series
Renesas RA4E1 MCU group
RA4E1 - 100MHz Arm® Cortex®-M33, the first entry-line product of the RA seriesThe RA41 if based on an 100 MHz Arm® Cortex®-M33 core, offers 512 kB code flash memory and 128 kB SRAM, and provides USB 2.0 Full-Speed, SDHI, Quad SPI, safety features, and advanced analog with one ADC unit.
The RA4E1 group is built on a highly efficient 40nm process and is supported by an open and flexible ecosystem concept, called Flexible Software Package (FSP), including FreeRTOS and Azure RTOS, but can be replaced and expanded by any other RTOS or middleware user’s need.
RA4E1 group is suitable for IoT applications requiring many connectivity options, large embedded RAM, 100 MHz performance and low power consumption.
READ MORE
Renesas RA4M2 MCU group
RA4M2 - High-performance 100 MHz Arm® Cortex®-M33 core based on latest Armv8-M architectureThe Renesas RA4M2 is based on the latest Arm® Cortex®-M33 core with TrustZone, and offers leading performance with an operating frequency of 100 MHz. With its additional memory options of of 512KB / 348KB / 256KB Flash as RA4M3’s 1MB / 768KB Flash, it enables a cost-effective design. The RA4M2 group also offers additional low pin count package options (LQFP and QFN 48-pin).
The RA4M2 is built on a highly efficient 40nm process and is supported by an open and flexible ecosystem concept, called Flexible Software Package (FSP), using FreeRTOS as base, but can be replaced and expanded by any other RTOS or middleware user’s need. RA4M2 is suitable for IoT application requiring strong Security, large embedded RAM with parity/ECC and low power consumption.
READ MORE
Renesas RA4M3 MCU group
RA4M3 - 100MHz ARM Cortex M33 with TrustZone, Security and Memory enhancementsWith the RA4M3 group, Renesas expands their Renesas RA4 series. It is based on the latest Arm Cortex-M33 Core with TrustZone with increased 1 MB on-chip Flash and 128KB on-chip RAM. The integrated Standby RAM along with the low power consumption of the Renesas 40nm process is the optimum for low power applications. The memory block swap feature, new on the RA4 series, in conjunction with the Security Crypto Engine is making the RA4M3 the best choice for applications where in field firmware updates are required.
The enhanced integrated Security Crypto Engine with several cryptography accelerators, key management support, tamper detection and power analysis resistance in concert with the newly implemented Arm TrustZone is making the devices the best choice in the market for connected applications. The memory can be extended with Quad-SPI, Octa-SPI and SD-Card interface for more memory hungry applications.
READ MORE
Renesas RA4M1 MCU group
48MHz ARM Cortex M4 with LCD controller and Cap Touch for HMIThe Renesas RA4M1 group uses the high-performance Arm® Cortex®-M4F core and offers a Segment LCD Controller and a Capacitive Touch Sensing Unit input for intensive HMI designs. The RA4M1 is built on a highly efficient low power process and is supported by an open and flexible ecosystem concept—the Flexible Software Package (FSP), built on FreeRTOS—and is expandable to use other RTOSes and middleware. The RA4M1 is suitable for applications where a large amount of Capacitive Touch channels and a Segment LCD controller are required.
Key features
- 48MHz Arm® Cortex®-M4F
- 256kB Flash Memory and 32kB SRAM
- 8kB DataFlash to store data as in EEPROM
- Scalable from 40pin to 100pin Packages
- Segment LCD Controller
- 14-bit A/D Converter
- Capacitive Touch Sensing Unit
- USB2.0 Full Speed
- CAN 2.0B
- SCI (UART, Simple SPI, Simple I2C)
- SPI/ I2C Multimaster Interface
Benefits
- Integrated Crypto Module with AES cryptography accelerator and Key management support
- Segment LCD Controller (up to 38seg/4com; 34seg/8com) for HMI
- 27 channels of Capacitive Touch Sensing Unit input for intensive HMI designs
Applications
- HMI with Seg/Com LCD display
- Security (Fire Detection, Burglar Detection, Panel Control)
- Industry (Door Openers, Panel Control)
- HVAC (Heating, Air Conditioning, Boiler Control)
- Home Appliance
- General Purpose
Evaluation kit
Full MCU evaluation including On-Chip debugger – Part name: RTK7EKA4M1S00001BUDOWNLOAD DATASHEET
RA TechTalks
Renesas
RA6M4 Group
RA6M4 is suitable for IoT application requiring strong Security, rich connectivity, large embedded RAM with parity/ECC and low power consumption.https://www.avnet.com/wps/portal/silica/products/new-products/npi/2020/renesas-ra6m4/
READ MORE
Renesas
RA TechTalk
We launched new videos & tutorials hosted by our RA MCU Expert Lou Leen. Learn how to develop your application with the Renesas RA MCU family.https://www.avnet.com/wps/portal/silica/resources/technical-videos/renesas-ra-techtalks/
WATCH VIDEOS NOW
Related Links |
Related Documents |
Have a question? Contact usEmail:For general questions: yourmessage@avnet.eu Local Avnet Silica offices: Click here to find contact information for your local Avnet Silica team. |
Bravo
If ARM was an arm, BRN would be its biceps💪!
"The choice is up to the partners"...interesting. Qualcomm is leaving the partners or the individual phone makers in this case, to decide what capabilities they want in the phones. I wonder if this could mean they're potentially leaving the door open to their partners to choose to implement Akida for additional AI capabilities at the edge.
It's not beyond the realms of possibilities it it?
au.pcmag.com
It's not beyond the realms of possibilities it it?
First Listen: Taking Qualcomm's Snapdragon Sound for a Spin
Qualcomm hopes to bring a handful of new experiences to next year's Bluetooth headphones and speakers, including dynamic spatial audio and lossless playback. We gave the upgrades a quick listen. Here are our thoughts.
alwaysgreen
Top 20
Proga
Regular
We haven't seen any revenue for those licencing agreements yet so they probably be in the next 4c if they've paid. It's not unusual for companies to take 60-90 days to pay invoices.
thelittleshort
Top Bloke
Just saw this on LinkedIn
Free online machine learning course from Stanford taught by Andrew Ng
www.coursera.org
Approximately 3 months to complete
Suggested pace of 9 hours/week
English
Subtitles: English
The Machine Learning Specialization is a foundational online program created in collaboration between DeepLearning.AI and Stanford Online. This beginner-friendly program will teach you the fundamentals of machine learning and how to use these techniques to build real-world AI applications.
This Specialization is taught by Andrew Ng, an AI visionary who has led critical research at Stanford University and groundbreaking work at Google Brain, Baidu, and Landing.AI to advance the AI field.
This 3-course Specialization is an updated version of Andrew’s pioneering Machine Learning course, rated 4.9 out of 5 and taken by over 4.8 million learners since it launched in 2012.
It provides a broad introduction to modern machine learning, including supervised learning (multiple linear regression, logistic regression, neural networks, and decision trees), unsupervised learning (clustering, dimensionality reduction, recommender systems), and some of the best practices used in Silicon Valley for artificial intelligence and machine learning innovation (evaluating and tuning models, taking a data-centric approach to improving performance, and more.)
By the end of this Specialization, you will have mastered key concepts and gained the practical know-how to quickly and powerfully apply machine learning to challenging real-world problems. If you’re looking to break into AI or build a career in machine learning, the new Machine Learning Specialization is the best place to start.
• Build machine learning models in Python using popular machine learning libraries NumPy and scikit-learn.
• Build and train supervised machine learning models for prediction and binary classification tasks, including linear regression and logistic regression.
• Build and train a neural network with TensorFlow to perform multi-class classification.
• Apply best practices for machine learning development so that your models generalize to data and tasks in the real world.
• Build and use decision trees and tree ensemble methods, including random forests and boosted trees.
• Use unsupervised learning techniques for unsupervised learning: including clustering and anomaly detection.
• Build recommender systems with a collaborative filtering approach and a content-based deep learning method.
• Build a deep reinforcement learning model.
Free online machine learning course from Stanford taught by Andrew Ng
Machine Learning
Offered by Stanford University and DeepLearning.AI. #BreakIntoAI with Machine Learning Specialization. Master fundamental AI concepts and ... Enroll for free.
www.coursera.org
Approximately 3 months to complete
Suggested pace of 9 hours/week
English
Subtitles: English
WHAT YOU WILL LEARN
- Build ML models with NumPy & scikit-learn, build & train supervised models for prediction & binary classification tasks (linear, logistic regression)
- Build & train a neural network with TensorFlow to perform multi-class classification, & build & use decision trees & tree ensemble methods
- Apply best practices for ML development & use unsupervised learning techniques for unsupervised learning including clustering & anomaly detection
- Build recommender systems with a collaborative filtering approach & a content-based deep learning method & build a deep reinforcement learning model
About this Specialization
431,767 recent viewsThe Machine Learning Specialization is a foundational online program created in collaboration between DeepLearning.AI and Stanford Online. This beginner-friendly program will teach you the fundamentals of machine learning and how to use these techniques to build real-world AI applications.
This Specialization is taught by Andrew Ng, an AI visionary who has led critical research at Stanford University and groundbreaking work at Google Brain, Baidu, and Landing.AI to advance the AI field.
This 3-course Specialization is an updated version of Andrew’s pioneering Machine Learning course, rated 4.9 out of 5 and taken by over 4.8 million learners since it launched in 2012.
It provides a broad introduction to modern machine learning, including supervised learning (multiple linear regression, logistic regression, neural networks, and decision trees), unsupervised learning (clustering, dimensionality reduction, recommender systems), and some of the best practices used in Silicon Valley for artificial intelligence and machine learning innovation (evaluating and tuning models, taking a data-centric approach to improving performance, and more.)
By the end of this Specialization, you will have mastered key concepts and gained the practical know-how to quickly and powerfully apply machine learning to challenging real-world problems. If you’re looking to break into AI or build a career in machine learning, the new Machine Learning Specialization is the best place to start.
Applied Learning Project
By the end of this Specialization, you will be ready to:• Build machine learning models in Python using popular machine learning libraries NumPy and scikit-learn.
• Build and train supervised machine learning models for prediction and binary classification tasks, including linear regression and logistic regression.
• Build and train a neural network with TensorFlow to perform multi-class classification.
• Apply best practices for machine learning development so that your models generalize to data and tasks in the real world.
• Build and use decision trees and tree ensemble methods, including random forests and boosted trees.
• Use unsupervised learning techniques for unsupervised learning: including clustering and anomaly detection.
• Build recommender systems with a collaborative filtering approach and a content-based deep learning method.
• Build a deep reinforcement learning model.
Proga
Regular
@Bravo, @Diogenese pointed out 8 Gen 2 have some low end SNN capabilities."The choice is up to the partners"...interesting. Qualcomm is leaving the partners or the individual phone makers in this case, to decide what capabilities they want in the phones. I wonder if this could mean they're potentially leaving the door open to their partners to choose to implement Akida for additional AI capabilities at the edge.
It's not beyond the realms of possibilities it it?
View attachment 22741
View attachment 22742
First Listen: Taking Qualcomm's Snapdragon Sound for a Spin
Qualcomm hopes to bring a handful of new experiences to next year's Bluetooth headphones and speakers, including dynamic spatial audio and lossless playback. We gave the upgrades a quick listen. Here are our thoughts.
buena suerte :-)
BOB Bank of Brainchip
Awesome ... And they rounded us out @ 70c very nice
alwaysgreen
Top 20
The half yearly report will be more interesting imo. It's not a cash report but shows all revenue generated in the last 6 months.
Fingers crossed for $10 mill plus.
Violin1
Regular
Totally agree FF. The company is moving forward but the trajectory not what we anticipated a year ago. Understandable and we should recognise and accept this publicly between now and release of the next 4c or we'll play into the hands of the dickheads once again. Seriously - we'll get there - but pleeeease don't roll-the-dice on this quarter's revenue. Everything is heading the right direction. I hate to say it again but listen to Master Po!!!
Patience grasshopper.....patience.....
Hi FMF,
Can you please give me a link to this quote??
“At the very low end, we have added an ARM M33 MCU and spiking neural network with BrainChip core licensed for selected applications – we have licensed what we need to license from BrainChip including the software to get the ball rolling.
Thanks heaps,
Iseki
Fullmoonfever
Top 20
About half way down.
ARM battles RISC-V at Renesas
Dr. Sailesh Chittipeddi, Executive Vice President and General Manager of IoT and Infrastructure Business Unit of Renesas Electronics talks to eeNews Europe
www.eenewseurope.com
Brilliant. Thanks so much!About half way down.
ARM battles RISC-V at Renesas
Dr. Sailesh Chittipeddi, Executive Vice President and General Manager of IoT and Infrastructure Business Unit of Renesas Electronics talks to eeNews Europe
I have previously looked and looked thru Renesas product catalogue to see where this chip is.
The whole computer chip business seems to be running to a totally new paradigm. Gone are the days you'd but some CPU and add some other microcontrollers all off the shelf. Now it seems to be all about hiring the designers from renesas, arm, or megachips or sifive and they'll design something uniquely for you.
Fullmoonfever
Top 20
No probs.
Someone posted the article when it first came out so occasionally I have a scrounge around for M4, M4f and M33 controllers to see what might turn up.
The RA series has potential as does the V2L I posted about in the Renesas thread a while back...imo.
If you have a look at the Edge Impulse public project pages I posted the other day, the dropdown menu for device targets also has a list of devices which I presume are compatible with the Akida NN as they put it.
These include a Texas Instrument, ST Micro, Sony Spresense amongst the usual suspects Arduino, Raspberry and others.
wilzy123
Founding Member
You used too many words. Readership does not contain enough neurons to compute.
TechGirl
Founding Member
hmmmm 
www.prophesee.ai
Researchers at the DLR Institute of Propulsion Technology of the German Aerospace Center explore the possibilities of harnessing event-based vision (EBV) for capturing flow fields in both air and water flows. Using a laser or other light source for illumination, the event-camera captures the motion of small (micrometer) sized particles carried with the flow.
Maps of the evolving flow field can then be obtained from the event tracks produced by particles. Compared to previous studies, a much higher number of particles can be tracked simultaneously, resulting in highly resolved velocity maps.
This has been made possible through recent advances in EBV hardware development featuring high-speed, high resolution systems such as the EVK4 HD by Prophesee.
Event data produced by small particles in a turbulent water flow
Flow velocity measurement
Within the ongoing study a variety of aspects from event-capture to algorithms capable of recovering flow velocity information are being explored. So far flow measurements of up to several meters per second have been demonstrated on a field of view of 100 mm width.
The high temporal resolution of the event data provides velocity information nearly equivalent to imagery obtained with costly high-speed cameras operating beyond 1000 frames per second.
The hardware demands for event-based flow velocimetry are minimal: it only requires an event camera and a small laser with light sheet forming lenses to capture the motion of small particles moving with the flow. This makes the system compact and very portable.
The “table-top” setup on the right was used to record the event data and flow fields above. Here, the water is “seeded” with small (10 µm) particles that scatter light from a thin (1 mm) laser sheet, that is introduced from the side.
Owing to the high sensitivity of the event camera, a laser power of less than 1 watt is already sufficient to capture the event imagery shown above.
Simple “table-top” setup for EBV imaging of a water flow involving a small water tank, laser light sheet and Prophesee event-based camera, here EVK 2 HD
Measurements close to surfacesMotion of individual particles can be efficiently reconstructedpossible by suppressing laser scatterusing the underlying flow field as guidanceBy synchronizing between different unitsA low-cost laser or LED light source can be usedModerate laser power is sufficient to imageMulti-view imagingReconstruction of particle paths in time and 2d spaceinstead of commonly used high-power lasersmicrometer-sized particles in water and airis readily extended to time-resolved 3d particle tracking
Event-based vision has many advantages for fluid flow measurement.
WATCH THE WEBINAR
The Institute of Propulsion Technology is strictly focused on the improvement of gas turbines in aviation and electric power generation by medium- to long-term exploitation of the inherent technical potentials. It is a part of DLR which is the Federal Republic of Germany’s research centre for aeronautics and space, conducting research and development activities in the fields of aeronautics, space, energy, transport, security and digitalization. At the DLR Institute of Propulsion Technology the department of Engine Measurement Systems is devoted to the development and application of laser optical measurement and imaging techniques for the aerothermodynamic investigation of aero engine components and related fields. The development and adaption of new measurement techniques contributes to the desired progress in turbomachinery research, with their application both at the test rigs of the Institute as well as on the facilities of external customers.
High efficiency real-time flow visualization with event-based vision
Researchers at the DLR Institute of Propulsion Technology are exploring the possibilities of harnessing the potentials of event-based vision for fluid flow measurement.
EVENT-BASED VISION ENABLES HIGH EFFICIENCY REAL-TIME FLOW VISUALIZATION
Real-time flow visualization High resolution flow field data resolution of tracked particlesResearchers at the DLR Institute of Propulsion Technology of the German Aerospace Center explore the possibilities of harnessing event-based vision (EBV) for capturing flow fields in both air and water flows. Using a laser or other light source for illumination, the event-camera captures the motion of small (micrometer) sized particles carried with the flow.
Maps of the evolving flow field can then be obtained from the event tracks produced by particles. Compared to previous studies, a much higher number of particles can be tracked simultaneously, resulting in highly resolved velocity maps.
This has been made possible through recent advances in EBV hardware development featuring high-speed, high resolution systems such as the EVK4 HD by Prophesee.
Event data produced by small particles in a turbulent water flow
(The field of view is about 55 mm wide and 30 mm high)
Turbulent flow in a water tank reconstructed from captured event data
Flow velocity measurement
Within the ongoing study a variety of aspects from event-capture to algorithms capable of recovering flow velocity information are being explored. So far flow measurements of up to several meters per second have been demonstrated on a field of view of 100 mm width.
The high temporal resolution of the event data provides velocity information nearly equivalent to imagery obtained with costly high-speed cameras operating beyond 1000 frames per second.
The hardware demands for event-based flow velocimetry are minimal: it only requires an event camera and a small laser with light sheet forming lenses to capture the motion of small particles moving with the flow. This makes the system compact and very portable.
The “table-top” setup on the right was used to record the event data and flow fields above. Here, the water is “seeded” with small (10 µm) particles that scatter light from a thin (1 mm) laser sheet, that is introduced from the side.
Owing to the high sensitivity of the event camera, a laser power of less than 1 watt is already sufficient to capture the event imagery shown above.
Simple “table-top” setup for EBV imaging of a water flow involving a small water tank, laser light sheet and Prophesee event-based camera, here EVK 2 HD
Measurements close to surfacesMotion of individual particles can be efficiently reconstructedpossible by suppressing laser scatterusing the underlying flow field as guidanceBy synchronizing between different unitsA low-cost laser or LED light source can be usedModerate laser power is sufficient to imageMulti-view imagingReconstruction of particle paths in time and 2d spaceinstead of commonly used high-power lasersmicrometer-sized particles in water and airis readily extended to time-resolved 3d particle tracking
Event-based vision has many advantages for fluid flow measurement.
- It is suitable for real-time flow visualization – The event tracks produced by the movement of the particle field directly visualize the flow field without the need of further processing.
- Event-records obtained from fast moving flows can readily be played back in “slow-motion” mode to visualize the flow dynamics.
- Event cameras are tolerant against intensity variations within the light sheet because its sensors only react on intensity changes.
- They are also tolerant against laser flare (light scatter on surfaces) and non-uniform background intensity. Their tolerance against high background intensity allows particles to be tracked in bright environments.
- Particle event tracks are clearly distinguishable from the background and can be extracted either by feature detection or by using the clusters formed in the motion-compensation analysis.
WATCH THE WEBINAR
ABOUT DLR INSTITUTE OF PROPULSION TECHNOLOGY
The Institute of Propulsion Technology is strictly focused on the improvement of gas turbines in aviation and electric power generation by medium- to long-term exploitation of the inherent technical potentials. It is a part of DLR which is the Federal Republic of Germany’s research centre for aeronautics and space, conducting research and development activities in the fields of aeronautics, space, energy, transport, security and digitalization. At the DLR Institute of Propulsion Technology the department of Engine Measurement Systems is devoted to the development and application of laser optical measurement and imaging techniques for the aerothermodynamic investigation of aero engine components and related fields. The development and adaption of new measurement techniques contributes to the desired progress in turbomachinery research, with their application both at the test rigs of the Institute as well as on the facilities of external customers.
David holland
Regular
Means Absolutely nothing
Similar threads
