Smart Cities Archives - Page 322 of 569 - SmartCitiesElectronics.com

AI controller and software detect humans, using less energy

Tue, Apr 20, 2021

Claimed to be the lowest power person detection system, the combination of Maxim Integrated’s MAX78000 artificial intelligence (AI) microcontroller and Aizip’s Visual Wake Words model bring human-figure detection to IoT imaging and video with a power budget of just 0.7 mJ per inference. This, says Maxim, is a 100 fold improvement on the performance of conventional software, making it “the most economical and efficient IoT person-detection solution available”.

The MAX78000 neural-network microcontroller detects people in an image using Aizip’s Visual Wake Words (VWW) model at just 0.7 mJ per inference. This allows 13 million inferences from a single AA/LR6 battery and is 100 times lower than conventional software solutions. It is, says Maxim, the most economical and efficient IoT person-detection solution available. The low-power network provides longer operation for battery-powered IoT systems that require human-presence detection, including building energy management and smart security cameras.

The MAX78000 low-power, neural network accelerated microcontroller executes AI inferences at less than 1/100th the energy of conventional software solutions to improve run-time for battery-powered edge AI applications, continues Maxim. The mixed precision VWW network is part of the Aizip Intelligent Vision Deep Neural Network (AIV DNN) series for image and video applications and was developed with Aizip’s proprietary design automation tools to achieve greater than 85 per cent human-presence accuracy.

The extreme model compression delivers accurate smart vision with a memory-constrained, low-cost AI-accelerated microcontroller and cost-sensitive image sensors, says Maxim.

The MAX78000 microcontroller and MAX78000EVKIT# evaluation kit are available now directly from Maxim Integrated’s website and through authorised distributors.

AIV DNN series models, tools and services are available directly from Aizip.

Aizip develops AI models for IoT applications. Based in Silicon Valley, Aizip provides design services with superior performance, quick turnaround time, and “excellent [return on investment] ROI.

Maxim Integrated offers a broad portfolio of semiconductors, tools and support to deliver efficient power, precision measurement, reliable connectivity and robust protection along with intelligent processing. Designers in application areas such as automotive, communications, consumer, data centre, healthcare, industrial and the IoT.

http://www.maximintegrated.com

> Read More

IoT kit demonstrates value for narrow band technology

Tue, Apr 20, 2021

Pre-integrated, pre-tested and pre-secure products are available in the IoT Network kit from Wittra which takes customers “straight to the ‘Proof of Value’,” says the Swedish company.

It is based on based on open standards to ensure interoperability and ease of integration, enabling users to collect, communicate and control assets. Devices run on a 6lowpan IP-based true mesh radio network which uses the sub-GHz spectrum providing long range and good penetration of structures for robust and reliable data delivery in any setting.

Wittra says its positioning technology is a simple, practical approach for tracking and monitoring assets, and enables total asset visibility in all environments which were not considered possible using narrow band technology.

To reduce the complexity of an IoT project, Wittra offers pre-integrated, pre-tested and pre-secure products for immediate deployment.

The IoT Network kit contains the Wittra gateway, sensor tags, mesh routers and all the associated accessories ensuring an IoT project is up and running in hours, by deploying the Wittra Solution via the cloud-based application programming interface (API), claims the company.

Each tag contains several sensors which include temperature, accelerometer, gyroscope, magnetometer, and positioning. Additional sensors can be added, covering humidity, ambient light, and air pressure. The mesh network is extended in range by Wittra’s mesh routers creating a multi-hop, self-forming and self-healing true mesh network.

Products are available through Mouser and Premier Farnell.

Wittra Sweden develops technologies and solutions within the IoT. Since 2012, Wittra has been able to secure a substantial portfolio of intellectual property (IP) and rights. The pre-integrated, pre-tested and pre-secure solutions allow users to collect, communicate and control their assets even in the toughest of environments, claims the company.

http://www.wittra.se

> Read More

Omnivision releases automotive image sensors for Nvidia AGX AI platform

Tue, Apr 20, 2021

Image sensors from OmniVision Technologies are compatible with Nvidia’s Drive AGX artificial intelligence (AI) computing platform for autonomous vehicles and advanced driver assistance systems (ADAS).

The company announced that it has joined the Nvidia Drive autonomous vehicle development system at the (virtual) Nvidia GTC21.

The first three qualified image sensor families offer unique benefits to automotive system designers. The OV2311 features 2Mpixel resolution for driver monitoring systems (DMS) applications, and was the industry’s first automotive-grade imager to offer a global shutter for minimal driver-motion artefacts. The new 5Mpixel version of the image sensor also adds OmniVision’s Nyxel near-infrared (NIR) technology, for the best image captures in low- to no-light conditions, says Omnivision. Nyxel technology is claimed to achieve the world’s highest quantum efficiency for DMS of 36 per cent at the invisible 940nm NIR light wavelength, providing the clearest driver images for use by AI software algorithms.

The second family is the OX08B40, believed to be the industry’s first automotive image sensor with 140dB HDR, LED flicker mitigation (LFM) and 8.3Mpixel resolution. The image sensor enables superior front view captures, regardless of external lighting conditions, claims OmniVision. OmniVision’s on-chip HALE (HDR and LFM engine) combination algorithm simultaneously provides industry-leading LFM and HDR over the entire automotive temperature range, claims the company.

The OX03C10 and OX03F10 are claimed to be the only automotive viewing camera image sensors to combine a large 3.0 micron pixel with the HALE algorithm for minimised motion artefacts and the best LFM performance. The OX03C10 also has the lowest power consumption of any 2.5MP LFM image sensor, says OmniVision. It is also the industry’s smallest package size, enabling cameras that continuously run at 60 fps to be fitted in even the tightest confines of automotive design.

The OX03F10 widens the vertical resolution to 1920x1536p, for image quality appropriate for which provides the high image quality needed when feeding surround view system (SVS) captures into autonomous machine vision systems. The OX03F10’s wider array is also essential in e-mirror applications, for greater coverage and to eliminate blindspots.

OmniVision’s OV2311 and a new 5Mpixel sensor are the exclusive driver monitoring system (DMS) image sensors for the Nvidia Drive Hyperion evaluation architecture.

“Paul Wu, automotive staff marketing manager at OmniVision, said: “We are excited to become an Nvidia Drive ecosystem partner, with three of our premium automotive image sensor families and many more to be added. Our goal is to accelerate time to market for autonomous and ADAS applications by reducing the development effort and cost”. The company will also provide sensors for the driver and occupant monitoring systems on the next-generation Nvidia Drive Hyperion platform which combines hardware and software.

The Nvidia Drive AGX AI computing platform offers automotive designers the flexibility to use Nvidia hardware with its full stack Drive software, or develop custom software. The platform can be used for in-cabin or for autonomous driving applications, or for both. OmniVision offers platform software drivers for its image sensors, as well as complete camera modules that customers can connect to the Nvidia Drive platform for immediate evaluation. OmniVision also provides custom tuning services.

“Adding innovative suppliers like OmniVision to our open AV development platform is a key element of accelerating automotive breakthroughs with Nvidia Drive,” said Glenn Schuster, senior director of sensor ecosystems at Nvidia.

All three initial Nvidia Drive AGX platform-compatible sensor families are available now for sampling and mass production, featuring advanced ASIL functional safety and AEC-Q100 Grade 2 certification. The platform drivers and evaluation camera modules are also available now.

http://www.ovt.com

> Read More

Mouser Electronics signs global distribution deal with QuickLogic

Tue, Apr 20, 2021

Embedded FPGA IP, low power, multi-core, voice-enabled SoCs and endpoint artificial intelligence (AI) from Quicklogic is now available from Mouser Electronics, following a global distribution agreement between the developers and the distributor.

The agreement means that Mouser now stocks QuickLogic’s EOS S3 microcontroller- and FPGA-based platform and its QuickFeather development kit.

The EOS S3 sensor processing platform is a multi-core SoC that enables an array of concurrent sensor applications, from basic to computationally demanding algorithms for smartphone, wearable, and IoT devices. EOS S3 integrates a low-power Arm Cortex-M4F microcontroller sub system and embedded FPGA (eFPGA) fabric, with optional sensory low power sound detect (LPSD) for on-chip voice recognition. The heterogeneous architecture enables designers to partition their software and use cases across the variety of cores to optimise a system for lowest power consumption.

The QuickFeather development kit is a small form factor system which enables the next generation of low power machine learning (ML) -capable IoT devices, explains Mouser. It is powered by the EOS S3 and is based on open source hardware. It is compatible with the Adafruit Feather form factor. The board is built around an open source software ecosystem, including SymbiFlow FPGA tools, Zephyr and FreeRTOS operating systems and Renode. Third-party software is available for sensor processing, voice, and AI/ML use cases, including SensiML and Google’s TensorFlow Lite for microcontrollers.

Mouser is a global authorised distributor, offering a wide selection of the newest semiconductors and electronic components which are in stock and ready to ship. To help speed customers’ designs, Mouser’s website hosts an extensive library of technical resources, including a Technical Resource Centre, along with product data sheets, supplier-specific reference designs, application notes, technical design information, engineering tools and other helpful information.

Mouser Electronics’ website is available in multiple languages and currencies and features more than five million products from over 1,100 manufacturer brands. Mouser offers 27 support locations worldwide and ships to 223 countries/territories from its distribution facilities in the Dallas, Texas, metro area in the US.

http://eu.mouser.com

> Read More

AI controller and software detect humans, using less energy

IoT kit demonstrates value for narrow band technology

Omnivision releases automotive image sensors for Nvidia AGX AI platform

Mouser Electronics signs global distribution deal with QuickLogic

About Smart Cities