Mixtile Edge AI solutions support ONVIF cameras and leverage object detection, object classification, real-time analytics, and intelligent surveillance to create IoT video analytics applications across security, transportation, logistics, retail, industrial, and agriculture sectors.
The hardware is based on the proven Mixtile Edge 2 Kit IoT Gateway powered by Rockchip RK3568 SoC with 1 TOPS NPU and 1080p60 VPU, or the Mixtile Blade 3 SBC with the more powerful Rockchip RK3588 octa-core SoC with 6 TOPS AI accelerator and 8Kp60 VPU – the former can analyze up to 10 streams in real-time, the latter a whopping 40 streams.
The novelty here is on the software side, developed by Mixtile, which works like this: ONVIF cameras send video streams over the network to the Mixtile Edge AI Box. The Edge AI Box then runs pre-trained AI models to analyze the streams in real time and sends the analyzed data to third-party video management software (VMS) so that it can trigger alarms, start recordings, or send notifications to administrators' mobile devices based on pre-defined rules. The diagram below summarizes how this works:
The Mixtile Edge AI solution supports four pre-trained models out of the box:
- Behavioral analysis Ensure safety and security by detecting falls among children, workers and elderly, spotting theft in stores and identifying fights.
- Facial Recognition It supports identifying attributes such as gender, hairstyle, clothing, etc. It is useful for airport security, office attendance management, and community access control.
- Quantitative Statistics The Mixtile Edge AI Box can be used to count visitor entries and exits at museums and art galleries to efficiently monitor engagement levels, and can also be used in parking lots to assess occupancy levels and in agriculture to estimate orchard yields.
- Object Classification They identify pedestrians and vehicles on the road to improve the safety of self-driving cars, analyze customer behavior to optimize sales in smart retail, and monitor forest fires to analyze air pollution emissions.
You can also train customized models with support for TensorFlow, Caffe, TFLite, PyTorch, ONNX NN, and Android NN. Workloads run on low power, consuming around 4W on the Mixtile Edge 2 Kit and around 9W on the Blade 3.
Mixtile supports its customers by responding to all technical inquiries for each project within 48 hours and also provides comprehensive documentation including SDKs, development documents, user manuals and tutorials. To learn more about Mixtile Edge AI box and solutions, Related pages on the Mixtile websiteand Contact the company Interested in our video analytics solutions?
This account is for paid sponsored posts. We do not take a sales commission. The content is usually provided by the advertisers themselves, but we sometimes write for our clients.
Support CNX Software! Donate with cryptocurrency, become a patron on Patreon, or buy our products on Amazon and Aliexpress
