Edge AI dev platform supports deep learning anomaly detection
The company’s Imagimob AI development platform for machine learning on edge devices now has added support for autoencoder networks, enabling developers to build anomaly detection in less time, and with better performance. A big strength with deep learning anomaly detection, says the company, is that it delivers high performance as well as eliminates the need for feature engineering – the act of converting raw observations into desired features using statistical or mathematical functions, which normally requires domain expertise and is in general very time consuming.
“Not only is deep learning anomaly detection better for eliminating the need for feature engineering but it can also leverage and deliver excellent performance on the new generation of powerful neural network processors that is now hitting the market.” says the company. “This means that when going to the edge customers can make the most of their hardware.”
The platform’s new anomaly detection features include:
- End-to-end training and deployment of convolutional autoencoder networks for anomaly detection/predictive maintenance
- Anomaly detection starter-project for rotating machinery to get developers up and running in minutes
Other improvements include:
- Support for quantization of models in the graphical user interface. This includes quantized models, reducing model size and decreasing inference time on MCUs without an FPU
- Improved model prediction – tracking of how models perform with millisecond resolution, before deploying given different confidence thresholds
- Faster training and model evaluation
- Increased support for large data sets
- Starter project for Renesas RA2L1 – Capacitive Touch Sensing Unit
- In total 8 starter projects, supporting sensors and MCUs from Texas Instruments, Renesas, STMicroelectronics, Acconeer and Nordic Semiconductors
The anomaly detection solution has been tested and verified on real-world machine and sensor data, says the company. The new release is available now.
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
