Imagimob AI, a development platform for building tinyML applications for edge devices, adds support for deep learning anomaly detection. Deep anomaly detection—the identification of rare items, events, or observations that differ significantly from the majority of data—can eliminate feature engineering, saving costs and reducing time-to-market.
Feature engineering, in simple terms, is the act of converting raw observations into desired features using statistical or mathematical functions. Feature engineering normally requires domain expertise and is generally very time consuming.
With the added support for convolutional autoencoder networks in Imagimob AI, developers can build anomaly detection for predictive maintenance in less time and with better performance. Imagimob has tested and verified the software on real-world machine and sensor data. To get developers up and running quickly, the company offers an anomaly detection starter project for rotating machinery.
This release of Imagimob AI also includes:
- Support for quantization of models in the GUI, including 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 projects supporting sensors and MCUs from Texas Instruments, Renesas, STMicroelectronics, Acconeer, and Nordic Semiconductors
Sign up for a free trial of Imagimob AI here. The trial program includes a number of project templates/content packs and one month of free usage.
Find more datasheets on products like this one at Datasheets.com, searchable by category, part #, description, manufacturer, and more.