Road to embedded world '23: Plan-les-Ouates, Switzerland, STMicroelectronics
February 23, 2023
Blog
“Today we are unveiling the world’s first MCU AI Developer Cloud, which works hand-in-glove with our STM32Cube.AI ecosystem. This new tool brings the possibility to remotely benchmark models on STM32 hardware through the cloud to save on workload and cost” said Ricardo De Sa Earp, Executive Vice President General-Purpose Microcontroller Sub-Group, STMicroelectronics.
And we are stopping by to get some first-hand knowledge of what is in store for us at embedded world 2023 from STMicoelectronics. Added bonus, we are in Switzerland, and I am eating, everything I can consume. So, as we all wait somewhat patiently for ew '23, let's consume all we can about our kindest of hosts.
STMicroelectronics will be highlighting the world’s first MCU Edge-AI Developer Cloud in hall 4A booth 148 at embedded world 2023. The MCU Edge-AI Developer Cloud is an online version of the STM32Cube.AIdesktop front-end tool including the resources for developers to validate and generate optimized STM32 AI libraries from trained Neural Networks.
The Cloud version delivers a range of industry-firsts:
- An online interface to generate optimized C-code for STM32 microcontrollers, without requiring prior software installation. Data Scientists and developers benefit from the STM32Cube.AI's proven Neural Network optimization performance to develop edge-AI projects.
- Detection, audio event detection for audio classification, and more.
- Hosted on GitHub, these enable the automatic generation of “getting started” packages optimized for STM32.
- Online benchmarking service for edge-AI Neural Networks on STM32 boards. The cloud-accessible board farm features a broad range of STM32 boards, refreshed regularly, allowing data scientists and developers to remotely measure the actual performance of the optimized models.
- STM32 model zoo, a repository of trainable deep-learning models and demos to speed application development. At launch, available use cases include human motion sensing for activity recognition and tracking, computer vision for image classification or object.
An Effective Use Case: Plant Leaf Disease Identification
Plant leaf disease identification is crucial for agriculture helping to prevent the spread of diseases, which can have a significant impact on crop yields and food security. Identifying the specific disease allows farmers to take appropriate measures to control or eradicate the disease, such as applying the correct pesticides only on targeted plants or implementing quarantine measures.
Approach
- The STM32 model zoo provides everything you need to train and retrain models with your own data
- The solution proposes a model trained on a public dataset providing very good accuracy while running on a STM32
- The model can be easily deployed on the STM32H747 discovery kit with the STM32 model zoo Python scripts
- The use case presented is based on a plant leaf dataset to identify diseases
Sensor
-
Vision:
- Camera module bundle (reference: B-CAMS-OMV)
Data
-
Dataset:
-
Data format:
- 39 different classes of plant leaf and background images
- RGB color images
Results
-
Model:
- Fast-downsampling MobileNet 0.25
-
Input size:
- 224x224x3
-
Memory footprint:
- 137 KB Flash for weights
- 152 KB RAM for activations
-
Accuracy:
- Float model: 99.82%
- Quantized model: 99.62%
-
Performance on STM32H747 (High-perf) @ 400 MHz
- Inference time: 63.2 ms
- Frame rate: 16 fps
“Our goal is to deliver the best hardware, software, and services to meet the challenges faced by embedded developers and data scientists so that they can develop their edge AI application faster and with less hassle,” said Ricardo De Sa Earp, Executive Vice President General-Purpose Microcontroller Sub-Group, STMicroelectronics.
For more information, visit stm32ai.st.com.
Chad Cox. Production Editor, Embedded Computing Design, has responsibilities that include handling the news cycle, newsletters, social media, and advertising. Chad graduated from the University of Cincinnati with a B.A. in Cultural and Analytical Literature.
