Distributor availability of STM32H747I-DISCO
Distributor Name | Region | Stock | Min. Order | Third party link |
---|---|---|---|---|
ANGLIA Live | EUROPE | 105 | 1 | Buy from Distributor |
MOUSER | WORLDWIDE | 279 | 1 | Buy from Distributor |
DIGIKEY | WORLDWIDE | 455 | 1 | Buy from Distributor |
Distributor reported inventory date: 3/29/2025
ANGLIA Live
Stock
105Min.Order
1MOUSER
Stock
279Min.Order
1DIGIKEY
Stock
455Min.Order
1Distributor reported inventory date: 3/29/2025
Stock
Min.Order
Distributor availability of B-CAMS-OMV
Distributor Name | Region | Stock | Min. Order | Third party link |
---|---|---|---|---|
Farnell Element14 | EUROPE | 84 | 1 | Buy from Distributor |
ANGLIA Live | EUROPE | 25 | 1 | Buy from Distributor |
MOUSER | WORLDWIDE | 74 | 1 | Buy from Distributor |
DIGIKEY | WORLDWIDE | 260 | 1 | Buy from Distributor |
Newark Element14 | AMERICA | 3 | 0 | Buy from Distributor |
Distributor reported inventory date: 3/29/2025
Farnell Element14
Stock
84Min.Order
1ANGLIA Live
Stock
25Min.Order
1MOUSER
Stock
74Min.Order
1DIGIKEY
Stock
260Min.Order
1Newark Element14
Stock
3Min.Order
0Distributor reported inventory date: 3/29/2025
Stock
Min.Order
This evaluation setup is based on the STM32H747XIH6 Discovery kit (STM32H747I-DISCO) with a high-performance, dual-core STM32H7 microcontroller running the computer vision artificial intelligence (AI) software (FP-AI-VISION1) and a camera module (B-CAMS-OMV) based on an image sensor with 5-Mpixel resolution and 8-bit color.
The FP-AI-VISION1 software running on the STM32H747XIH6 microcontroller performs edge-based people counting in a scene based on outputs generated by neural networks (NN). The AI model was optimized by ST’s AI expert team in terms of computation and memory footprint while preserving the accuracy in usual office environment conditions. Optimized C code is then generated using the STM32Cube.AI expansion pack (X-CUBE-AI). The LCD included in the STM32H747XIH6 Discovery kit completes the setup to display in real-time the number of counted persons, the captured video and the bounding boxes corresponding to people's positions in the video.
The object-detection convolutional neural network (CNN) model is part of STMicroelectronics model zoo and is derived from the YOLO model, quantized with 8-bit precision and expecting a 240 × 240 pixel resolution input image in RGB888 format.
The people counting application implements a full internal memory allocation scheme and enables image-based people recognition, counting and tracking applications.
Get the Software
Part Number | General Description | Marketing Status | Download | |
---|---|---|---|---|
FP-AI-VISION1 | STM32Cube function pack for high performance STM32 with artificial intelligence (AI) application for Computer Vision | Active | Get latest |
All Evaluation Features
-
- Ready-to-use people counting application based on an object detection convolutional neural network (CNN) model
- Complete application firmware for camera capture, frame image preprocessing, inference execution and output post-processing
- Support for camera frame capture to enable image dataset collection