Environment

Agriculture

Transfer learning applied to forest fire detection 

Image classification on high-perf MCU.

Transfer learning applied to forest fire detection  Transfer learning applied to forest fire detection 
Transfer learning applied to forest fire detection  Transfer learning applied to forest fire detection 
Fork on GitHub Read the tutorial

Environment

Agriculture

STM32Cube.AI

Image classification

Vision

Fork on GitHub Read the tutorial
Data collection and annotation is often a tedious and time consuming task to get satisfying image classification results. The transfer learning technique enables to overcome this challenge by reducing the number of images and the training time needed to add a new class. It is applied here to fire detection in the wild, detecting if there is a fire or no fire but can apply to many other situations. 

Approach

  • The tutorial presents how to use a technique called "Transfer learning" to quickly train a deep learning model to classify images.
  • The tutorial is based on the computer vision function pack FP-AI-VISION1

Sensor

Vision: Camera module bundle (reference: STM32H747I-D + B-CAMS-OMV)

Data

Dataset Dataset for forest fire detection (License CC BY 4.0) 
Data format
2 classes: fire and no fire
RGB image 250x250x3 

Results

Model MobileNetV2 alpha 0.35 
Input size: 128x128x3
Memory footprint:
403 KB Flash for weights
225 KBRAM for activations
Accuracy:
Float model: 98% 
Quantized model: 98% 
Performance on STM32H747 (High-perf) @ 400 MHz 
Inference time: 112 ms
Frame rate: 8.9 fps

Optimized with

STM32Cube.AI

Optimized with

Compatible with

STM32H7 series

Compatible with

Resources

Optimized with STM32Cube.AI

A free STM32Cube expansion package, X-CUBE-AI allows developers to convert pretrained AI algorithms automatically, such as neural network and machine learning models, into optimized C code for STM32.

Optimized with STM32Cube.AI

Compatible with STM32H7 series

The STM32 family of 32-bit microcontrollers based on the Arm Cortex®-M processor is designed to offer new degrees of freedom to MCU users. It offers products combining very high performance, real-time capabilities, digital signal processing, low-power / low-voltage operation, and connectivity, while maintaining full integration and ease of development.

Compatible with STM32H7 series

You also might be interested by

Agriculture | Appliances

Coffee bean recognition

Classify coffee beans thanks to an image classification model from STM32 Model Zoo running on a STM32H7 microcontroller.

Environment

Predicting extreme weather temperatures

Predict the next day's maximum temperature to better prepare for potential disasters.

Agriculture

Monitoring quality in a food production line

Create an AI model that predicts the quality of processed food instead of measuring it.