This guide shows you how to train a neural network that can recognize fire in images. The ability to recognize fire means that the neural network can make fire-detection systems more reliable and cost-effective. This guide shows you how to use the Python Application Programming Interfaces (APIs) for the Arm NN inference engine to build a sample application that classifies images as fire or non-fire.
This guide uses a Raspberry Pi 3 or 4 device. The Raspberry Pi device is powered by an Arm CPU with Neon architecture. Neon is an optimization architecture extension for Arm processors. Neon is designed for:
- Faster video processing
- Image processing
- Speech recognition
- Machine Learning
Neon provides Single Instruction Multiple Data (SIMD) instructions, where multiple processing elements in the pipeline perform operations on multiple data points simultaneously. Arm NN provides the APIs to harness the power of the Neon backend.
At the end of this guide, you will be able to:
- Run a Python script that predicts whether a supplied image contains fire or not
- Explain what the differences are between Arm NN and PyArmNN