Neural Network on Minized - Number recognition




Features

• Deep learning capability on low-end device XC7Z007s
• Multilayer Perceptron (MLP) network topology
• Extreme quantized network using Binarized Neural Networks*
• 95.8% accuracy on MNIST
• Small Fully Connected Layer: 256 nodes
• Jupyter notebook over built-in Wifi
• 91%/66% (LUT/FF) Resource Utilization

*Yaman, “FINN: A Framework for Fast, Scalable Binarized Neural Network Inference”


Key Components

• Zynq XC7Z007S
• Dialog PMIC
• Murata Wifi Module


Target Applications

• Smart camera for surveillance
• Smart camera for industrial automation


Block Diagram





For further enquiries, please contact us at Asia-GDS-Contact@Avnet.com.