Deep Learning: Image Recognition

Learn how to design, build, and deploy a deep neural network to serve as an image recognition system.

Introduction

• Build cutting-edge image recognition systems
• What you should know
• Exercise files

1. Setting Up Your Development Environment

• Installing Python 3, Keras, and TensorFlow on macOS
• Installing Python 3, Keras, and TensorFlow on Windows

2. How Image Classification Works

• What is a neural network?
• Coding a neural network with Keras
• Feeding images into a neural network
• Recognizing image contents with a neural network
• Adding convolution for translational invariance

3. Designing a Deep Neural Network for Image Recognition

• Designing a neural network architecture for image recognition
• Exploring the CIFAR-10 data set
• Loading an image data set
• Dense layers
• Convolution layers
• Max pooling
• Dropout
• A complete neural network for image recognition

4. Building and Training the Deep Neural Network

• Setting up a neural network for training
• Training a neural network and saving weights
• Making predictions with the trained neural network

5. Fine-Tuning Pre-trained Neural Networks

• Pre-trained neural networks included with Keras
• Using a pre-trained network for object recognition
• Transfer learning as an alternative to training a new neural network
• Extracting features with a pre-trained neural network
• Training a new neural network with extracted features
• Making predictions with transfer learning

6. Using an Image Recognition API

• When to use an API instead of building your own solution
• Introduction to the Google Cloud Vision API
• Setting up Google Cloud Vision account credentials
• Recognizing objects in photographs with Google Cloud Vision
• Extracting text from images with Google Cloud Vision

Conclusion

• Next steps