Deep Learning: Image Recognition (2018)

1h 44mIntermediate2020-01-15

Authors

Adam Geitgey

Developer and Machine Learning Consultant

Course details

Thanks to deep learning, image recognition systems have improved and are now used for everything from searching photo libraries to generating text-based descriptions of photographs. In this course, learn how to build a deep neural network that can recognize objects in photographs. Find out how to adjust state-of-the-art deep neural networks to recognize new objects, without the need to retrain the network. Explore cloud-based image recognition APIs that you can use as an alternative to building your own systems. Learn the steps involved to start building and deploying your own image recognition system.

Topics include:
Classifying images
Designing an image recognition system
Building a deep neural network
Training a deep neural network
Modifying pre-trained neural networks
Using image recognition APIs
Deploying a deep neural network

Skills covered

Real-TimeNeural Networks and Deep LearningMachine LearningPythonVisualization and Real-TimeAECProduct and ManufacturingArtificial Intelligence (AI)Open SourceDeep Dive (X:Y)

Concepts

0. Introduction

01 - Build cutting-edge image recognition systems
02 - What you should know
03 - Exercise files

1. Setting Up Your Development Environment

04 - Installing Python 3, Keras, and TensorFlow on macOS
05 - Installing Python 3, Keras, and TensorFlow on Windows

2. How Image Classification Works

06 - What is a neural network
07 - Coding a neural network with Keras
08 - Feeding images into a neural network
09 - Recognizing image contents with a neural network
10 - Adding convolution for translational invariance

3. Designing a Deep Neural Network for Image Recognition

11 - Designing a neural network architecture for image recognition
12 - Exploring the CIFAR-10 data set
13 - Loading an image data set
14 - Dense layers
15 - Convolution layers
16 - Max pooling
17 - Dropout
18 - A complete neural network for image recognition

4. Building and Training the Deep Neural Network

19 - Setting up a neural network for training
20 - Training a neural network and saving weights
21 - Making predictions with the trained neural network

5. Fine-Tuning Pre-trained Neural Networks

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

6. Using an Image Recognition API

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

Conclusion

33 - Next steps