This device is not compatible.
PROJECT
Data Augmentation for ML Datasets
In this project, we’ll get hands-on experience with data augmentation. We’ll explore different libraries for data augmentation, including OpenCV, TensorFlow, and imgaug. Finally, we’ll learn about advanced augmentation functions in this project.
You will learn to:
Apply data augmentation.
Perform image preprocessing.
Make image augmentation pipelines.
Handle image datasets.
Skills
Data Visualization
Machine Learning Fundamentals
Prerequisites
Intermediate knowledge of image processing
Basic understanding of computer vision
Intermediate knowledge of Python
Basic understanding of TensorFlow
Technologies
Python
OpenCV
Project Description
An important way to increase the performance and outcomes of a machine learning model is through data augmentation. In the case of a computer vision model, it can increase the number of images and help the model to make better generalizations.
We’ll start by making small transformations to images using OpenCV. Next, we’ll make transformations to a batch of images using TensorFlow. Finally, we’ll make a data augmentation pipeline with imgaug that will apply multiple random transformations to an image.
Project Tasks
1
Getting Started
Task 0: Introduction
2
OpenCV
Task 1: Load the Dataset
Task 2: Translate an Image
Task 3: Rotate and Scale an Image
Task 4: Add Perspective
3
Tensorflow
Task 5: Apply Affine Transformations
Task 6: Apply Batch Transformations
4
imgaug
Task 7: Apply Sequential Transformations
Task 8: Add Randomness
Task 9: Apply More Affine Transformations
Congratulations
Relevant Courses
Use the following content to review prerequisites or explore specific concepts in detail.