Book Image

OpenCV 4 Computer Vision Application Programming Cookbook - Fourth Edition

By : David Millán Escrivá, Robert Laganiere
Book Image

OpenCV 4 Computer Vision Application Programming Cookbook - Fourth Edition

By: David Millán Escrivá, Robert Laganiere

Overview of this book

OpenCV is an image and video processing library used for all types of image and video analysis. Throughout the book, you'll work with recipes to implement a variety of tasks. With 70 self-contained tutorials, this book examines common pain points and best practices for computer vision (CV) developers. Each recipe addresses a specific problem and offers a proven, best-practice solution with insights into how it works, so that you can copy the code and configuration files and modify them to suit your needs. This book begins by guiding you through setting up OpenCV, and explaining how to manipulate pixels. You'll understand how you can process images with classes and count pixels with histograms. You'll also learn detecting, describing, and matching interest points. As you advance through the chapters, you'll get to grips with estimating projective relations in images, reconstructing 3D scenes, processing video sequences, and tracking visual motion. In the final chapters, you'll cover deep learning concepts such as face and object detection. By the end of this book, you'll have the skills you need to confidently implement a range of computer vision algorithms to meet the technical requirements of your complex CV projects.
Table of Contents (17 chapters)

Detecting features quickly

The Harris operator proposed a formal mathematical definition for corners (or, more generally, interest points) based on the rate of intensity changes in two perpendicular directions. Although this constitutes a sound definition, it requires the computation of the image derivatives, which is a costly operation, especially considering the fact that interest point detection is often just the first step in a more complex algorithm.

In this recipe, we present another feature point operator, called FAST (short for Features from Accelerated Segment Test). This one has been specifically designed to allow quick detection of interest points in an image; the decision to accept or not to accept a keypoint is based on only a few pixel comparisons.

How to do it...

...