Chapter 1, Laying the Foundation, will lay the foundation for the basic data structures and containers in OpenCV for example, the Mat, Rect, Point and Scalar objects . It also explains the need for each of them as a separate data types by offering an insight into the possible use cases. A major portion of the chapter focuses on how OpenCV uses the Mat object to store images so that your code can access them, basics of scanning images using the Mat object and the concept of R, G and B color channels. After the readers are comfortable with working with images, we introduce the concept of pixel-based image transformations and show examples of simple code that can achieve contrast/brightness modification using simple Mat object traversals.

Chapter 2, Image Filtering, progresses to slightly advanced pixel traversal algorithms. We introduce the concepts of masks and image filtering and talk about the standard filters such as box filter, median filter, and Gaussian blur in OpenCV. Also, we will mathematically explain the concepts of convolution and correlation and show how a generic filter can be written using OpenCV’s filter2D method. As a practical use case, we implement a popular and interesting image manipulation technique called the Vignette filter.

Chapter 3, Image Thresholding, talks about image thresholding, which is yet another process that frequently comes up in the solution to most computer vision problems. In this chapter, the readers are made to understand that the algorithms that fall into this domain basically involve operations that produce a binary image from a grayscale one. The different variants such as binary and inverted binary, are available as part of the OpenCV imgproc module are explained in detail. The chapter will also briefly touch upon the concepts of erosion and dilation (morphological transformations) as steps that takes place after thresholding.

Chapter 4, Image Histograms, talks about aggregating pixel values by discussing image histograms and histogram-related operations on images such as calculating and displaying histograms, the concept of color and multi-dimensional histograms. 

Chapter 5, Image Derivatives and Edge Detection, focuses on other types of information that can be extracted from the pixels in our image. The readers are introduced to the concept of image derivatives and the discussion then moves on to the application of image derivatives in edge detection algorithms. A demonstration of the edge detection methods in OpenCV for example Sobel, Canny and Laplacian is presented. As a small practical use-case of the Laplacian, we demonstrate how it can be used to detect the blurriness of images.

Chapter 6, Face Detection Using OpenCV, talks about one of the most popular and ubiquitous problems in the computer vision community--detecting objects, specifically faces in images. The main motivation of this chapter is to take a look at the Haar-cascade classifier algorithm, which is used to detect faces and then go on to show how the complex algorithm can be run using a single line of OpenCV code. At this point, we introduce our programming project of predicting gender of a person from a facial image. The reader is made to realize that any system involving analysis of facial images (be it face recognition or gender, age or ethnicity prediction) requires an accurate and robust face detection as its first step.

Chapter 7, Affine Transformations and Face Alignment, serves as a natural successor to the preceding chapter. After detecting faces from images, the readers will be taught about the post-processing steps that are undertaken--geometric (Affine) transformations such resizing, cropping and rotating images. The need for such transformations is explained. All images in the gender classification project have to go through this step before feature extraction.

Chapter 8, Feature Descriptors in OpenCV, introduces the notion of feature descriptors. The readers realize that in order to infer meaningful information from images, one must construct appropriate features from the pixel values. In other words, images have to be converted into feature vectors before feeding them into a machine learning algorithm. To that end, the chapter also goes on to introduce the concept of local binary pattern as a feature and also talks about the details of implementation. We demonstrate the process of calculating the LBP features from the cropped and aligned facial images that we obtained in the previous chapters.

Chapter 9, Machine Learning with OpenCV, teaches the readers different classifiers and machine learning algorithms available as part of the OpenCV library and how they can be used to make meaningful predictions from data. The readers will witness the learning algorithms accept the feature vectors that they had computed in the previous chapters as input and make intelligent predictions as outputs. All the steps involved with using a learning algorithm--training, testing, cross validation, selection of hyper-parameters and evaluation of results--will be covered in detail.

Appendix, Command-line Arguments in C++, talks about command line arguments in C++ and how to extract the best possible use of them while writing OpenCV programs.

Learning OpenCV 3.0 Application Development is the perfect book for anyone who wants to dive into the exciting world of image processing and computer vision. This book is aimed at programmers having a working knowledge of C++. A prior knowledge of OpenCV is not required. The book also doesn't assume any prior understanding of computer vision/machine learning algorithms as all concepts are explained from basic principles. Although, familiarity with these concepts would help. By the end of this book, readers will get a first-hand experience of building and deploying computer vision applications using OpenCV and C++ by following the detailed, step-by-step tutorials. They will begin to appreciate the power of OpenCV in simplifying seemingly challenging and intensive tasks such as contrast enhancement in images, edge detection, face detection and classification.

The book assumes a basic, working knowledge of C++. However, prior knowledge of computer vision, image processing or machine learning is not assumed. You will need an OpenCV 3.1 installation in your systems to run the sample programs spread across the various chapters in this book. The setup and installation details have already been shared.

In this book, you will find a number of text styles that distinguish between different kinds of information. Here are some examples of these styles and an explanation of their meaning.

Code words in text, database table names, folder names, filenames, file extensions, pathnames, dummy URLs, user input, and Twitter handles are shown as follows: "We can include other contexts through the use of the include directive."

A block of code is set as follows:

typedef Size_<int> Size2i; 
typedef Size2i Size;

New terms and important words are shown in bold. Words that you see on the screen, for example, in menus or dialog boxes, appear in the text like this: "Clicking the Next button moves you to the next screen."

Feedback from our readers is always welcome. Let us know what you think about this book—what you liked or disliked. Reader feedback is important for us as it helps us develop titles that you will really get the most out of.

To send us general feedback, simply e-mail [email protected], and mention the book's title in the subject of your message.

If there is a topic that you have expertise in and you are interested in either writing or contributing to a book, see our author guide at www.packtpub.com/authors.

Now that you are the proud owner of a Packt book, we have a number of things to help you to get the most from your purchase.