OpenCV Essentials: Acquire, process, and analyze visual content to build full-fledged imaging applications using OpenCV

First things first, OpenCV Essentials is definitely not a book for an image processing/computer vision novice.If you’ve been using OpenCV and programming in C++ for at least a few months and want a resource which you can quickly refer to when you’re delving into some of the deeper functionality of OpenCV, then I can thoroughly recommend this book to you.I’ve been using OpenCV for about one and a half years, and over that time I have witnessed a definite improvement in the documentation available online. However, good examples of some of the deeper functionality are still difficult to come across, and the ability of this book to plug this gap is one of its chief virtues.

ひと昔まえなら貧弱なCPUとあいまって実現不可能な画像処理がOpenCVを使えばいとも簡単に（＜－開発経験者にとって）できてしまうことがよく理解できます。サンプルコードと、処理前後の写真を見るだけでも、古いソフトエンジニアには、唸ること間違いなし。英文も難しい言い回しもなく、大きな勘違いもなく、理解できるはずです。Kindleだと辞書で単語調べるのはラクラクですし。

I received a copy of this from the publisher, and I am very glad I did. It's a wonderful book for everyone who wants to get fast solid knowledge of OpenCV.I have done Image processing in my University program and if I had the book at that time, the course would be much more easy and interesting primarily because "OpenCV Essentials" delivers all information in a straightforward way supported by lots of examples. The book is perfect for people with basic C++ knowledge who want to dwelve in the deeps of OpenCV with no complications like advanced mathematics and programming. I found it a pleasure to dip into different sections of the book with no previous knowledge of the topic.I would recommend the book to all newcomers in OpenCV.

A few weeks ago I have been selected by Packt Publishing to review the new book “OpenCV Essentials” by Oscar Deniz Suarez et.al. Thus here is my objective review:This book is for people who have at least basic knowledge in OpenCV and Computer Vision, if you are new in OpenCV maybe this book won’t be the suitable one. Additionally all the examples are in C++ for Visual Studio, I did not like the last restriction, but it was necessary for some examples. This book show some upcoming OpenCV functions for the next release 3.0 (available at this time in Alpha version), and there are a lot of very useful functions.Let’s begin with the main review. First chapters are about OpenCV basic functionalities (load images, brightness, contrast, color spaces, arithmetic and geometrical transforms). For every topic they show the code, explanation and results. The book shows tips in each topic, giving you the opportunity to investigate more about the functionality, and also they usually share links to Scientific Papers, thus, if you have a better idea to improve that function it helps you to do research about it.I found Threshold topic in Chapter 4 (What’s in the Image? Segmentation) very useful and didactic, a lot of developers (also experts) made mistakes in this topic when choosing the type of threshold. Until here, Chapters were about main OpenCV functions, but here on-wards the book becomes better and very useful with content about Pattern Recognition, Classification, also Video Processing. Thus, I will discuss separated every chapter.Chapter 5: Focusing on the Interesting 2D Features: This Chapter is totally based in Keypoints and its variations, the book provide a lot of examples and also the matching between images using Keypoints. Additionally some new OpenCV feature detectors of the new release are explained (KAZE and AKAZE).Chapter 6: Where’s Wally? Object Detection: The main topics are Cascade detectors and Latent SVM. The book shows an example of basic pedestrian detection in a few lines of code using HOG cascade detector. Also, if you need to train your own cascade, you will find a detailed explanation about how to do it.Chapter 7: What is he Doing? Motion: Due to I developed this kind of systems, I particularly liked this Chapter. Topics like Video tracking, Motion and Background subtraction are now included in the new OpenCV 3.0 and certainly will be very useful for everyone who is developing/researching this area. The book does not show the typical Optical Flow, but explains and optimized and faster Lucas-Kanade optical flow instead.Chapter 8: Advanced Topics: If you are interested in Machine learning, Classification, or CUDA GPU programming this Chapter will be very useful too. It shows how to use the Random Forest classifier, one of the best classifiers available in recognition power and efficiency. Also the typical SVM is explained with a simple version of a recognition system.To summarize, if you are using OpenCV for basic image processing certainly you will be more interested in Chapter 1 to 4, and maybe 5. If you need more depth knowledge and want to do complex Pattern Recognition systems you will be interested in Chapter 5 on-wards. The main weak point of this book is about pre-processing image explanation, although authors show some of them, they do not explain the real potential of the pre-processing step for this kind of systems. In all my experience I used a lot of pre-processing functions to improve results. Simple functions like erode, dilate and smoothing to avoid noise in images are very useful in all Computer Vision projects. Afterwards, I found this book very useful and recommendable.Overall, this is a must-read book for programmers, researchers, and people eager with basic OpenCV, Computer Vision and C++ knowledge.

Book is very good recipe for those who has prior knowledge of opencv, I did my engineering project using opencv so I was already familiar with this topic. Examples given in this book are for visual studio, but in first chapter it also explains how to use QT grphics library and integrating opecv with QT.In 3rd Chapter book covers basic image types, fundamentals of image and data types used to represent an image.In segmentation chapter author explains algorithms with proper examples and pictures which helps alot in understanding segmentaion and it usability in image processing.In Chapter 6 and 7 author describes SVM , Object detection algorithms and then how it works in OpenCV, This is what makes this book unique in its category. I always wanted to know how can I use machine learning algorithms in opencv, I had basic idea of opencv but problem was from where to start. This is a book which answers all those basic questions.We know that number of standard algorithms are already implemented in opencv but how can we use and what prior data is required to use those functions. This book genuinely helps to build concepts in image processing and opencv.