OpenCV Computer Vision with Python

This book covers exactly what it claims to cover - an introduction to using the OpenCV library with the Python language. It is not a Python book, and it is not an OpenCV book, as it explains neither in depth. But for someone who wants to start playing around with OpenCV algorithms, using Python is a good starting point (as it's simple and fast for prototyping), and this book will help you get started right away without having to look at too many online sources.

Gives the reader a very good intro to using OpenCV for facial recognition. I recommend writing the code yourself rather than just downloading the example code.

Complete waste of money. Very elementary. Very sparse book mostly about how to install opencv. Don't waste money on this book.Go on the internet and look the excellent tutorials by Abid K. Free and much more useful than this.

OpenCV is probably the most extensive open source graphics package and language out there on the net. So using it via Python coding is a natural need for some readers. The problem addressed by the text is that OpenCV is not the easiest of packages to learn to use in Python. (In case you wonder, it is not much easier in C++ or java either.) The extensiveness of chapter 1 in dealing with installing OpenCV on your computer is the first indication of its complexity.The Python examples are deliberately in an object oriented style. Though given the necessarily brief nature of the examples, the author could have coding procedurally. Which style you adopt is largely up to you.The text assumes that even if you have not used OpenCV in other contexts, you are at least conversant about key graphics ideas, like a CMV colour space. Or the alternate colour spaces of RGV and RC (which I have never used).One nifty example which could appeal to some readers is the emulation of common types of photographic film, like the Fuji Provia and Velvia or the Kodak Portra. When you had to use those films, you were intrinsically stuck with the highlighting ranges of each. Nothing you could alter. Now it is remarkable how easy the book demonstrates about just toggling between each film, using the filters provided.Another chapter of the book delves into using Haar filters, or cascades as it terms them. The idea of Haar filters is a key one in image manipulation. For this, you might need a more extensive text on graphics that gives a thorough exposition. The current text will quickly walk you through using them.

I've been working with OpenCV with Python for about 2 years. I have developed some applications with Python using OpenCV. All of my knowledge has been self taught with respect to OpenCV with Python as there were no books covering Python API. Python API for OpenCV has come a long way since then. If you are looking to start with OpenCV with python this is a decent addition to your collection. Working with Python is great because of the speed of development. Many organizations use Python for prototyping their Computer Vision applications as most of the software written for scientific apps (written with Numpy and Scipy) can easily be combined with Python API for OpenCV. The "cv2" module of OpenCV is a great advancement compared to the old "cv" module and this module is covered in this book. If one compares this book with Packt's OpenCV C++ version you could see that a lot of stuff is missing(array manipulation, machine learning, etc). This is so because that missing content becomes equivalent of Numpy and other libraries but not OpenCV in Python API. Some of the other useful stuff in the book is training of haar cascades xml files (but that doesn't fall under Python specifically). Overall, I would give a 4.0/5.0 to this book.

If you have some experience with the Python programming language and your aim is start to use the OpenCV library for creating computer vision / image processing applications, this book is certainly a good start point.The author focus the first chapter into explaining the several installation options that exists for the different platforms (Win, Mac, Linux). This is good since such a huge library has several ways to be installed, with the possibility of enabling/disabling some features. I can remember how I spent much time dealing with these kind of issues in my first contact with OpenCV some years ago.In the rest of the chapters some of the basic key issues of the library are introduced to the readers in order to create a simple application which involves:- Input / output features of the library (Dealing with images, videos, devices, etc.).- Basic filtering and color-space conversions.- Face tracking with Haar cascades classifiers.- Depth information manipulation (depth sensors such as kinect).All the code is designed using the Object oriented paradigm in order to be extensible and reusable. This is one of the things I more liked about this book, since many of the python examples (as well as C/C++ ones) provided by the OpenCV library are programmed like independent programs/scripts that could not be extensible.The only negative point I can found in this book is its short content. I miss the explanation of some basic features/algorithms I use to employ daily when I'm working with OpenCV (in C++), but since the book is meant for beginners I can understand it.It is important to note that this book assumes you have basic knowledge about Python programming.Overall, I would give a 3.5/5.0 to this book.

Get this book if you are trying to install OpenCV on Linux and cannot get it to work. It gives clear directions and scripts that work. I used it to on a Rasperry pi and got it to build and install the latest version including face recognition.

Very limited amount of use cases.

excellent book for beginners.

パソコン用のＫｉｎｄｌｅソフトで読めません。英語のサイトから買えば、読めるようですね。