Customer Reviews

Introduction to Data Compression, Second Edition (The Morgan Kaufmann Series in Multimedia Information and Systems)

5 star:		(8)
4 star:		(5)
3 star:		(2)
2 star:		(1)
1 star:		(0)

 

 

While some books may cover individual compression techniques in finer detail, Sayood's book provides the best comprehensive overview of the data compression field including video, voice, images, text, music, etc. The reader gains a deep respect for the underlying mathematical basis for compression without being inundated with complex mathematics. I use this book for a senior undergrad / beginning grad student level course at the University of Delaware, and strongly recommend it to others.

The book is one of the clearest I have read as a text book. Why cant everybody write like this?? There is very good flow throughout the text.

Only complaint is the software. It looks the software has not kept pace with the book itself. Some additional software has to be added (for Transform coding, for instance), and some references in the text book to the software are incorrect. If the accompanying software is upgraded, as it should be, I will rate this book a clear 5-star.

Many reviewers tried to say that this book should avoid the word "INTRODUCTION" in it's name because of it's intensive mathematics. But I'm totally disagree. Mathematics and all theoretical aspects explained in this book is just a preliminary review compare to what the researchers all around the world are doing in the data compression area. This book is really an "INTRODUCTION" book.

This book is very comprehensive. You'd absolutely need it if you are going to start a research in this area. If you're just wonder how data can be compressed, you can try other easy books like Peter Wayner's :"Compression Algorithms for Real Programmers". Or if you need compression as a feature in your software product, I'd suggest you not to implement it yourself. Try 3rd party library. Data Compression is the job for specialists.

Amazon claims that this is "the definitive guide". I have to agree with them because this book is the only broad grey pages introduction to data compression that I have been able to find, and it is very well written.

If you are only casually interested in data compression this book is not for you.

If you are interested in adding compression to your application and your data falls into a common category, sound, video, text ect this book is probably not for you. You should look to the open source community or buy an off the shelf product.

But if your data is odd or unique like say telemtry data (I'm sure there are other examples I just can't think of any) and you need to design a compression scheme for your data this book is "the only" book for you.

If you want to begin research into data compression and you are a newbie this book is a must have.

Appropriate examples in this book help understanding of this field, even if readers don't have enogh grounding in mathematics. It should be also mentioned that this second edition covers new technologies such as wavelet-based compression. This is one of the best textbook to study still-image compression, while may be somewhat insufficient to study video compression.

This textbook is fairly comprehensive on the various compression algorithms, and the theory behind them. However, a CD with included sample code and the "example" pictures mentioned in the text would have helped tremendously in conveying the material. As it is, the text tries to explain the algorithms using math with no sample code whatsoever.

The best thing about this book is the coverage and organization of the material. Sayood covers a wide variety of compression topics without getting into the nitty gritty details of them all. Thats why its an "Introductory" book. This book is a valuable resource for those who want to know the basics of various compression techniques and can be used as a starting point for further details. Some topics like arithmetic coding are covered in more detail than others. The book is also organized nicely with mathematical foundations provided as and when necessary.

This book is very comprehensive, so it isn't just an introduction. Some mathematical maturity is required to understand parts of it. Nevertheless, it is an excellent reference for professionals in this field.

This is one of those books that only gets a new edition when the author has something genuinely new to say, and this third edition of Sayood's excellent introduction to data compression is no exception. This particular edition is different from the second mainly in that there is a new chapter on audio compression that includes a description of the mp3 algorithm. Also there is additional information on the new video coding standards as well as the new facsimile standards.

As to the target audience for this book, if you are tasked with designing hardware or software implementations of data compression algorithms and you have some background in either electrical engineering or computer science, then this is a good book from which to learn and then to practice what you learn via some very good exercises. Some prior knowledge of information theory and random processes wouldn't hurt either. There is also an abundance of examples that are sprinkled throughout the book to illustrate concepts as they are presented. The author's approach in each chapter is to explain each concept in as an accessible manor as possible, present relevant equations, and then work an example using what has just been presented.

The book presents the mathematical preliminaries in chapter 2, and chapters 3 and 4 are dedicated to coding algorithms which include Huffman coding, arithmetic coding, Golumb-Rice codes, and Tunstall codes. Chapters 5 and 6 describe many of the popular lossless compression methods and their applications. These methods include LZW, BWT, and DMC. Chapter 7 describes various lossless image compression algorithms such as JBIG as well as their applications. Chapter 8 discusses the mathematical background of lossy compression standards. Chapters 9 and 10 concentrate on quantization since it is the basis of most lossy compression schemes. Chapter 11 discusses differential encoding techniques such as DPCM and delta modulation. Included is a discussion of the CCITT G.726 standard.

Chapter 12 is the third and final chapter dedicated to mathematical foundations. It is meant to prepare the reader for the chapters on transform, subband, and wavelet based methods that encompass the following three chapters. The JPEG standard is covered in chapter 13, the CCITT G.722 standard in chapter 14, and the EZW, SPIHT, and JPEG2000 standards are covered in chapter 15. Chapter 16 focuses on audio compression and includes descriptions of the various MPEG audio compression schemes including mp3. Chapter 17 switches gears somewhat and covers techniques in which the data to be compressed is analyzed and a model is produced. This model is then used to synthesize the data and is quite useful in speech compression. Chapter 18 deals with video compression and diverges from the book's central theme of dealing with techniques rather than applications. The chapter discusses the H.261 standard as well as MPEG-1, MPEG-2, and MPEG-4 standards.

The website for the book, found at the publisher's site, contains a large number of C programs dealing with compression. I haven't tried to use any of these yet, so I can't speak to their validity.

This book has all the ingredients for a great textbook. It provides good theoratical background without going into unnecessary details, gives lot of discussion about applications, provides great exercise problems, and above all it has outstanding examples that makes some of the difficult concepts easy to understand.

Data compression needs a lot of background in information theory and other areas specific to speech, image processing etc. It is impossible to give a rigourous theoratical treatment of all of those in one volume. A strong point of this book is that it gives you just enough background on a variety of topics - without making the whole book obscure. In that respect, it is very application and implementation oriented. It is in fact what it says it is: A very good "INTRODUCTION to Data Compression"