views:

445

answers:

5

Hi,

I am interested in computer vision and image processing (object recognition). However my knowledge about these is equal to 0. I'm trying to find some recommendation about books that I should start with...

I found this book really interesting, at least by looking at table of contents and few customer reviews. They all seemed fine, but I still decided to ask here for other recommendations or if anyone here has read this book.

P.S. Please excuse me that I'm asking again similar question, but ones I found here doesn't mention this book so I'm wondering did I missed something .

EDIT: the reason why I'm interested in this topic is that I have an (possible great) idea, that among other complicated things include recognizing and extracting specific objects from an image:)

Many thanks

+4  A: 

When I was in school, I used Gonzalez & Woods during my undergrad and Milan Sonka during my grad .That is for basics. If you are into perspective geometry, try Olivier Faugeras. ofcourse you would need to program stuff, MATLAB is a good start. And for remote sensing, Robert Schowengerdt is THE GUY (I took his classes when in U of A, and he was one of the very well funded research profs there)

ram
+5  A: 

I know you don't have knowledge in CV but it would be nice if you can tell your goals and interests as CV is a huge subject. Computer Vision can be used in many fields and which one use some techniques more than others, ex:

  • Illumination: industrial illumination techniques for instance
  • Image segmentation: the ability to separate regions in a image based in some criteria
  • Image filtering: Removal of parts of the image that can cause misinterpretation (noise for example)
  • Pattern recognition: ability to identity groups of pixels (objects) and classify then, given a meaning to that object.

What I can recommend:

Here you'll find some lectures about computer vision

Books:

  • This book is very nice and has a practical math approach
  • That book about patterns recognition (gonzalez & Tou) is quite old but one of the best. The problem it's no easy to find.
  • Also in the field of feature extraction, this book is specific to shape features, very good material with a good introduction to the necessary math and image processing needed.

Finally if you are serious about diving into CV, I recommend you a specialization in the field.

Gook luck!

Andres
Thanks for an info! I've updated my post with reason why I need this, and yes, I think of specializing in this field, even though it is a quite complex :)
sinec
+1  A: 

The zisserman and hartley book "multi view geometry" and forsynth 's computer vision a modern approach are 2 extremely loaded books; I used this recently for my grad level course. but for a starter Gonzalez and Woods is a good starting point. The former are more modern in the content and upto date with current algorithms and state of the art technology while the later is more classical in it nature and starts with the basics

Egon
+4  A: 

As a graduate student working in computer vision. I must say that there are lots of Math in it. But if you are going to spend your time learning the Math it will never finish because the Mathematics in computer vision is just too broad. So start reading about the computer vision and learn the necessary Math as you go. Here are some of the books that I can recommend.

Image Processing:

  • Digital Image Processing by Gonzalez and Woods

Computer Vision:

  • Multiple View Geometry by Hartley and Zisserman

  • Computer Vision: A Modern Approach by Forsyth Ponce

    If you find Multiple View Geometry too hard to read try reading this one for the basics.

Machine Learning

  • Pattern Recognition and Machine Learning by Bishop

    This book is very Mathematical especially in areas of Probability and Statistics. You may need to brush up your Math skills before reading this book.

Programming

  • Learning OpenCV: Computer Vision with the OpenCV Library by Bradski and Kaehler]

    A little outdated as the book is based on OpenCV 1.0 while the current version is 2.0. However it gives lots of good code samples. You should consult the OpenCV documentation for the latest changes.

You should have a good knowledge in Linear Algebra as well. If not try the MIT video lectures on Linear Algebra.

Sunny
Heh heh... Bishop is a dangerous recommendation for a beginner. Gonzalez-Woods is good, though.
Steve
True!! But you really can't do much without machine learning right :-)
Sunny
+1  A: 

Hi,doing my research I find one amazing free to download book about Computer Vision. The contents of this book allow the reader to know more technical aspects and applications of computer vision. The intended audience is anyone who wishes to become familiar with the latest research work on computer vision, especially its applications on robots This book features representative work on the computer vision, and it puts more focus on robotics vision and omnidirectional vision. This is the link where you can find it: http://www.intechopen.com/books/show/title/computer_vision

Hanibal Lecter