My father used to say that every page that you read in a scientific journal, book or magazine should be worth $1, adding to your knowledge of how to solve specific problems by leveraging the talents of others. Over the years I'd like to think that the pages ofVision Systems Design magazine have somewhat lived up to his expectations. At the same time, however, I have learned that some pages are worth more than others. Last month, my belief in this was confirmed after I received a copy of a new book published by Apress (New York, NY, USA; www.apress.com).
Although his book "Computer Vision Metrics - Survey, Taxonomy and Analysis" is somewhat awkwardly titled, Scott Krig, founder of Krig Research (www.krigresearch.com) has written a book unlike any other on the subject of image processing I have ever read. Is it full of the mathematical formulations found in classics such as "Digital Image Processing" by Gonzalez and Wintz (http://amzn.to/1rznwlK) or Jain's "Fundamentals of Digital Image Processing" (http://amzn.to/1rOOJ6m)? It is not. Does it cost the hundreds of dollars commanded by these worthy tombs? I am happy to report that it does not. It is free of charge from his publisher at http://www.apress.com/9781430259299.
Why then is this book worth reading? First, it assumes that the reader is familiar with the basic principles of image processing and machine vision. Further, a description of the typical hardware components -lenses, lighting, frame grabbers, PLCs and monitors used in such systems is not included. This is a software-centric publication.
In essence, this nearly 500 page paperback (I was sent the printed version) sums up just about every algorithm I have ever heard of - including my favorite - the Mahalanobis distance! Better yet, the over 500 references in the publication draws readers into further research should they wish to do so.
Scott Krig's book should be on every software engineer's shelf (or computer). He is to be commended for such a fine publication. Why his publisher, Apress, offers this publication free of charge, however, one can only guess. I would have paid $100, although that is still less than the $1 per page my father would have recommended.