What do you think?
Rate this book
Machine Vision, Second Edition: Theory, Algorithms, Practicalities
The field of machine vision has expanded extensively since the First Edition of Machine Vision was published by Academic Press in 1990. As a result, this Second Edition contains significant amounts of new material on artificial neural networks, mathematical morphology, motion, invariance, texture analysis, x-ray inspection, and foreign object detection. Intermediate level vision is examined in depth (especially Hough transforms), and automated visual inspectionis discussed. The author takes care to consider theoretical aspects as well as practical applications, including perspective invariants and robust statistics. Written in a user-friendly style and full of up-to-date methods, Machine Vision, Second Edition will be an essential volume for students and professionals in the field.
Key Features
* Gives considerable emphasis to robust analysis of images to demonstrate how problems of occlusion, noise, distortion, and variability may be overcome
* Introduces Hough transforms as an integral part of the text and shows how they may be applied in a variety of situations
* Presents the topic of robust statistics non-mathematically in the context of vision algorithms
* Considers the role of neural networks in machine vision
* Shows how the concepts of perspective invariance provide basic strategies for 2-D and 3-D vision
* Studies image transformations and the prespective n-point problem systematically to clarify how interpretation may proceed in various geometrical situations
* Pays special attention to the detection of defects, foreign objects, and real-time implementation hardware in consideration of automated visual inspection
Key Features
* Gives considerable emphasis to robust analysis of images to demonstrate how problems of occlusion, noise, distortion, and variability may be overcome
* Introduces Hough transforms as an integral part of the text and shows how they may be applied in a variety of situations
* Presents the topic of robust statistics non-mathematically in the context of vision algorithms
* Considers the role of neural networks in machine vision
* Shows how the concepts of perspective invariance provide basic strategies for 2-D and 3-D vision
* Studies image transformations and the prespective n-point problem systematically to clarify how interpretation may proceed in various geometrical situations
* Pays special attention to the detection of defects, foreign objects, and real-time implementation hardware in consideration of automated visual inspection
- GenresProgramming
750 pages, Hardcover
First published November 1, 1990
2 people are currently reading
30 people want to read
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 of 1 review
February 28, 2017
Extremely comprehensive overview of computer vision techniques prior to the current deep learning wave. The textbook has since been updated to a newer edition. While this won't teach you how to build a convolutional neural network to recognize pictures of cats, it covers almost every other conceivable vision technique, and I felt much more grounded after having read the whole thing cover to cover. (I am actually going over parts of it again to make sure I got all the math).
Beyond its comprehensiveness, the book is particularly notable for Davies' insistence on principled, rather than ad-hoc approaches. Again and again he begins a chapter describing a simple trick people once used, notes where the technique is weak, breaks down how the technique works - sometimes down to the math of the distance of pixels - and uses it to design a new, better technique, for which he presents empirical evidence of its superiority (usually in a paper he wrote or co-wrote :-D).
Short story: read the new edition!
Beyond its comprehensiveness, the book is particularly notable for Davies' insistence on principled, rather than ad-hoc approaches. Again and again he begins a chapter describing a simple trick people once used, notes where the technique is weak, breaks down how the technique works - sometimes down to the math of the distance of pixels - and uses it to design a new, better technique, for which he presents empirical evidence of its superiority (usually in a paper he wrote or co-wrote :-D).
Short story: read the new edition!
Displaying 1 of 1 review