6.Feature Extraction
Features play a central role in image processing. These numerical values, derived from direct calculations on the image, allow:
to represent each pixel i by a vector , each of the components reflecting a feature relevant to the problem at hand;
to detect points of interest in the image, for further processing (registration, matching, …).
We illustrate the first aspect in Sections 7.2 and 6.3. We focus on the second aspect in the two other sections of this chapter, and we more particularly search for particular geometric objects (points, shapes) in the image. First, we detail a method dedicated to the detection of points of interest and more particularly corners. Then we focus on a method for the detection of parametric ...
Get Digital Image Processing with C++ now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.