region detection

Terms from Artificial Intelligence: humans at the heart of algorithms

Page numbers are for draft copy at present; they will be replaced with correct numbers when final book is formatted. Chapter numbers are correct and will not change now.

Region detection is a phase in computer vision which attemtps to identify areas of the image that are a single object or part of an object, for example, one of the walls of a house. Traditional algorithms often use similarity of colour or texture. Deep neural networks can also be used, sometimes combining region detection and labelling.

Defined on page 260

Used on Chap. 12: pages 244, 258, 260, 261, 263, 285