image recognition

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.

Image recognition refers to the recognition of objects or ther features in an image. This can be useful in day-to-day aplications, for example, labelling people in a personal photo collection, and also more life-critical situations such as looking for cancerous cells in X-rays.. It is also critical in autonomous vehicles identifying road markings, obstacles, other cars and pedestrians. A more controversial application is in CCTV surveliance, partcularly facial recognition, which can be used to track down criminals, but also to control citizens in authoritarian regimes. There are also potential problems of bias in both therapeutic and law-enforcement applications.

Used in Chap. 8: page 111; Chap. 17: page 285; Chap. 20: page 344; Chap. 21: pages 352, 359; Chap. 22: page 369; Chap. 23: page 387