Fisher keys for content based retrieval

In the classic computer science paradigm of data searching, the data is sor ted according to a key and then inserted into a hash table or tree for fast access. How would this paradigm work for images? What key function would b e best? This paper examines the problem of efficient indexing of large imag e databases using the concept of image keys. The ideal image key maximizes the probability that the key of a computed image copy is closer to the key of the original than the key to a different image in the database. The case of optimal linear image keys turns out to be similar to Fisher's linear di scriminant. Results on image collections with real world noise are presente d. (C) 2001 Elsevier Science B.V. All rights reserved.