A number of groups have developed packages of standard image processing data structures and operations to aid research in computer vision. Ones currently in use include the following.
Although they differ in many details, these systems all share a common approach: provide the user with a set of pre-written C functions. Such pre-written code is, of course, very helpful. However, it is not a long-term solution to the vision programming problem, because these packages are large, hard to maintain, and not standard across sites.
Ownership, Maintenance and Disclaimers
Manual Top Page