Scene Analysis Using Regions
Abstract
One of the vision projects of the Stanford Research Institute Artificial Intelligence Group is described. The method employed uses regions as basic data and progresses by successive partitioning of the picture toward an interpretable "goal partition," which is then explored by a heuristic decision tree. A general structure is discussed and an example problem is shown in detail. One important component of the SRI automaton project is a set of programs that provide the automaton with a means of interpreting visual data. This automaton, or robot, described in detail by Nilsson (1969) and Munson (1970), is equipped with touch sensors, a range finder, and a standard vidicon camera. The output of the camera is transmitted by microwave to an analog-to-digital converter which produces a 120 X 120 digitized picture with 16 levels of gray. In the early stages of the robot project a prime concern was the simple task of navigation, for which the vision programs needed only to find areas of environment through which the robot could freely move. More recently, however, interest has turned to the more advanced tasks of collecting various objects and using tools to manipulate these objects. For these tasks it has become necessary for the programs to isolate and to recognize objects.
Document Details
- Document Type
- Technical Report
- Publication Date
- Apr 01, 1970
- Accession Number
- ADA638144
Entities
People
- Claude L. Fennema
- Claude R. Brice
Organizations
- SRI International