Accession Number : ADA298528

Title :   The Application of Artificial Neural Networks to Object Direction In Digital Images,

Corporate Author : AIR FORCE INST OF TECH WRIGHT-PATTERSON AFB OH

Personal Author(s) : Housholder, Steacy W.

PDF Url : ADA298528

Report Date : AUG 1995

Pagination or Media Count : 74

Abstract : This report surveys the topic of object detection in digital images. It outlines a standard method of analysis for the perception process. Several imaging and recognition issues are identified and explored through the human visual system as a model which is then used for constructing an artificial perception system. This leads to the study of self-organized artificial neural networks. The study includes the Kohonen feature map and several of its variations. The report will next focus on the application of these neural networks to object recognition tasks.

Descriptors :   *NEURAL NETS, *SELF ORGANIZING SYSTEMS, DIGITAL SYSTEMS, DETECTION, PERFORMANCE(HUMAN), HUMANS, COGNITION, SURVEYS, IMAGES, STANDARDIZATION, RECOGNITION, ARTIFICIAL INTELLIGENCE, MAPS, VISUAL PERCEPTION, LEARNING.

Subject Categories : Cybernetics

Distribution Statement : APPROVED FOR PUBLIC RELEASE