OCCLUDED OBJECT DISCRIMINATION BY A MODIFIED HOPFIELD NEURAL NETWORKPOMPES PAR TRANSITIONS MULTIPLES

منو

" OCCLUDED OBJECT DISCRIMINATION BY A MODIFIED HOPFIELD NEURAL NETWORKPOMPES PAR TRANSITIONS MULTIPLES "

Document Type	:	Latin Dissertation
Language of Document	:	English
Record Number	:	152429
Doc. No	:	ET24221
Main Entry	:	DEXIANG LUO,B.ENG
Title Proper	:	OCCLUDED OBJECT DISCRIMINATION BY A MODIFIED HOPFIELD NEURAL NETWORKPOMPES PAR TRANSITIONS MULTIPLES
Note	:	This document is digital این مدرک بصورت الکترونیکی می باشد
Abstract	:	This thesis presents a new method of the 2-D partially occluded objectdiscrimination for the computer vision application. A binary modified Hopfield neuralnetwork was applied to perform the global feature matching. To obtain the featurepoints of the object, a Gaussian hction was implemented to smooth the objectboundary curve and a curvature estimation method was used to extract the dominant.
Subject	:	Electericl tess
	:	برق
electronic file name	:	TL48423.pdf
Title and statement of responsibility and	:	OCCLUDED OBJECT DISCRIMINATION BY A MODIFIED HOPFIELD NEURAL NETWORKPOMPES PAR TRANSITIONS MULTIPLES [Thesis]