خط مشی دسترسیدرباره ما
ثبت نامثبت نام
راهنماراهنما
فارسی
ورودورود
صفحه اصلیصفحه اصلی
جستجوی مدارک
تمام متن
منابع دیجیتالی
رکورد قبلیرکورد بعدی
Document Type : Latin Dissertation
Language of Document : English
Record Number : 149484
Doc. No : ET21276
Main Entry : WAD HATEM
Title Proper : A HYBRID MULTIVARIATE CLASSIFICATION TECHNIQUE AND ITS APPLICATION IN TISSUE IMAGE ANALYSIS
Note : This document is digital این مدرک بصورت الکترونیکی می باشد
Abstract : Discriminant analysis and neural network techniques were used in this work todevelop a hybrid mukivariate classification machine. The motivation behind the design ofthis machine was a theoretical interpretation of human recognition of patterns. Thisinterpretation considers the pattern recognition task as a two-step process: differencedetection and classifier training. Based on this theory, a machine of classifiers (MC) wasdesigned to do classification through two steps in two separate components. The firstcomponent performs unsupervised clustering of data based on an iterative Fisherdiscriminant technique. The second component was a supervised network designed on thebasis of the committee machine technique of neural network. MC was tested onartificially-generated multivariate data. The results.
Subject : Electericl tess
: برق
electronic file name : TL44432.pdf
Title and statement of responsibility and : A HYBRID MULTIVARIATE CLASSIFICATION TECHNIQUE AND ITS APPLICATION IN TISSUE IMAGE ANALYSIS [Thesis]
 
 
 
(در صورت عدم وضوح تصویر اینجا را کلیک نمایید)