research centers


Search results: Found 2

Listing 1 - 2 of 2
Sort by

Article
Estimation of Wavenet Optimistic Values by using Genetic algorithm to Recognize Colored images

Author: Dr.Entidhar Mhawes Zghair
Journal: journal of the college of basic education مجلة كلية التربية الاساسية ISSN: 18157467(print) 27068536(online) Year: 2011 Volume: 17 Issue: 71 Pages: 43-60
Publisher: Al-Mustansyriah University الجامعة المستنصرية

Loading...
Loading...
Abstract

The purpose of this work is to use Wave net and genetic algorithms for the prediction of the presented an original initialization procedure for the parameters of feed-forward wavelet networks, prior to training by gradient-based techniques.
Genetic algorithm global optimization techniques can be providing near optimal value for learning rate, translation, dilation, and hidden nodes. Then the net work can be learning in short learning time (reduce %50 from number of iteration) with desired accuracy is more than the desired accuracy without using genetic algorithm.
Then it uses the genetic algorithm method to determine a set of best wave net whose translation and dilation parameters with optimal value of learning rate. The hidden nodes are used as initial values for subsequence training. The results show high accuracy in classification and recognize of color medical images convert as gray level of applying of (WN).

Keywords


Article
Implementation of Artificial Neural Network and Random Iteration Algorithm Uses in Medical Images

Authors: Dr. Maha Abdul Ameer Kadhum --- Dr. Entidhar Mhawes Zghair
Journal: University of Thi-Qar Journal for Engineering Sciences مجلة جامعة ذي قار للعلوم الهندسية ISSN: 26645564/26645572 Year: 2013 Volume: 4 Issue: 1 Pages: 100-112
Publisher: Thi-Qar University جامعة ذي قار

Loading...
Loading...
Abstract

This research present a multiply connected neural network designed to estimate the fractal dimension (Df) using the Random Iteration Algorithm (IFSP). Fractal analysis is a powerful shape recognition tool and has been applied to many pattern recognition problems. Additionally, the one of the most adaptive Random Iteration Algorithm (IFSP) methods for estimating (Df). The architecture presented separates the calculation of (Df) into two sections, a data sampling section and a linear regression section. The data sampling section provides the ability to dyadic ally sample the data. The linear regression section simply calculates the slope of the best line through the sampling results. Instructional program was designed and built according to Knirk and Gustafson Design model, as one of the instructional design model in order to comprehension of the integrated ideas and concepts related to research .

يقوم هذا البحث بدراسة شبكة عصبية متعددة الارتباطات صممت لحساب البعد الكسوري باستخدام طريقة الدالة المكررة العشوائية .التحليل الكسوري هو اداة تعريف او تمييز بشكل فعال وله تطبيقات عديدة بمسائل التعريف النسقي بالاضافة الى ذلك , طريقة الدالة المكررة العشوائية هي واحدة من الطرق الشائعة لحساب البعد الكسوري . تم فصل حساب البعد الكسوري الى قسمين قسم عينة البيانات وقسم التراجع الخطي . قسم البيانات العينية يجهز للقابلية لتجزئة البيانات العينية. قسم التراجع الخطي ببساطة يقوم بحساب المنحني لاحسن خط موجود في نتائج العينات .برنامج تعليمي تم تصميمه وبناءه نسبة الى نموذج تصميم ( Knirk and Gustafson) كواحد من النماذج التعليمية المصممة حتى يتم استيعاب الآفكار المتكاملة و المعطيات المتعلقة بالبحث .

Listing 1 - 2 of 2
Sort by
Narrow your search

Resource type

article (2)


Language

English (2)


Year
From To Submit

2013 (1)

2011 (1)