research centers


Search results: Found 3

Listing 1 - 3 of 3
Sort by

Article
Neuro Fuzzy Network and Wavelet Gabor For Face Detection

Author: Raidah Salim
Journal: Journal of Kufa for Mathematics and Computer مجلة الكوفة للرياضيات والحاسوب ISSN: 11712076 Year: 2013 Volume: 1 Issue: 8 Pages: 48-57
Publisher: University of Kufa جامعة الكوفة

Loading...
Loading...
Abstract

This paper presents a face detection technique based on two techniques: wavelet Gabor filter for extract features from the localized facial image and neuro fuzzy system used as classifier depending on the features that extract , where it is used to determine the faces in the input image by draw boxes around the faces. The neurofuzzy network will be train on 128 image (69 face and 59 non face, size of each image 16*27 pixel in gray scale , this mean it trained to choose between two classes “face” and “non-face” images. Our approach has been tested on eight common images with different face number in image and different number of fuzzy set. We got the best detection rate is 89.3% in case threshold equal 0.2 and in case number of fuzzy set equal 2. The stages of this work are implemented in MATLAB 7.0 environment.


Article
Takagi-Sugeno-Kang(zero-order) model for diagnosis hepatitis disease

Author: Raidah Salim
Journal: Journal of Kufa for Mathematics and Computer مجلة الكوفة للرياضيات والحاسوب ISSN: 11712076 Year: 2015 Volume: 2 Issue: 3 Pages: 72-83
Publisher: University of Kufa جامعة الكوفة

Loading...
Loading...
Abstract

The aim of this paper is to use Takagi-Sugeno-Kang(zero-order) model as fuzzy neural network for the medical diagnosis of hepatitis diseases which represent a major public health problem all around the world . For further improve the accuracy and the speed of the diagnosis, the Microarray Attribute Reduction Scheme (MARS) for reduction features (or attributes) and Mean Imputation (MI) method for treatment the missing values were used in this work. The used data source of hepatitis diseases was taken from UCI machine learning repository.After treat the missing values problem by apply MI method, the dataset is partitioned into three training–testing partitions (30%–70%, 40–60% and 20%–80% respectively) and apply MARS with different values of thr(from 0.1-0.9 ) in order to determine the number attributes (that represent the number of inputs to the fuzzy neural network), the results record in each case of thr values and each case of partitions. The high diagnosis accuracy has been achieved for the 40–60% training–testing, namely, 100% for training and 95.77% for testing with thr equal to 0.4 and with less training cycle and fuzzy sets number. This work was implemented in MATLAB 7.0 environment.


Article
Hybrid System Geno-Fuzzified Neural Network For Solving Some Classification Problems
نظام هجين: تضبيب شبكة عصبية-جينية لحـــل بعـض مسائـل التصنيــف

Author: Raidah Salim Khaudeyer & Shatha Faleh Hendy رائدة سالم خضير و شذى فالح هندي
Journal: Journal of Basrah Researches (Sciences) مجلة ابحاث البصرة ( العلميات) ISSN: 18172695 Year: 2011 Volume: 37 Issue: 2B Pages: 104-117
Publisher: Basrah University جامعة البصرة

Loading...
Loading...
Abstract

This paper presented a hybrid method consisting of three intelligent systems (artificial neural networks, fuzzy logic, genetic algorithms), since these systems are effective in solving different issues but all the system suffers from some problems that reduce efficiency, so it was to integrate these systems with some To give the system benefit from the advantages of each method and encroaches on the disadvantages. We used in this research method of a hybrid resulting from a combination of fuzzy logic and neural networks, as used fuzzy logic to fuzzified training data and weights used in the neural network, and this method is called fuzzified neural networks, which gives the network a greater ability to generalize and accelerate the convergence process, but this method suffers of a problem in determining the number of fuzzy sets and optimal fuzzy weights, as the experiment method used to select it. To resolve this problem, genetic algorithm was used to determine the best number of fuzzy sets and the best fuzzy weights through research, which makes the network more efficiently trained.

قَدَّم هذا البحث طريقة هجينة تتكون من الأنظمة الذكية الثلاثة الشبكات العصبية الاصطناعية، المنطق المضبب، الخوارزميات الجينية، إذ تعد هذه الأنظمة فعالة في حل مسائل مختلفة إلا أن كل نظام يعاني من بعض المشكلات التي تقلل كفاءته، لذا تم دمج هذه الأنظمة مع بعض لتعطي نظاماً يستفيد من محاسن كل طريقة ويتعدى عن مساوئها.استعملنا في هذا البحث طريقة هجينة ناتجة من الجمع بين المنطق المضبب والشبكات العصبية، إذ استعمل المنطق المضبب لتضبيب بيانات التدريب والأوزان المستعملة في الشبكة العصبية وتسمى هذه الطريقة تضبيب الشبكات العصبية، الذي يعطي الشبكة قدرة أكبر على التعميم ويسرع من عملية التقارب، ولكن هذه الطريقة تعاني من مشكلة في تحديد عدد المجاميع المضببة والأوزان المضببة المثلى، إذ تستعمل أسلوب التجربة لتحديدها. لحل هذه المشكلة استعملت الخوارزمية الجينية لتحديد أفضل عدد من المجاميع المضببة وأفضل أوزان مضببة من خلال البحث ، مما يجعل الشبكة تتدرب بصورة أكفأ.

Listing 1 - 3 of 3
Sort by
Narrow your search

Resource type

article (3)


Language

English (2)

Arabic (1)


Year
From To Submit

2015 (1)

2013 (1)

2011 (1)