Keywords : neural network


Textures Recognition using Elman Neural Network

Fawziya Mahmood Ramo; Alaa Anwr Mohamed

AL-Rafidain Journal of Computer Sciences and Mathematics, 2014, Volume 11, Issue 1, Pages 97-108
DOI: 10.33899/csmj.2014.163741

         In this research building system to recognition  texture images using artificial neural networks. The system consists of two phases: phase extraction important feature of each texture by using an algorithm Principal Components Analysis (PCA)   and recognition phase which recognize these feature by using  Elman network were trained network on a number of various texture  models down to the steady-state network  and then test the network by input  samples  of textures.  The experiments show that the method achieves high performance and produces 92% recognition rate.
 

Application of Chaotic Neural Network for Authentication using the Database

Ammar Thaher Yaseen Abd Alazeez

AL-Rafidain Journal of Computer Sciences and Mathematics, 2014, Volume 11, Issue 1, Pages 81-95
DOI: 10.33899/csmj.2014.163740

  In this paper a new algorithm is suggested to encrypt data, since it was to take advantage of properties of the chaotic, so it was entered as a key in encryption and hiding  by entering values to the artificial neural network for training as well as hide in the picture, beside use the database for storage and retrieval information and increase the secret system. Through the overlap between the results of stages encryption and hide and artificial neural network algorithm was obtained exciting new strength from where you can not detect secret text only after obtaining random values of the chaotic algorithm and information about neural network algorithm as well as algorithm of work.
 

Neural Network Using for Extracting Hidden Information in Images

safwan hasoon; Farhad M. Khalifa

AL-Rafidain Journal of Computer Sciences and Mathematics, 2013, Volume 10, Issue 3, Pages 113-130
DOI: 10.33899/csmj.2013.163539

Steganography technique widely spread and varied. With the widening in using steganography, its misuse alarmists arisen. Thus steganalysis comes into sight to deter unwanted secret communications.
In this paper a new scheme proposed for extracting hidden information, this scheme relies on the capability of artificial neural networks for prediction to estimate the original values of the pixels which values of some of them were changed by the affection of data embedding process, and then the present pixel values will be compared with estimated values to identify the embedded data. Multilayer Perceptron MLP neural network used in this scheme to estimate the pixel's original value using its neighbor pixels. The proposed schemes programmed using Matlab v. 7.10.0.499. The proposed schemes has been trained and tested using a data base prepared for this purpose. Then its performance compared with another work in the same field applied in similar conditions. The results showed that the proposed scheme has the ability to achieving the desired with a good rate of success.
 

Electronic Forecasting of Women's Jumping Events using Neural Networks

Fares Ghanem Ahmed; Aida Younis Muhammad; Hala Nafi Fathi

AL-Rafidain Journal of Computer Sciences and Mathematics, 2013, Volume 10, Issue 1, Pages 309-321
DOI: 10.33899/csmj.2013.163461

The aim of this research is to use neural network in future forecasting field to show the of jumping competitions in international Olympics for (2016-2024).
Expert system named (AAA) is designed by using neural network in  future forecasting field for period chain of data from 1984 to 2012,which represents 8 years period.  The data represent the first three winners in running competition for (100 m., 200 m., 400 m., 100 m. Hurdles, 400 m. Hurdles, 4×100 Relay, 4×400 Relay), The prepared programs for this research has been done C++.  Then it forecast three future levels represented in (2016, 2020, 2024),where the Olympic Cycle take place each 4 years.
Throughout the results it found that forecasting values are the best by using neural networks then other traditional methods used before.
This  paper is depended  on the results of athletes who take Olympic medals in women jumping events (long, triple, high and pole-vault) in 8  Olympic cycles, since Tokyo cycle (1984) till the last Olympic cycle in (2012) . The cycle on (1984) was used as the  beginning of  study as it considered as the first Olympic cycle.
 

Steganalysis Using KL Transform and Radial Basis Neural Network

Safwan Hasoon; Farhad M. Khalifa

AL-Rafidain Journal of Computer Sciences and Mathematics, 2012, Volume 9, Issue 1, Pages 47-58
DOI: 10.33899/csmj.2012.163670

The essential problem in the security field is how to detect information hiding. This paper proposes a new steganalysis scheme based on artificial neural network as a classifier to detect information hiding in colored and grayscale images. The statistical features extracted from Karhunen-Loève (KL) transform coefficients obtained from co-occurrence matrix of image. Then radial basis neural network (RBNN) trained using these features to discriminate  whether the image contains hidden information or not. This system can be used to prevent the suspicious secret communication.
 

Neural Network with Madaline for Machine Printed English Character Recognition

Shaymaa M. Al-mashahadany

AL-Rafidain Journal of Computer Sciences and Mathematics, 2011, Volume 8, Issue 1, Pages 47-58
DOI: 10.33899/csmj.2011.163607

The recognition of optical characters is known to be one of the earliest applications of Artificial Neural Networks, which partially emulate human thinking in the domain of artificial intelligence.
In this research the application of neural networks to the problem of identifying English machine printed characters in an automated manner is developed. A preprocessing step is implemented to separate each character from the others. After that a feature extraction process is applied on each character to obtain the minimum nodes by using Mean, Standard Deviation, and Variance. Madaline neural network is trained on a 26 alphabetical English characters with a standard font and size. And tested on these characters to verify each character image belongs to which type of character. This is done by using MATLAB®2008a.
 

Genetic Pattern Recognition using Intelligent Techniques

Basil Younis Thanoon; Omar Saber Qasim

AL-Rafidain Journal of Computer Sciences and Mathematics, 2010, Volume 7, Issue 3, Pages 161-172
DOI: 10.33899/csmj.2010.163935

In this research a light is shed on converting DNA series to amino acid being responsible for forming protein through intelligent techniques. Some comparisons have been made between particularly Artificial Neural Network, Fuzzy Logic and Genetic Algorithms for discovering the powerful and the week ones in particle way. The hybrid operation has been made between Artificial Neural Network and Fuzzy Logic t to get a hybrid technique in a new formula having robust results than the original ones.
 
 

General Regression Neural Network Application for Dynamic Data Compression and Decompression

safwan hasoon; Susan Hassan Mohammed

AL-Rafidain Journal of Computer Sciences and Mathematics, 2010, Volume 7, Issue 3, Pages 73-80
DOI: 10.33899/csmj.2010.163912

In the last decade have been witnessed a great development in artificial intelligence especially in neural network .This paper have been employed neural networks techniques for data compression and decompression.
 The application of General Regression Neural Network (GRNN) in data compression is very important of data transmission, because this technique offers less than memory storage and time for transferring of the data over computer networks or internet. Taking into consideration the data compression provides security of these data.
 The matlab version (R2009a ) is used for designing the propose system of neural network (GRNN) to dynamic data compression and decompression .
 

Adoption of Neural Networks to Classified the Gender of the Speaker

Khalil I. Al-Saif; Mason Kh. Al-Nuaimi

AL-Rafidain Journal of Computer Sciences and Mathematics, 2010, Volume 7, Issue 3, Pages 47-56
DOI: 10.33899/csmj.2010.163926

In this research the neural network was adopted to classified the gender of the spoken, by creating the two dimension matrix from the parameters of the spoken speech signal which normal was snigle dimension array.
            The porpose algorithm in this research divided in two stage :-
            In the first stage the seven moment were calculated for a set of spoken signal of 50 persons , to be followed creating database depend on the seven moments .This database will be used to find the threshould value for both genders (male/female) which will be trained  by neural network to classify any  input tothe network.
In the second stage , speech  of any spoken will be selected and the same feature will be extracted , as in the first stage  , to be used as input to the neural network which was traind previously for gender recognition.
Back propagation neural network was achieved for recognition. The result of the applied algorithem on 10 spoken passed on 8 of them and 2 of them was failed .
 

Recognition of Eudiscoaster and Heliodiscoaster using SOM Neural Network

Raid R. AL-Nima

AL-Rafidain Journal of Computer Sciences and Mathematics, 2010, Volume 7, Issue 3, Pages 141-152
DOI: 10.33899/csmj.2010.163918

This research is aimed to design an Eudiscoaster and Heliodiscoaster recognition system. There are two main steps to verify the goal. First: applying image processing techniques on the fossils picture for data acquisition. Second: applying neural networks techniques for recognition.
The image processing techniques display the steps for getting a very clear image necessary for extracting data from the acquisition of image type (.jpg). This picture contains the fossils. The picture should be enhanced to bring out the pattern. The enhanced picture is segmented into            144 parts, then an average for every part can easily be computed. These values will be used in the neural network for the recognition.
For neural network techniques, Self Organization Maps (SOM) neural network was used for clustering. The weights and output values will be stored to be used later in identification. The SOM network succeeded in identification and attained to (False Acceptance Rate = 15% - False Rejection Rate = 15%).
 
 

Audio File Compression Using Counter Propagation Neural Network

Saja J. Mohammed

AL-Rafidain Journal of Computer Sciences and Mathematics, 2010, Volume 7, Issue 1, Pages 153-168
DOI: 10.33899/csmj.2010.163869

In this paper audio files are compressed using counter propagation neural network (CPNN) which is one of the fastest neural networks in multi media. The utilized counter propagation neural network was trained on uncompressed sound file to obtain the final weights of this CPNN (Kohonen layer, Grossberg layer ).
In compression operation: the sound signal segmented to number of frames equal in size. Then these frames are applied step by step, to the first layer of the neural network(kohonen layer) to obtain some compression results. The decompression operation done by retrieve stored information  in resulted file. This information is applied to second layer of this CPNN (Grosberg layer) which will perform decompression operation and retrieve the original sound file. The proposed algorithm is applied on (.wav) audio files , The results show high performance in addition to short time in compression and decompression operation.
 

Design and Implementation of Stream Cipher Using Neural Network

Siddeq Y. Ameen; Mazin Z. Othman; Safwan Hasoon; Moyed Abud Al-Razaq

AL-Rafidain Journal of Computer Sciences and Mathematics, 2009, Volume 6, Issue 1, Pages 237-249
DOI: 10.33899/csmj.2009.163781

The centaral problem in stream cipher cryptograph is the  the difficulty to generate a long unpredicatable sequence of binary signals from short and random key. Unpredicatable sequence are desirable in cryptography because it is impossible, given a reasonable segment of its signals and computer resources, to find out more about them. Pseudorandom bit generators have been widely used to construct these sequences.
The paper presents a PN sequence generator that uses neural network. Computer simulation tests have been carried out to check  the  randomness  of the generated  through statistical tests. There tests have shown the successful PN sequence generator passes all the recommended tests. The paper also proposes and validates the data encryption and decryption process using neural network instead of using traditional methods (Exclusive or). This task increases the difficulty in the breaking the cipher.