Keywords : artificial neural networks


Distinguish of Fingerprint Based on Artificial Neural Networks

Zahraa Maen Al-Qattanz

AL-Rafidain Journal of Computer Sciences and Mathematics, 2014, Volume 11, Issue 1, Pages 143-162
DOI: 10.33899/csmj.2014.163744

In this research use one of artificial intelligence techniques which is artificial neural networks was used to distinguish fingerprint from a range of fingerprints belong to a unified database, based on a set of properties to the texture of image and which are extracted and analyzed using co-occurrence  matrix (Event), These properties are (contrast ,correlation, determined, homogeneity), and after extracting properties, a combination of neural networks (Cascade Neural Network CNN and Radial Basis Functions netwoek RBFN and Elman Neural Network ENN) used to distinguish fingerprint, and the results of training 100%  for the three networks after being trained on the network (18) sample where each person(3)samples.
Network efficiency was measured in recognition by using scale (training rate) and scale (recognition rate RR) for comparison between these networks to see the best network in the recognition.
 

Use Artificial Neural Network Neococcontron in Distinguishing Handwritten Arabic Numerals

Laheeb Mohammad Ibrahim; Hanan H. Ali

AL-Rafidain Journal of Computer Sciences and Mathematics, 2009, Volume 6, Issue 1, Pages 47-59
DOI: 10.33899/csmj.2009.163784

Artificial Neural Networks have wide applications now a day, Among these are in the field of pattern recognition and image processing. This is due to the fact that it has a good performance and advanced mathematical computation power particularly its flexible adaptation to parallelism technique. That is why this research is conducted for the recognition of hand written Arabic numbers (0 - 9). Recognition artificial neural network is simulated the human eye for tracking the property of entered image (Feature extractor).
The systems  examined on  samples of Arabic  numbers its performance  was found to be balanced in spite of the variations in position and direction of the recognized number.
 

Integration Method with Backpropagation

Nidhal H. AL-Assady; Jamal S. Majeed; Shahbaa I. Khaleel

AL-Rafidain Journal of Computer Sciences and Mathematics, 2005, Volume 2, Issue 1, Pages 49-68
DOI: 10.33899/csmj.2005.164073

In this research, a new method is discovered (combined method) to accelerate the backpropagation network by using the expected values of source units for updating weights, we mean the expected value of unit by the sum of the output of the unit and its error term multiplied by the factor Beta to accelerate the algorithm and also adjust the value of learning coefficient continuously if the value of energy function E decreases the learning rate is increased by a factor , if the value of the energy function E increases , the value of the learning rate is decreased by a factor . To obtain the optimal weight with minimum iteration and minimum time, we applied a new method on many applications to prove the result of this method (pattern compression, encoding and recognition on Arabic, English digits and alphabetic.